11th EUROPEAN CONFERENCE FOR RESEARCH ON LEARNING AND INSTRUCTION Biennial Meeting

11th EUROPEAN CONFERENCE FOR RESEARCH ON LEARNING AND INSTRUCTION Biennial Meeting

11th EUROPEAN CONFERENCE FOR RESEARCH ON

LEARNING AND INSTRUCTION

Multiple Perspectives on Effective Learning Environments

Biennial Meeting

Nicosia, Cyprus

August 22-27, 2005

ABSTRACTS

Edited by

C. P. Constantinou, D. Demetriou, A. Evagorou, M. Evagorou,

A. Kofteros, M. Michael, Chr. Nicolaou,

D. Papademetriou, N. Papadouris

University of Cyprus

This book provides the abstracts for papers, symposia, panel discussions, interactive poster presentations, CIT sessions and thematic poster presentations of the EARLI 2005 Conference.

Abstracts are in chronological order by slot number of dates.

A list of participants’ e-mail adresses is included at the end.

European Association for Research on Learning and Instruction

11th Biennial Meeting

August 22-27, 2005, Nicosia, Cyprus

July 2005

© University of Cyprus

ISBN 9963-607-65-9

Printed by

Kailas Printers and Lithographers Ltd,

Nicosia, Cyprus

– 2 –

Contents

A 1 ................................................28

A 2 ................................................32

A 3 ................................................35

A 4 ................................................39

A 5 ................................................43

A 6 ................................................48

A 7 ................................................52

A 8 ................................................56

A 9 ............................................... 61

A 10 ................................................65

A 11 ................................................70

A 12 ................................................73

A 13 ................................................77

A 14 ................................................81

A 15 ................................................86

A 16 ................................................91

A 17 ................................................96

A 18 ................................................100

A 19 ................................................103

A 20 ................................................107

A 21 ................................................111

A 22 ................................................115

B 1 ................................................116

B 2 ................................................120

B 3 ................................................123

B 4 ................................................126

B 5 ................................................130

B 6 ................................................133

B 7 ................................................137

B 8 ................................................139

B 9 ................................................142

B 10 ................................................145

– 3 –

B 11 ................................................148

B 12 ................................................151

B 13 ................................................155

B 14 ................................................158

B 15 ................................................161

B 16 ................................................164

B 17 ............................................... 167

B 18 ................................................170

B 19 ................................................174

B 20 ................................................177

B 21 ................................................180

C 1 ................................................183

C 2 ................................................186

C 3 ................................................190

C 4 ................................................193

C 5 ................................................199

C 6 ................................................202

C 7 ................................................207

C 8 ................................................210

C 9 ................................................214

C 10 ................................................220

C 11 ................................................223

C 12 ................................................226

C 13 ................................................231

C 14 ................................................235

C 15 ................................................239

C 16 ................................................243

C 17 ................................................246

C 18 ................................................251

C 19 ................................................255

C 20 ................................................260

C 21 ................................................264

C 22 ................................................268

C 23 ................................................273

C 24 ................................................276

D 1 ................................................280

D 2 ................................................282

E 1 ................................................342

– 4 –

E 2 ................................................346

E 3 ................................................351

E 4 ................................................356

E 5 ................................................360

E 6 ................................................364

E 7 ................................................369

E 8 ................................................372

E 9 ................................................377

E 10 ................................................381

E 11 ................................................386

E 12 ................................................390

E 13 ................................................394

E 14 ................................................398

E 15 ................................................402

E 16 ................................................407

E 17 ................................................411

E 18 ................................................415

E 19 ................................................418

E 20 ................................................421

F 1 ................................................423

F 2 ................................................426

F 3 ................................................430

F 4 ................................................433

F 5 ................................................436

F 6 ................................................439

F 7 ................................................442

F 8 ................................................445

F 9 ................................................448

F 10 ................................................451

F 11 ................................................454

F 12 ................................................457

F 13 ................................................459

F 14 ................................................463

F 15 ................................................466

F 16 ................................................468

F 17 ................................................472

F 18 ................................................475

F 19 ................................................477

– 5 –

F 20 ................................................479

F 21 ................................................483

F 22 ................................................486

F 23 ................................................489

F 24 ................................................492

G 1 ................................................495

G 2 ............................................... 499

G 3 ................................................504

G 4 ................................................508

G 5 ................................................512

G 6 ................................................517

G 7 ................................................521

G 8 ................................................525

G 9 ................................................529

G 10 ................................................532

G 11 ................................................536

G 12 ................................................541

G 13 ................................................546

G 14 ................................................550

G 15 ................................................555

G 16 ................................................559

G 17 ................................................563

G 18 ................................................567

G 19 ................................................570

G 20 ................................................575

G 21 ................................................578

G 22 ................................................582

G 23 ................................................586

H 1 ................................................589

H 2 ................................................591

I 1 ................................................652

I 2 ................................................657

I 3 ................................................661

I 4 ................................................665

I 5 ................................................670

I 6 ................................................676

I 7 ................................................681

I 8 ................................................686

– 6 –

I 9 ................................................690

I 10 ................................................694

I 11 ................................................699

I 12 ................................................703

I 13 ................................................70

I 14 ................................................712

I 15 ................................................716

I 16 ................................................721

I 17 ................................................725

I 18 ................................................729

I 19 ................................................734

I 20 ................................................737

I 21 ................................................741

I 22 ................................................746

I 23 ................................................750

J 1 ................................................757

J 2 ................................................760

J 3 ................................................763

J 4 ................................................767

J 5 ................................................770

J 6 ................................................773

J 7 ................................................776

J 8 ................................................779

J 9 ................................................782

J 10 ................................................785

J 12 ................................................790

J 13 ................................................793

J 14 ................................................797

J 15 ................................................800

J 16 ................................................803

J 17 ................................................806

J 18 ................................................809

J 19 ................................................812

J 20 ................................................814

J 22 ................................................820

J 23 ................................................823

K 1 ................................................826

K 2 ................................................830

– 7 –

K 3 ................................................834

K 4 ................................................840

K 5 ................................................844

K 6 ................................................848

K 7 ................................................852

K 8 ................................................857

K 9 ................................................862

K 10 ................................................866

K 11 ................................................870

K 12 ................................................874

K 13 ................................................878

K 14 ................................................883

K 15 ................................................886

K 16 ................................................890

K 17 ................................................893

K 18 ................................................897

K 19 ................................................900

K 20 ................................................905

L 1 ................................................909

L 2 ................................................912

M 1 ................................................970

M 2 ................................................973

M 3 ................................................976

M 4 ................................................978

M 5 ................................................982

M 6 ................................................985

M 7 ................................................988

M 8 ................................................991

M 9 ................................................994

M 10 ................................................998

M 11 ................................................1001

M 12 ................................................1004

M 13 ................................................1007

M 14 ................................................1010

M 15 ................................................1013

M 16 ................................................1016

M 17 ................................................1019

M 18 ................................................1022

– 8 –

M 19 ................................................1025

M 20 ................................................1029

M 21 ................................................1031

M 22 ................................................1035

N 1 ................................................1038

N 2 ................................................1041

N 3 ................................................1044

N 4 ................................................1047

N 5 ................................................1050

N 6 ................................................1053

N 7 ................................................1057

N 8 ................................................1060

N 9 ................................................1063

N 10 ................................................1066

N 11 ................................................1068

N 12 ................................................1071

N 13 ................................................1074

N 14 ................................................1077

N 15 ................................................1079

N 17 ................................................1086

N 18 ................................................1089

N 19 ................................................1093

N 20 ................................................1096

N 21 ................................................1099

N 22 ................................................1101

N 23 ................................................1104

P 1 ................................................1107

P 2 ................................................1112

P 3 ................................................1116

P 4 ................................................1120

P 5 ................................................1125

P 6 ................................................1131

P 7 ................................................1134

P 8 ................................................1138

P 9 ................................................1143

P 10 ................................................1147

P 11 ................................................1151

P 12 ................................................1156

– 9 –

P 13 ................................................1159

P 14 ................................................1163

P 15 ................................................1166

Q 1 ................................................1170

Q 2 ................................................1173

Q 3 ................................................1176

Q 4 ................................................1179

Q 5 ................................................1183

Q 6 ................................................1185

Q 7 ................................................1188

Q 8 ................................................1192

Q 9 ................................................1195

Q 10 ................................................1198

Q 11 ................................................1202

Q 12 ................................................1205

Q 13 ................................................1208

Q 14 ................................................1211

Q 15 ................................................1214

Q 16 ................................................1217

Q 18 ................................................1224

Q 19 ................................................1227

– 10 –

Junior Researchers of EARLI (JURE) Pre-conference

Workshop 1

Conversation Analysis

Alan Zemel,

Drexel University, USA

Conversation analysis studies the order/organization/orderliness of social action, particularly those social actions that are located in everyday interaction, in discursive practices, in the sayings/tellings/doings of members of society (Psathas, 1995, p. 2).

- As analysts, we concern ourselves with the local relevance of actions to the participants themselves.

- Why?

- Because we do not possess a privileged view of social action.

- What is the basis for this claim?

- Because we use the same tools, methods and procedures for doing sense making work that are used by the interactants whose actions we observe. CA is an analytical methodology that attempts to describe the actions of participants in terms of the relevances demonstrated by participants in and as their interaction (Ten Have

1999, Psathas 1995, Pomerantz and Fehr 1997). This methodology privileges the perspective of the participants over the analyst’s perspective (Pomerantz and Fehr

1997). Actions are seen as situated within a stream of ongoing action and are sequentially organized. Furthermore, conversation analysts presume that actors design and ‘customize’ their action for the particular circumstances in which they are accomplished. In this workshop, we will discuss some of the basic principles of conversation analysis and ethnomethodology and engage with each other and data in a data session. Through this work, we will consider how to analyse different kinds of interactional data (video, audio, chat logs, etc.), and try to identify specific mechanisms by which participants do sense-making together.

– 11 –

Workshop 2

Cognitive Measurements to Design Effective Learning Environments

Fred Paas

1

and Slava Kalyuga

2

1

Open University of The Netherlands,

2

University of New South Wales, Australia

The aim of researchers in the field of Cognitive Load Theory (CLT) has been to engineer the instructional control of cognitive load to provide the means to optimise cognitive load in learning arrangements. In this workshop, the current state of cognitive load measurement in the CLT context is presented. Specifically, the different measurement techniques are described with respect to their contribution to the theory. First, the construct ‘cognitive load’ and related concepts are defined. Then, the instrumentation for the measurement of cognitive load is explained and an overview is presented of the different measurement techniques that have been used in cognitive load research. Special attention is given to computational approaches for visualizing the relative mental efficiency and involvement of instructional conditions based on mental effort and performance measures. Finally, the role of cognitive load measurement in the advancement of the theory is discussed.

Another point of focus in this workshop is the possibility to use cognitive load based measurements to design learning arrangements that manage and present complex instructional information in flexible, accessible, and learner-individualized formats.

Cognitive load experienced by learners depends significantly on their prior knowledge and experience. Therefore, instructional methods need to change dynamically with alterations in expertise to reduce extraneous and increase germane cognitive load during instruction. To optimise cognitive load in adaptive learning environments, it is necessary to have a simple, rapid measure of learners’ knowledge suitable for real-time assessment of expertise. The workshop will present some preliminary results of developing and applying a schema-based rapid method of evaluating learners’ knowledge structures in several task domains (algebra, arithmetic word problems, kinematics, reading skills) based on cognitive load theory. This diagnostic technique is based on evaluating an immediate content of memory as learners approach a task or solve a problem.The workshop will also discuss the suitability of several methods of combining rapid measures of learners’ performance with measures of cognitive load into integrated indicators of instructional efficiency for optimising adaptive e-learning environments.

– 12 –

Workshop 3

Analysis of Hierarchical Data

Johannes Hartig, German Institute for International Educational Research,

Germany

Hierarchical data structures are a common phenomenon in educational research.

For example, a population of interest may consist of schools, classrooms, and students within classes. For empirical research questions, schools may be sampled from the population and students may be sampled within the schools. In such samples, the data of individuals within the same schools are not independent. Standard statistical analysis techniques like linear regression or analysis of variance do not take into account these dependencies, and results obtained by these methods are biased. Furthermore, data can be collected on all levels of the hierarchy: Variables describing students can be assessed on individual level, and additionally information characterising schools or classrooms may be obtained by teacher questionnaires or classroom observations. Statistical analysis procedures are needed that can take dependencies of data within groups into account and that allow to simultaneously handle measurements made at different levels of the hierarchy. The workshop will give an introduction into the nature and handling of hierarchical data structures. Specifically, the basic ideas of hierarchical linear regression and multilevel structural equation modelling will be outlined and illustrated. Additionally, an overview of statistical software available to conduct analyses with hierarchical data will be given. The workshop is open to all participants who are interested in the topic, though it might be helpful to have some knowledge of qua ntitative analysis techniques.

Workshop 4

Item-Response Theory and Standard-Based Assessment

Nina Jude,

German Institute for International Educational Research, Germany

At present, a trend towards standard-based educational reforms and corresponding assessment approaches can be observed throughout Europe. By setting standards and monitoring students’ levels of competence using standard-based assessment, policy makers hope to identify strengths and weaknesses of the educational system, to increase the overall outcome level and to reduce inequalities. For these evaluations, tests and questionnaires are developed, presupposing that one can more or less directly infer a person’s position on a

– 13 –

competence scale from his/hers responses to the tests. In contrast to traditional testing approaches, latest educational assessments use measurement models basedon

Item-Response-Theory. This allows for a linkage between the specific skills being assessed and the actual competence shown by the test-takers, as those models assume a probabilistic interrelation between answers on test-items and the underlying latent competence. In IRT-based tests, test taker abilities and item difficulties are estimated on the same scale, thus making it possible to describe students’ competencies directly using item demands. Good examples are the tests used in PISA 2000 and 2003 with their linkage of specific items to sections on the competence scales.

This enables sophisticated statements about levels of competencies Ð the standards

- achieved by the students and also about still-to-reach steps in the learning process.

Furthermore, standard-based models of competence that imply theoretical assumptions about item difficulties due to their cognitive demand or their reference to the curriculum can be tested empirically using IRT-modelling.

The workshop will provide an overview over the basic principles of Item-Response-

Theory in contrast to Classical Test Theory. Practical examples for standard-based interpretation of competence scores will be given using samples from the PISA studies. The workshop is open to all participants who are interested in the topic, though it might be helpful to have some knowledge of psychometric statistics.

Workshop 5

Qualitative Research Interviewing: Free Attitude Techniques

Gert van der Westhuizen, University of Johannesburg, South Africa

The focus of this workshop is on the concepts and processes of Free Attitude interviewing. The purpose is to familiarise participants with this approach as a form of non-directive depth interviewing. Emphasis is placed on specific techniques of probing and reflecting of information. The workshop is offered to a group limited in number. It is experiential, and some sense-making activities are included. Apart from the questioning techniques, specific interviewer qualities such as open-mindedness, respect, and listening are emphasised.

– 14 –

Workshop 6

Quality Control in Qualitative Research: Audit for Determining the Trustworthiness of Educational Research

Wilfried Admiraal and Sanne Akkerman,

University of Utrecht, The Netherlands

Qualitative research mostly is characterized by complex studies in which standardized procedures and rules for the collection and analyses of data cannot be used to solve the research problem. Research methodologies in this kind of research often trigger much interpretation of the researcher, collecting real-life data from different sources and analysing this data iteratively. Criteria of transparency (are research decisions made explicit), confirmability (are research decisions justified and grounded) and dependability (are these decisions acceptable according to rules and standards in the particular domain) are more difficult to reach. Up until now, there is not much literature on how to control the quality of this kind of qualitative research.

We suggest an audit procedure described as a procedure in which the scientific quality of the study and the research process is verified by an assessor on the basis of a priori guidelines.In this workshop, the participants will set up an audit trail for their own research project. This means that they have to decide what quality criteria they will use for their research project, which documents they will put in the audit trail, who the auditor will be, what tasks and roles the auditee and auditor will have, how the auditor will perform the audit, and how they proceed with the audit procedure and the results of it. Participants will work in small groups of 3 or 4 on their audit.

Participants are asked to prepare for this workshop by reading a paper on the audit procedure and describing their research project in about 250 words.

Workshop 7

The Process of Writing Scientific Articles

Kirsti Lonka, Karolinska Institutet, Sweden / University of Helsinki, Finland

The problem with teaching academic writing is that important tacit knowledge, silent and procedural in nature, is generally left untaught. According to Boice (1993), part of the problem is that university professors prefer demonstrations of brilliance rather than the acquisition of it, and this preference denies many students the chance to become successful writers. One reason for this is that tacit knowledge is, by definition, hard to teach and difficult to find in written and substantive form. This workshop is very practical. The aim is to describe typical thoughts about scientific writing that may hinder success. The second step is trying to eliminate and modify

– 15 –

these thoughts in a way that promotes productivity and creativity in a research team.

In academic writing, the sociocultural view (Dysthe, 2003) emphasizes that knowledge and understanding are constructed in social interaction. Second, language is the key cultural tool, which mediates learning. Third, learning takes place in ‘a community of practice’ that also includes well functioning group processes. Fourth, because knowledge is always situated, motivation to learn is largely dependent on the learning culture. Every research group creates its own specific learning culture.

I claim that in many cases, the culture does not support the process of writing in the context of educational psychology and education. One problem is that the idea of a research group may appear strange and researchers do not develop social practices that would support publishing in international refereed journals. In this workshop the idea of process-writing is applied in the setting of writing scientific articles. My approach is a combination of courses in cognitive strategies with generative writing and shared revision, whereas practical advice on stylistic rules and grammar are not emphasised. The aim of the intervention is to reveal and then revise practices and ideas of writing that usually remain tacit. The theories and methods are applied by

Bereiter and Scardamalia (1987), Olson (1994), Bjørk & Røisønen (1996), Boice

(1993); Tynjøl?, Mason, & Lonka (2001), and Lonka & Ahola (1995). The idea is to put these theories in action in a very demanding real- life situation. The methods are focused free-writing exercises, using multiple drafts, training peer-feedback strategies, revealing the myths and revising mental models of writing (for instance, by sharing research evidence on writing making tacit knowledge overt to discussion, and reflecting on the participants’ own writing practices and cultures).

Paper session 1: Interaction Patterns and Classroom Practices

Chair/Discussant: Senem Sanal-Erginel, Middle East Technical University,

Turkey

Variation Theory as a Tool for Planning and Analysis of Literacy Lessons in

Swedish Education

Laila Gustavsson, Kristianstad College University, Sweden

In this paper results from a praxis-oriented research project, which aims to focus on teachers’ development during an in service training, is presented. The raised question is if and how teachers’ in service training affects pupils’ learning. By the introduce of a theory about learning, the variation theory, the teachers gradually develop a more powerful understanding about the connections between what they do in the classroom and what pupils actually have possibilities to learn. The more

– 16 –

developed understanding is shown by the way the teachers discuss instruction and pupils’ learning in a theoretical way, together with the outcome of results shown by the pupils when they take tests about the learning object. The used method is learning study. In this paper results from the first and third learning study cycles (2x3 lessons) in Swedish as a first language will be presented, to show how the theoretical insights about learning have changed and developed. Secondly the results of the pupils’ learning outcome are presented to give a picture of if and in what way the teachers’ development is connected to the learning outcome shown by the pupils.

A Methodology for Analysing the Relations between Interaction Patterns and

Modes of Talk in the Classroom

Hayuta Yinon, Haifa University, Israel

As part of a study, that investigated relations between interaction patterns and modes of talk in frontal language lessons in one seven grade classroom in northern Israel, a unique method of analysis was developed, which allowed for examining modes of talk in relation to interaction patterns in the classroom. The data comprised of interviews with language teachers and pupils, observations of lessons, and collection of school documents. This qualitative study adopted a classroom discourse analysis approach, and followed grounded theory procedures. As a result, the new method of analysis was developed. The method includes several stages. The main stage is based on identifying ‘interaction cycles’ in the observations, a new construct that emerged from the study, and which defined the completion of an entire cycle of the

IRE/F sequence. The uniqueness of the analytical tool can be attributed to three factors. First, it is pioneer in that it allows for analysing classroom discourse in relation to interaction patterns and modes of talk. Second, the method allows for tracking of the inner dynamics of classroom interaction, which is usually only recognized by its actual participants. Third, the method allows for interpreting any kind of event in the classroom.

Innovating Mother-Tongue Education in Pre-Vocational Education: Fostering a

Community of Learners

Anne Toorenaar and Gert Rijlaarsdam, University of Amsterdam, The Netherlands

In Dutch pre-vocational education teachers of mother-tongue education (L1) are focussed to innovate their educational practice. They deal with students with motivational problems and a poor level of communicative competences. To improve this,

L1-teachers look for a meaningful and rich context Ð linked up with the vocational

– 17 –

program - in which students cooperatively acquire and reflect on their communicative competences in realistic whole language tasks. By doing so, students should be able to connect the different practices Ð L1 ?vocational program ?internship Ð in which they participate as language learners and users. This innovation links up with the concepts of ‘community of learners’ and ‘community of practice’. In the next two years, researchers and teacher will furtherance the L1-innovation and elaborate a conceptual framework based upon the two concepts of ‘community of learners’ and ‘community of practice’. This paper shows the results of the first case study. In two schools (4 teachers, 4 classes 9th grade) data has been gathered by observations and stimulated recall interviews. Results shows in what way and to what extent learning activities of students and instructional behaviours of teachers Ð needed to transform the classroom into a community of learners - have already taken shape.

Paper session 2: Competency Assessment

Chair/Discussant: Judith Gulikers, Open University of the Netherlands

Students’ Self-Assessment on Their Competencies in Mathematics - Development of the Questionnaire COMA

Holger Gaertner, Free University Berlin, Germany

The goal of this study is to construct a reliable and valid questionnaire that provides teachers with information about several mathematics-specific competencies of their students. These competence domains (subject, methodical, personal and social) correspond to recent German curricula for teaching mathematics. The four competencies were operationalized as follows: The methodical competence comprises work-organization and self-assessment ability. The social competence includes cooperation, communication as well as the ability to handle criticism. The personal competence includes mathematics-specific interest, self-concept and self-efficacy.

The subject-specific competences were measured with items according to the educational standards of the curriculum. 24 classes (grade 5-8) participated in this study

(N=499). The reliabilities for the different scales were between .63 and .87. The confirmatory factor analysis (CFA) supported the assumption that the questionnaire measures the four domains of competence (RMSEA=.08; CFI=.96). In addition, the CFA revealed high inter-correlations between the four domains of competence, which were particularly high for elementary school students. In comparison, Secondary school students judged their subject competence as more independent from the other competencies. As expected, the class means differed by grade, especially on the subject competence. Teachers participating in the study confirmed that all

– 18 –

relevant aspects of the new curriculum were covered.

Output Orientated Course Evaluation: Berlin Instrument for Evaluation of Students’ Competencies

Edith Braun, Free University Berlin, Germany

Because of the lack of output orientated questionnaires for higher educational courses a new evaluation instrument has been developed. This instrument, so called

Berliner Evaluation Instrument of Students? Competencies, measures the subjective gain of competencies in four areas. The areas of competencies are following: knowledge, methodology, social skills (meaning communication and cooperation skills), and personal proficiency. In winter 2004/05 this instrument was used in 396 courses at six universities in Germany. 1901 student responded which is a total of

3413 completed questionnaires because students visited several courses and filled it out more often. The classical test criteria are fine; especially the confirmatorical factoring verified the theoretical areas of competence. Furthermore the internal consistencies of the scales are good.

The Wheel of Competency Assessment: Presenting Quality Criteria for Competency Assessment Programmes

Liesbeth K. J. Baartman, Theo J. Bastiaens, and Paul A. Kirschner,

Open University of The Netherlands

Instruction and learning are increasingly based on competencies, increasing the call for assessment methods to adequately determine competency acquisition. One single assessment method appears not to be sufficient, necessitating a Competency Assessment Programme (CAP) that combines different methods, from classical methods to recently developed alternatives. A problem is that many quality criteria for classical methods cannot be applied to CAPs, which use a combination of different methods. This article presents a model of ten quality criteria for CAPs: authenticity, cognitive complexity, meaningfulness, transparency, fairness, transparency, directness, educational consequences, costs & efficiency, reproducibility of decisions and comparability. An expert meeting where participants entered quality criteria into a

Group Support System was used to test this model. The results confirm the framework (9 out of 10 criteria) and expand it with three additional criteria: fitness for purpose, acceptability, and fitness for self-assessment. Based on the results, and adapted and layered framework is presented.

– 19 –

Paper session 3: Teacher-Classroom Interaction

Chair/Discussant: Stavroula Philippou, Primary School Teacher/Educational

Researcher, Cyprus

Learning from Ones Practise: Teacher’s Growth during In-Service Training in the

Classroom

Anna Wernberg, Kristianstad College University, Sweden

In service training is in many ways important for teachers’ development in their profession. In many cases one or two teachers are sent to a continuation course and at the best they return with some idea of how to improve the practise. The problems then occur of how to implement these new ideas to the other teachers who did not attend the course, and also the problem of putting the learned theories into practise.

In this paper I will show how a teacher develops through in service training in the classroom. The teacher participated in a research project, The pedagogy of learning, where one aim is to develop teachers’ learning. The method used in this project is

Learning Study. Learning Study is an iterative process where teachers in collaboration with a researcher plan and conduct three research lessons, with an intention to improve educational settings by revision based upon the analyses of the data (video recording from the research lesson and post-test) which indicates the students’ development.

The Same Teacher, the Same Intended Curriculum, but Different Classrooms: Is It the Same Implemented Curriculum?

Tammy Eisenmann and Ruhama Even, Weizmann Institute of Science, Israel

The focus of this study is the implemented curriculum and the role played by the classroom in determining the implemented curriculum. We examine a case study of one teacher who teaches the same intended mathematics curriculum in two 7th grade classrooms, each in a school with different socio-cultural background. Main data sources include observations of the teaching of the beginning of the topic of equivalent algebraic expressions (about 20 lessons in each class). The data are analyzed both qualitatively and quantitatively. The paper presents an analysis of two elements of the implemented curriculum: (1) Instructional materials, (2) Instructional organizations. The main findings show there were similarities and differences in the way the mathematics materials were used in the two classes and in the main activities that took place during the lessons. We propose some explanations to the differences we found that connect between the implemented space (schools, students and

– 20 –

teacher) and the implemented curriculum.

Changing a Primary Teacher’s Interaction Strategies with Second Language

Learners of Mathematics

Gerda-Eva Swank, Utrecht University of Professional Education, The Netherlands

Active participation in classroom interaction is considered to be crucial for children’s language development, both in specific language teaching as in content lessons delivered through the medium of the second language. A growing number of studies, however, report on teachers discouraging student participation, tending to a teacher fronted approach. A case study explored how two primary teachers changed their interaction patterns through inventions. This paper reports on findings with the changing of the interaction strategies of a math teacher in a primary school with second language learners.

The research questions were

1 Which strategies do teachers use to promote students’ active verbal participation in classroom interaction in content teaching?

2 How do teachers’ interaction patterns change during professional development activities?

3 How does their thinking change during professional development activities?

4 Which interventions could explain the changes in teacher behaviour and thinking?

Data on teachers’ behaviour consisted of videotaped lessons. Data on teachers’ cognitions were collected trough stimulated recall interviews and a concept map.

The teacher received feedback on her work from an in-service trainer. The results show that through the weeks, the teacher succeeded in opening up more discussion and stimulating students to talk about math in Dutch. Her behaviour and cognitions are closely related and changeable.

Paper session 4: Student Attitudes and Beliefs

Chair/Discussant: Nina Jude, German Institute for International Educational

Research, Germany

You Say TomAto, I say Tomato. How Different Subjects Foster Different Beliefs in the Students that Study Them

Cathal Siochru, Liverpool Hope University, UK

Recent studies on epistemology, an individual’s beliefs about the way knowledge

– 21 –

works, suggest that one individual can have two independent sets of epistemological beliefs relative to two different knowledge domains (Hofer, 2000). This study aimed to assess the reliability of Hofer’s (2000) findings and her Epistemological

Belief’s Questionnaire (EPQ). It looked at the epistemologies of different students studying one of five different subjects, as well as looking at the epistemology(s) of students studying two of those subjects at the same time. The results of the reliability analysis showed that only three of Hofer’s four factors were reliable (Certainty : a=0.709; Source : a=0.651; Justification : a=0.538; Simplicity : a=0.197).

A MANOVA confirmed that different students from different subjects had different epistemologies (Certainty : df=4, F=17.780, p< 0.001; Source : df=4, F=15.425, p< 0.001; Justification : df=4, F=3.101, p<0.05). However the results from the

Dependent T-Tests comparing responses from the same individual regarding their epistemological beliefs in Psychology and Sociology didn’t show evidence of two significantly different epistemologies (Certainty : t=-1.084, df=27, p=0.288; Source

: t=-0.904, df=27, p=0.374; Justification: t=-1.039, df=27, p=0.308). Results are discussed in terms of the reliability of the EPQ and the issue of homogeneity of epistemological beliefs across domains.

The Nature of Life Meanings of Pre-adolescents

Martin Ubani, University of Helsinki, Finland

The aim of this study is to examine the spirituality of the Finnish 12-13 year-old pre-adolescents (N=26). The recent studies in RE emphasise that spirituality is a universal human characteristic. In other words, people are spiritual regardless of their religious orientation. In the Nordic context spirituality is often connected to search for meaning and existential questions. In this study, spirituality is explored by examining the children’s views on the meaning of life. The results echo the recent studies that show that children have a rich spiritual life. The meanings that the students gave to life included three dimensions. They are called referential, relational and revelatory life meanings. It is suggested that these dimensions are integral elements of children’s spirituality. The study uses a narrative phenomenological approach.

Environmental Knowledge and Attitudes of Turkish Elementary School Students:

The Effect of Gender and Parents’ Education Level

Elvan Alp, Hamide Ertepinar, Ceren Tekkaya, Middle East Technical University,

Turkey and Ayhan Yilmaz, Hacettepe University, Turkey

– 22 –

The present study aims to (a) determine Turkish elementary school students’ environmental knowledge and attitudes, (b) examine the effect of gender and parents’ education level on children’s environmental knowledge and attitudes, (c) explore how children’s environmental friendly behavior related to environmental knowledge, environmental affects and behavioral intentions. The data was collected by the administration of Children’s Environmental Attitudes and Knowledge Scale

(CHEAKS) as a measuring instrument to 782 eight grade students from 18 randomly selected elementary schools. The results indicated that children had favorable attitudes toward the environment but their environmental knowledge was fragmentary and incorrect. There was a significant effect of gender on environmental attitudes of students in favor of females. However, no significant effect of gender was found on environmental knowledge. Environmental knowledge was significantly related to parents’ education level. Results showed that the mean knowledge score of students with fathers having university education was significantly higher than that for students with fathers having less than university education. The same general trend was observed for the mothers’ education level. No significant effect of parents’ education level was found on students’ attitudes toward the environment. Environmental friendly behaviors of the students were highly correlated with behavioral intentions and environmental affects but there was no statistically significant relationship between students’ environmental knowledge and environmental friendly behaviors. These results indicated that behavioral intentions and environmental affects might be the determinants of environmental friendly behaviors.

Paper session 5: Evaluating the Use of ICT in Education

Chair/Discussant: Herman J. Abs, German Institute for International

Educational Research, Germany

Intentions of Integrating ICT in Campus-Based University Teaching: Qualitative

Results and Further Research Prospects

Claudia Hauswirth, University of Dortmund, Germany

Up to spring 2004 university teachers with different experiences in using educational media and in strategic positions in a German campus-based university were interviewed about their experience and assessments of instructional benefits of ICT.

One aim of the study was to identify intentions of Integrating ICT into the teaching context of university teachers. Four dimensions of intentions could be identified: teaching strategy-oriented, learning-strategy-oriented, social-communicative and strategic. This paper summarizes the main findings of the dissertation project and

– 23 –

refers to further research prospects.

Influences of the Integration of Cyprus into the European Union on ICT Policy and the Construction of ‘ICT in teaching and learning’ in Cypriot Primary

Schools

Christina Hadjithoma, University of Bristol, UK

Beyond doubt, the attempts for introducing ICT in educational systems in European countries have been increasingly taking place, and this phenomenon is rationalised through arguments that can be found in various discourses. Focusing on the way that ICT policy in Cyprus was influenced by the European Union agenda, and the implications that this had on the construction of ‘ICT in Teaching and Learning’ by policy-makers and educators in Cyprus, this paper claims that the study of related discourses and policies is crucial, when studying ICT and learning and instruction.

The findings of the research, which employed qualitative methods of analysis, that is discourse analysis of policy-documents (EU and Cypriot official documents) and interviews with key officials who are involved in the implementation of ICT policy are discussed in this paper. The main thesis of the paper is that the way that ICT is embedded and used in teaching and learning is a result of a combination of influences, arising from national, international and global forces, and although this has implications mostly for policy-oriented research, it is also important for research into the development of future models of teaching and learning with ICT.

Evaluating Computer Technology Integration in Cyprus Elementary Schools

Nikleia Eteokleous, P.A. College, Cyprus

The study evaluated the current situation in Cyprus elementary classrooms regarding computer technology integration. It examined how Cypriot elementary teachers use computers, and the factors that influence computer integration in their classroom practices. To address the study’s research questions, an evaluative case study design was applied. It employed a mixed method approach through the usage of structured questionnaires and semi- structured, open-ended interviews as the major methods of data collection. Quantitative and qualitative data gathered from a sample of Cypriot teachers in which high, moderate, and low computer use teachers were identified. The results of the quantitative analysis indicated that while Cypriot teachers use computers rather extensively for their own purposes, they use them less frequently in their classes. When they do use them in their classes, it tends to be in a rather sporadic fashion, more as extras or fancy chalkboards than as true

– 24 –

learning tools. Few teachers were found to use computers in any sort of constructivist or progressive way. Qualitative analysis summarizes the factors that influence teachers in applying computers in their classroom practices. At the end, the study provides a discussion of the results, implications for policymakers and educators, as well as questions for further study.

Paper session 6: The Ability to Comprehend Written and Spoken Language

Chair/Discussant: Nele McElvany, Max Planck Institute for Human

Development, Germany

The Influence of Learning to Read Music On the Ability to Learn how to Decode

Verbal Language

Yehudit Carmon, Bar-Ilan University, Israel

Reading acquisition remains a worldwide problem, unresolved through traditional reading instruction approaches. Analyzing music and verbal reading differences, we found basic shared components: audio-visual integration; successive signs in a specific direction; the development of hearing memory to accumulate signs in the working memory until acquiring the cognitive/emotional message, and perceiving the alphabetical principle. The reading instruction methods do not pay enough attention to these elements. This fact, together with an overflowing quantity of reading components, might cause initial reading problems. With a new musical method,

Toy Musical Notes (TMN,) using just a few signs, the shared skills are explicitly acquired, creating a child’s first reading scheme in the brain. The verbal reading becomes a second, much easier reading. Hundred and fifty preschool children were tested in three intervention groups: the TMN method, conventional music, and nonmusic enrichment. Their text reading acquisition parameters were tested in first grade. The TMN group findings were significantly higher than the conventional music group, and the latter scored higher than the non-music group in all reading parameters: number of mistakes; comprehension; vocal reading time, and more.

The research found the TMN method especially effective for the acquisition of reading skills.

– 25 –

Verbal Working Memory Skills as Predictors of Listening Comprehension of Text in the Preschool and Elementary School Years

Elissavet Chrysochoou, Aristotle University of Thessaloniki, Greece

Childrens’ language comprehension abilities have been investigated in relation to various cognitive skills. Significant research has been conducted in the frame of

Baddeley and Hitch s (1974) working memory (WM) model. The model s central executive (CE) is thought to be involved in syntactic and semantic processing, in storing the products and in retrieving relevant information from long-term memory.

It would, therefore, be expected to significantly contribute to text comprehension.

The models phonological loop (PL) is thought to limit its contribution to language comprehension in maintaining a phonological record that can be consulted during off-line processing. Currently, a Working Memory Test Battery for Children (Pickering and Gathercole, 2001) has enabled a fine-grained WM assessment from the preschool years. Its PL and CE tasks, a vocabulary task and texts with accompanying comprehension questions were translated to Greek and were adapted. Tasks were administered to 180 children, divided into three groups (C.A. 5:7, 7:7 and

9:7, respectively). Our aim was to shed further light on the processing and storage functions involved in young children s listening comprehension of text. Statistical analyses indicated the structural validity of the adapted assessment tools and their sensitivity to age-related differences. The results of stepwise and hierarchical regression analyses are discussed and implications are indicated.

The Relationship Between Rapid Automatized Naming Components and Reading

Ability:A Longitudinal Study from Kindergarten Until the End of Grade 1

George K. Georgiou, University of Alberta, Canada

The present study examines (a) how RAN components Ð articulation time, pause time, and pause time consistency Ð develop from kindergarten to the end of first grade, (b) how RAN components are related to different reading measures, and (c) which RAN components differentiate poor from skilled readers. Sixty-two children were administered RAN tasks in kindergarten and at the beginning and end of grade

1. Performance on Colour and Letter Naming was recorded and analysed. Reading accuracy and reading fluency measures were used as the criterion variables. Results indicated that pause time developed significantly from kindergarten to the end of grade 1 and was highly correlated with both reading accuracy and reading fluency measures. Articulation time did not develop and was only weakly correlated with the reading measures. Pause time consistency shared most of its variance with the

– 26 –

pause time. Finally, only pause time differentiated between poor and skilled readers.

Paper session 7: Contextual Learning/School Management

Chair/Discussant: Debbie Bond, Syracuse University, USA

Learning in a Personal-Context: Levels of choice in a Free-Choice Learning Environment at Science and Natural History Museums

Yael Bamberger and Revital T. Tal, Technion Israel Institute of Technology

The study aims to characterize contextual learning processes during class visits to science and natural history museums. It is based on the assumption that outdoors learning is different from classroom-based learning and strongly relies on personal context. Acquiring learning skills in free choice outdoors environments such as science museums is extremely important. Therefore, the learning activities during class visits to museums are studied, focusing on levels of choice provided. The study follows up four museums of different size, location, visitor number and foci.

Participants are 700 students in grades 4-8. A descriptive-interpretative approach was adopted, with data sources comprising observations, semi-constructed interviews with students and teachers, and museums’ working sheets. Four levels of choice were identified within the museums’ provided activities, which affect the effectiveness of learning: no choice, some choice, subject choice and free choice.

The effectiveness of learning was evaluated as well by questions and answers of the students, interactions, management of group behaviour, off-task behaviour, linkage to students’ prior knowledge and experience, and linkage to the school’s science curriculum.

Learning a Foreign Language - a Purpose or a Means?

Tatiana Olevsky, Hebrew University of Jerusalem, Israel

Learning a foreign language in an early age is although more productive, but at the same time it sets a number of motivational problems. When formulating a purpose of studies, if we move the focus from improving language skills per se to developing child’s leading activity (as defined by Leontiev) that corresponds to his/her age, we can significantly diminish the motivational problem that naturally exists in the traditional approach. The paper describes an application of the principle explained above. Simple arithmetic tasks have been used in order to teach English to a group of 5-year old Israeli kids.

– 27 –

EARLI 2005 Conference

A 1 23 August 2005 14:30 - 16:30 Room A008

Symposium

Collaborative learning

FACTORS OF EFFECTIVE COOPERATIVE AND COLLABORATIVE

LEARNING

Chair: Anne Huber, University of Education Weingarten, Germany

Organiser: Anne Huber, University of Education Weingarten, Germany

Discussant: Jan Terwel, Vrije University Amsterdam, Netherlands

Guenter Huber, University of Tuebingen, Germany

This symposium elaborates on factors influencing the effectivity of cooperative and collaborative learning. Cooperative and collaborative learning is organized in various learning environments, varying from primary and secondary school to college and professional training contexts (e.g. teacher training). In these contexts learning aims at knowledge building, problem solving as well as at attitude change. Three different types of factors can be distinguished: (1) characteristics of the individual learner, for example cognitive ability, previous knowledge, gender or ethnicity. (2) internal and external characteristics of the cooperative learning method, for example resource interdependence versus resource independence or different instructions on how to go about learning and (3) characteristics of the learning process, for example whether learners are involved in careful problem definition, cognitive conflicts or in social comparison processes. We hope to shed light on which developments are necessary in order to improve scientifically and educationally significant outcomes of research on cooperative and collaborative learning.

Resource interdependence and note taking as a means to improve peer learning

Celine Buchs, University of Geneva, Switzerland

Fabrizio Butera, University of Lausanne, Switzerland

Previous studies have indicated that resource interdependence (information distribution) elicits two different dynamics in regard to student interactions and learning. Working on complementary information produces more positive interactions;

– 28 –

however, a good quality of information transmission is needed in order to guarantee learning. Working on identical information produces confrontations of points of view as well as a focus on social comparison of competence, which is threatening for one's own competence; this threat is detrimental to learning. A pilot study underlined that the quality of information transmission was higher when students could take notes. On the other hand, not taking notes permitted to reduce both confrontations and competition reported by students working on identical information. In the present study, students worked cooperatively in dyads with the same partner during

3 sessions, either on complementary information (positive resource interdependence) or on identical information (resource independence); they were allowed vs. not allowed to take notes. Results indicated that perception of confrontations and of competence stake was greatest when students worked on identical information while taking notes; students also reported the lowest efforts in explaining information in that condition. As for individual learning, the marginal interaction between resource independence and note-taking indicated that when students worked on identical information, they performed better without taking notes, whereas when students worked on complementary information, not taking notes was not beneficial (and the listeners' learning could be disturbed when their partner could not take notes). Directions for educational settings include: a) when students work on identical information (e.g. the same pedagogical materials), not taking notes could emphasize the efforts needed and reduce the stake of competence, which in turn favours performance in that condition; b) when students work on complementary information, a good transmission of information is required, and therefore taking notes could be necessary.

The role of strategy instructions and cognitive ability for learning with the Jigsaw method

Anne Huber, University of Education Weingarten, Germany

The Jigsaw method is a cooperative learning method. In a first phase learners become experts for different parts of the learning subject. In a second phase experts teach each other their expert knowledge. In a third phase learners initiate and monitor activities for elaborating and deepening their knowledge. The paper presents a study, which tries to find out the role of learning instructions and cognitive ability within the Jigsaw method. From self-determination theory of motivation and from the cognitive-elaboration perspective of cooperative learning it can be derived, that strategy instructions are very important in order to obtain positive results in achievement, intrinsic motivation and perceived competence. Strategy instructions

– 29 –

help learners to learn more thoroughly and therefore have positive effects on the expected outcomes (see above). From research on cognitive ability and achievement we know that students with higher cognitive ability usually reach higher levels of achievement than students with lower cognitive ability. Cognitive ability therefore is an important factor for achievement results. In order to test these hypotheses a 2 x 2 factorial design was realised with the factor ìlearning instructions: ìJigsaw with learning instructions versus ìJigsaw without learning instructions and the factor

ìcognitive ability: ìhigh cognitive ability versus ìlow cognitive ability. The Jigsaw method was used over eight weeks in four secondary school classes in order to teach biology. Results of this study confirm the expectations largely. Most of the hypotheses could be approved. Unexpected results are discussed within the framework of self-determination theory of motivation. Important advises for teachers can be derived from the study. It can be shown how to use the Jigsaw method best in order to account for interindividual differences.

The structure and dynamics of the collaborative problem solving process

Hannu Soini, University of Oulu, Finland

Katri Jaemsae, University of Oulu, Finland

Antti Rantanen, University of Oulu, Finland

Matleena Sedergren, University of Oulu, Finland

Our earlier findings have shown that peer consultation is a promising method when we try to increase higher education students problem solving skills or commitment to their studies. Peer consultation takes advantage of students mutual co-operation, dialogue and reflectivity by encouraging them to accept more responsibility for learning by their peers. We have examined the impact of peer consultation on the case problem solving process among medical students on a paediatrics course.

The aim of this method is for the whole group to be able to put themselves in the position of one of the group members for a certain period of time, to listen to this group member and to really consider his/her problems. Consultative work means creative work in small groups under the guidance of a supervisor. Peer consultation helps students to learn for themselves. It is also a tool with which the students can evaluate their own learning process. The contribution of peer consultation to the problem oriented learning is to concentrate on the careful definition of the problem and to identify collaboratively the different aspects of the question. In this paper we will present our preliminary findings of the underlying structure and dynamics of the problem solving process in different learning environments.

– 30 –

External and internal characteristics of cooperative learning

Mary Koutselini, University of Cyprus, Cyprus

The study presents the results of an educational intervention during in-service training of secondary school teachers. The aim was to help teachers simulate cooperative learning and construct the characteristics that differentiate it from plain group work. The results of the study indicated that teachers have negative attitudes towards cooperative learning because they dont know how to ensure collaboration, coherence and interaction of the members of the group; their attitudes changed gradually during the study, and after the experience that only cooperative learning with concrete internal and external characteristics resulted to learning outcomes for

ALL students.

Computer Supported Collaborative Learning: Individual differences in participation and outcomes

Fleur Prinsen, Vrije University Amsterdam, Netherlands

Monique Volman, Vrije University Amsterdam, Netherlands

Jan Terwel, Vrije University Amsterdam, Netherlands

Computer Supported Collaborative Learning is often presented as a promising innovation. But CSCL faces some well known problems together with some more specific new challenges. Some of the already mentioned problems in co-operative learning research appear in a new fashion in the context of CSCL. Apart from the question whether working with CSCL generates satisfying learning outcomes, an important question is whether all participants profit from collaboration with the computer as a communication device. Is it possible that especially high achievers profit, as is often noticed with innovative learning environments. How do low achievers participate in this environment? Do the gender differences that are found in co-operative learning research also occur in CSCL? And what is known about the relationship between social-ethnic background and participation in CSCL? Firstly, this paper will review what is known in the literature about differences in participation and learning outcomes of students differing in gender, ability, pre-knowledge and social-ethnic background when working in a CSCL environment. We argue that, if there are differences between these student categories in participation in

CSCL environments, there should be more attention paid to ways of including students in the collaboration in a way that can lead to good results. This is one of the most important lessons from a long research tradition in co-operative learning. Secondly, this paper offers a description of a new, special designed program for CSCL

– 31 –

for 10 to 12-year - old students. The content of the program is about nutrition and health within the integrated subject matter World orientation and Science. Thirdly, this paper describes the implementation and effects of the experimental program in

5 classes with in total 120 students in primary schools. Special attention is given to the differential effects of pre-knowledge, gender and ethnicity on the participation and learning outcomes of students.

A 2 23 August 2005 14:30 - 16:30 Room A009

Symposium

Teaching and Instructional Design

FACTORS INFLUENCING THE STABILITY OF THE MODALITY EF-

FECT

Chair: Geraldine Clarebout, K.U.Leuven, Belgium

Organiser: Geraldine Clarebout, K.U.Leuven, Belgium

Jan Elen, K.U.Leuven, Belgium

Discussant: Wolgang Schnotz, University of Koblenz: Landau, Germany

This symposium aims at gaining insight in the stability of the modality effect. This effect is related to the split attention effect that occurs when learners have to attend to and integrate multiple information sources (e.g., Mayer, 2001; Sweller &

Chandler, 1994). If the different information sources all require visual information processing, the visual information processing channel might get overloaded.

The modality effect postulates that providing some of the information in an auditory format reduces the load of the visual information processing channel. Different studies in multimedia learning environments (e.g. Moreno, Mayer & Lester,

2000, Mousavi, Low & Sweller, 1995) provide evidence for this effect. Students get higher scores on recall and transfer tests when instructional messages are provided through narration (and visuals) rather than only visually.. However, most of these studies have been carried out in more closed environments', with little learner control and a well-defined problem. In this symposium, the different contributions address the modality effect in different contexts and/or in relation to different variables. In a first contribution by Moreno a theory of multimedia learning is proposed and the modality principle is tested for three different media. Paas et al. investigate the relation between the modality principle and pacing. Clarebout and Elen studied the modality effect in an open learning environment. The final contribution by Elen

– 32 –

and van Gorp researched the relation between modality and learner control.The discussant critically reflects on the different research studies and findings and tries to integrate the different results to come up with a proposal for a research agenda.

Does the modality principle hold for different media?

Roxana Moreno, University of New Mexico, United States

Does the modality principle hold for different media? The goal of this presentation is to answer this question by examining the classic distinction between the roles of media versus method in promoting learning. To this end, I first propose a cognitive theory of multimedia learning from which a set of instructional design principles are derived. Then, for three different media (i.e., multimedia explanations, animated pedagogical agents, and virtual reality), I review a set of experiments where one of such design principles was tested: the modality principle. Finally, I discuss the theoretical and practical implications of the findings.

Observational learning from cognitive models: uncovering the cognitive processes

Pieter Wouters, Educational Technology Expertise Center, Netherlands

Fred Paas, Educational Technology Expertise Center, Netherlands

Jeroen van Merrienboer, Educational Technology Expertise Center, Netherlands

Theories of multimedia learning advocate the use of dynamic visualizations in combination with explanatory verbal text. Information that has been encoded in two ways can be retrieved and used more easily. Modern educational theories also regard the application of cognitive modelling with its focus on authentic tasks as a promising instructional approach. Cognitive modelling not only shows what is happening during the performance of a task, but also why this is happening. The application of dynamic visualizations in cognitive modelling offers interesting affordances because conceptual or abstract processes can be visualised. Cognitive load theory contends that instruction should minimize the extraneous cognitive load. One of the most investigated phenomena regarding extraneous cognitive load is the split-attention effect. With respect to dynamic visualizations in cognitive modelling several split-attention effects may occur. Beside spatial split-attention, which occurs when visual material and its explaining text are separated from each other, dynamic visualizations may impose specific split-attention effects, such as the intra-representation split-attention and transient split-attention. Two design guidelines are often proposed in order to meet these split-attention effects. The modality principle, which states that explanatory text should be presented orally,

– 33 –

and pacing, which enables learners to adapt the presentation speed to their cognitive needs. In this study, a 2*2 factorial design was used to investigate the relation between modality and pacing. The focus, however, was not on the effect on transfer performance, but on the uncovering of cognitive processes that underlie the difference in transfer performance. Two valuable methods were used in this exploratory study, the think aloud method and the analysis of eye movements. The analysis of data will be completed in April 2005 and final results are presented at the EARLI.

Modality effect in open learning environments

Geraldine Clarebout, K.U.Leuven, Belgium

Jan Elen, K.U.Leuven, Belgium

Different studies of Moreno and Mayer (e.g., Moreno, Mayer & Lester, 2001) revealed a modality effect of pedagogical agents on learning: an agent using narration lead to deeper learning than an agent using on-screen text. The aim of the present study is to test the modality effect in open learning environments with an agent providing metacognitive support. Participants were 42 first year educational sciences and psychology students. A quasi experimental design was used. In this study no confirmation was found of the modality effect, suggesting that this effect cannot be generalised to open learning environments where the agent provides support on a metacognitive level, rather than a content specific level which is the case in the

Moreno et al. studies.

The modality effect and primary school learning materials

Jan Elen, K.U.Leuven, Belgium

Els Van Gorp, K.H.Kempen, Belgium

Numerous experimental settings have illustrated the modality effect. These studies have a number of clear characteristics: respondents are mainly higher education students, information addresses procedures or processes, and there is no learnercontrol with respect to the pace of the presentation. In order to explore the boundaries of the modality effect in this study an experiment was carried out with primary schools children and 12 conditions: three conditions with respect to the relationship between text and pictures (text separate, text integrated, and text auditory), two image modalities (static versus dynamic) and two conditions with respect to learner-control. These twelve conditions were tested for two sets of materials (one procedural and the other purely factual), resulting in a total number of 24 conditions. Each condition had 10 participants. Learning gains (knowledge and transfer)

– 34 –

were calculated and total watching time as well as watching time for each screen were registered. Analyses of variance revealed no main effects of independent variables for learning gains with respect to retention or transfer.

A 3 23 August 2005 14:30 - 16:30 Room A109

Symposium

Social Interaction in Learning and Instruction

INVESTIGATING CLASSROOM INTERACTION: METHODOLOGIES

IN ACTION (PART I)

Chair: Paivi Kristiina Kumpulainen, University of Oulu, Finland

Organiser: Paivi Kristiina Kumpulainen, University of Oulu, Finland

Margarida Cesar, Universidade de Lisboa, Portugal

Discussant: Michele Grossen, University of Lausanne, Switzerland

This symposium realized in two sessions (Part I and Part II) introduces strands of research on classroom interaction whose logic of inquiry produce different approaches, analyses and interpretations of social interactions and discourses in contemporary classroom settings. The methodological approaches which are introduced and discussed in the symposium within the context of empirical investigations of classroom interactions draw on studies of language and discourse, ethnography, as well as on sociological, psychological, and domain-specific analyses. In recognizing the complexity and challenges in mapping out the complex research territory focusing on classroom interactions, the prime goal of the symposium is to build a complimentary context for discussion of the ways in which different approaches to classroom interaction are realized and how they produce different analyses because of their purpose, conceptual framework, and methodological choice. The illumination of diverse approaches to classroom interaction and discourse is believed to demonstrate the potential and challenges each strand of research is likely to bring towards understanding the psychological, social and cultural life of the classroom and how these mediate the situated practice of teaching and learning in today's schooling.

– 35 –

The analysis of talk as data in educational settings

Neil Mercer, Open University, UK, United Kingdom

Researchers from a range of disciplinary backgrounds - including psychologists, sociologists, anthropologists, linguists - have studied talk in educational settings, and they have used a variety of methods to do so. The methods they have used reflect their research interests and orientations to research. That is, particular methods are associated with particular research perspectives or approaches; and each approach always, even if only implicitly, embodies some assumptions about the nature of spoken language and how it can be analysed. My main aim in this paper is to provide a basic guide to ways of analysing talk which can be used in educational research. I begin with a review of approaches and methods, and then discuss some of the key issues involved in making methodological choices. Given limited space, I have not attempted to go into detail about any of the methods involved, but instead I shall highlight their key features and compared their strengths and weaknesses. My review of existing methodologies describes eight approaches which have provided analytic methods for educational research. These are (1) systematic observation, (2) ethnography, (3) sociolinguistic analysis, (4) linguistic discourse analysis, (5) sociocultural discourse analysis, (6) conversational analysis, (7) discursive psychology, and (8) computer-based text. These categorizations need to be viewed as conceptual abstractions since, in practice, approaches overlap, and researchers often (and increasingly often) use more than one method. The paper finishes by encouraging researchers interested in educational talk to consider a range of methodological options for addressing any research questions, to make explicit one's reasons for selecting or combining methods, and most of all to avoid reducing methodological problems to simplistic choices, such as that between 'quantitative' and 'qualitative' methods.

– 36 –

(Re)formulating opportunities for learning and academic identities: Interactional ethnography and the study of intertextuality and consequential progression

Judith Green, University of California Santa Barbara, United States

Elizabeth Yeager, University of California Santa Barbara, United States

Laura Hill-Bonnet, University of California Santa Barbara, United States

Audra Skukauskaite, University of California Santa Barbara, United States

Carol N. Dixon Dixon, University of California Santa Barbara, United States

LeAnn Putney, University of Nevada, United States

Ana Floriani, Illinois Wesleyan University, United States

This paper presents Interactional Ethnography, an approach to studying the social construction of everyday life in classrooms and ways in which classroom practices are supported and/or constrained by school, district, state and national policies (Castanheira, Crawford, Green & Dixon, 2002; Dixon, Green, Yeager, Franquiz & Baker, 2000). Our goal is two-fold: 1) to demonstrate how Interactional

Ethnography's constitutes a logic of inquiry, entailing multiple layers of analyses needed to make visible knowledge and practices constructed by the collective, and the ways in which individuals-within-the-collective take up and use these opportunities to (re)formulate local knowledge and academic identities; 2) to demonstrate how changes in policies at the school, district, state and national levels support and/or constrain the opportunities teachers are able to construct with their students.

To achieve these goals, we focus on the social construction of social science in a

ìbilingual classroom in three years (1993-1994; 1995-1996; 1996-1997) prior to a policy shift to ìEnglish Only. The contrastive analyses of years pre-post a major reform (1998-1999) make visible lost opportunities for learning and academic identity when Spanish was no longer a resource for constructing public texts. Contrastive analyses show how change(s) in multiple policy contexts led to changes in the ways in which the teacher was able to formulate academic identity potentials and opportunities for student learning to be social scientists (e.g., historians and ethnographers), to provide equity of access across years and to create common knowledge for her students. Through contrastive analyses of social science in this 5th grade bilingual class across years, we illustrate the principles of practice guiding the multiple layers of analysis necessary to explore both the in-the-moment, face-to-face, discursive construction of life, and the across time construction of opportunities for learning and identity formulation made available to the group and individuals within the group.

– 37 –

Framing activities: exploring the demands on pupils in new classroom activities

Asa Makitalo, Gothenburg University, Sweden

A. Jakobbson, University College of Malmo and Lund University, Sweden

Roger Saljo, Gothenburg University, Sweden

The purpose of the present paper is to contribute to the understanding of the problems of framing (in the Goffmanian sense) educational discourse that students might experience. The context is a science classroom and pupils are working on the issues of climate change. In our analysis of how the participants concretised this rather abstract notion in their work, we found that the various framings that are possible in this case (a political, an historical, an economic, a scientific version etc.) were mixed in the pupils' interpretations of what the issue was all about. Thus, their analytical endeavours were characterised by a mix of voices that frame the issue of climate change in diverse social languages. Learning to see what framing that is relevant in a particular setting, and what arguments it implies and that are relevant, is an important learning task in modern pedagogy. Learning of this kind is very different from the traditional approaches of institutionalised schooling, and the skills students need to master this involve insights both into the different framings that are relevant, and they also need to understand from which position a particular argument or hypothesis is formulated.

Defending and positioning: Studying discursive practices in inquiry mathematics and science classrooms

Ellice Forman, University of Pittsburgh, United States

Ellen Ansell, University of Pittsburgh, United States

Many mathematics and science classrooms through out the industrialized world are undergoing a revolution in instructional practices. One of the most obvious results is a shift of focus from the memorization of facts, practice in algorithms, and use of routine laboratory assignments to collective inquiry activities organized around complex problems. A key component of this instructional revolution is that classrooms dominated by teacher talk and individual seatwork are now busy, noisy places. That change means that educational researchers must have tools available that enable them to assess the effectiveness of classroom talk. The aim of this chapter is to present a conceptual framework for understanding the form and function of discourse in inquiry mathematics and science classrooms and a set of methodological tools for studying these talk-rich classrooms. Our primary focus will be on the classroom as a site for the enculturation of mathematical and scientific argumenta-

– 38 –

tion and identity. The paper will be organized around three conceptual and methodological themes: investigating how classroom norms for scientific argumentation are co-constructed; studying how teacher and students co-create scientific arguments; and assessing the creation of mathematical and scientific identities through discourse. We will begin by providing a justification for the study of argumentation and identity in K-12 classrooms. We will also include a discussion of the differences between mathematical and scientific arguments. Next, we will present some promising methods for studying argumentation and identity formation in inquiry classrooms. Finally, we will evaluate the limitations of current theories and methods and suggest new directions for this line of research.

A 4 23 August 2005 14:30 - 16:30 Room E010

Symposium

Teacher Education

VIDEO BASED RESEARCH TO PROMOTE TEACHER EXPERTISE

Chair: Helmut Fischler, Free University of Berlin, Germany

Organiser: Helmut Fischler, Free University of Berlin, Germany

Discussant: Steeve Wheeler, University of Plymouth, United Kingdom

Teachers are one of the most influential context factors in students' learning environments. Based on this, the conference theme gives rise to several important questions concerning teachers' learning processes, one of which will be addressed in this symposium: How does the use of video enhance teacher learning and achieve that best qualified teachers work in our schools? What better way for novices or experienced teachers to improve their teaching than to analyse videotapes of themselves or colleagues in action? Brophy (2004) points at the potential video techniques have for pre-service or in-service teacher education. In various teacher training projects videos are a highly appreciated tool to give (prospective) teachers the chance to integrate vivid examples into their studies about teaching and learning. But what do we really know about the outcomes of these programs? In the symposium, a second powerful field of application is described: Videos are used in the course of research activities that observe and promote processes and results within programs aiming at enhancing teachers' expertise. Four examples of using videos as a means for controlling effects of efforts to promote teachers' expertise will be presented.

These effects are observed and analysed with the help of videotaped lessons cover-

– 39 –

ing a longer period of time in which changes of teachers' teaching competencies are expected. The research examples presented include general aspects of teaching like adaptive teaching as well as subject specific features like activating students in science lessons.

Video test to evaluate the development of ‘adaptive teaching competency'

Christian Bruehwiler, School of Teacher Education, St. Gallen, Switzerland

Matthias Baer, School of Teacher Education, St. Gallen, Switzerland

Marion Rogalla, School of Teacher Education, St. Gallen, Switzerland

In this paper findings from our research project on 'Adaptive Teacher Compentency' involving 50 teachers at primary and secondary schools are presented. Our research is aimed at capturing the interplay between teacher knowledge and teacher action in enabling understanding-centred as well as individually oriented learning of each student in a classroom. The concept of 'Adaptive Teacher Competency' is proposed to allow an understanding of the conditions of planning and acting in instruction.

The project included an intervention aimed at developing teachers' adaptive competency which was evaluated in a quasi-experimental design. To measure the effects of the intervention a variety of instruments were developed employing quantitative as well as qualitative methods. However, the focus of the presentation is on the video test capturing adaptive teaching competency, since it is believed that this instrument could be crucial for measuring the effects of teacher education. The video test enables a measurement of participants adaptivity of instruction performance which is close to the teaching situation but also standardised to support comparison.

In this way the video test is a new instrument to gain a better understanding of the various concepts of instruction teachers deploy and how the teachers implement these. The methodology of the video test will be discussed as well as the findings of its statistical analyses. Comparing experimental and control group within the four dimensions of Adaptive Teaching Competency (Subject Knowledge, Diagnosis of

Students Learning, Methods of Instruction, and Classroom Management), significant effect size was found in the dimension of Methods of Instruction: the competency of the teachers in the experimental group increases. Conclusions on the future development of teacher education are drawn.

– 40 –

Preconditions for learning in a school for all: development of a video-based study group method for junior high school teachers

June Junge, University of Stavanger, Norway

Elaine Munthe, University of Stavanger, Norway

This paper addresses the pilot phase of the Preconditions for Learning in a School for

All-project (2004-2005). Here, a learning method is being developed in close collaboration with a group of teachers at one junior high school. The method involves teachers and their students being video filmed in the classroom, and then analysing and discussing this video in a group of teachers who all teach the same students.

An aid in these discussions is a manual developed by researchers at the University of Virginia, USA (CLASS). This manual was developed to analyze video films of classroom interaction in relation to variables that research has shown as effective in enhancing academic and social growth. Objectives of this pilot study phase is to test this manual as a learning tool in relation to video, and to develop this material further, making it relevant for Norwegian junior high schools teachers. The discussions carried out in the group are also video filmed, and are analysed and discussed by a group of researchers (including the researcher who leads the discussion in the group of teachers) at the CBR to study and further advance the communication that goes on in the group as well as to further develop the CLASS manual for junior high school teachers. Outcomes presented in this paper will focus mainly on how the use of video contributes to learning among teachers. Sources of information to discuss this will be the videotaped discussions, the teachers' own perceptions as they are expressed verbally, and analyses of classroom videos carried out throughout one school year to measure change.

Using video data in subject related pedagogical coaching processes

Helmut Fischler, Free University of Berlin, Germany

After a period of about 20 years with research on students' ideas about concepts in natural sciences (e.g. force, energy, atom) and with investigations of learning processes in science lessons it becomes more and more important to reflect on how the findings can be transferred into daily teaching processes. That means: How can teachers be helped to include principles of effective teaching derived from these findings into their scope of action? To identify the supporting and restricting factors in the processes of enhancing teachers expertise is the main goal of a project in which subject related coaching is used as a means to modify teachers' conceptions about teaching and learning as well as their acting in classrooms, in short: to foster

– 41 –

teachers professional development. The theoretical background is shaped by theories of cognitive and emotive therapy. These theories focus on the transformation of an individual's problem specific frame of references and of his/her corresponding behavior. Video taped teaching episodes or lessons meet different demands appearing in the coaching process:- Observing their own behavior in classroom processes and teacher-students interactions help teachers to perceive their emotive and cognitive state. In coaching processes, videotaped episodes serve as stimuli for reflections. - In coaching processes it is necessary that the coach has gained insight not only into a teacher's conceptions about teaching and learning but also into characteristics of his/her teaching behaviors. Videotaped episodes facilitate the analysis of teachers' teaching practices. - The effects of coaching processes have to be controlled on the level of teachers' conceptions as well as on the level of their decision making in classrooms. The various functions of videotaped teaching episodes in coaching processes will be described in detail.

Professional development of physics teachers and in-service teacher training: Video-based studies

Hans Fischer, University of Duisburg-Essen, Germany

During the last ten years techniques have been developed to investigate the effects of teacher further education on teachers' activities and on students' behaviour in physics classrooms. Video analysis based on theoretical models of teaching and learning physics, and several methods of investigating performance and motivation as well as controlling different variables of the teaching process have been worked out. As a result, a set of research tools came into being, which facilitates analysing and enhancing teacher training and its effects in more detail. The triangulation of the different data capture allows a sophisticated and divers interpretation of the data describing learning and teaching in the physics classroom and tracking the effects of teachers' professional development. In addition, specific category sets and basic coding of videos are used as interactive methods of teacher training, known as ìvideo feedback methodology. The framework of the project will be outlined.

It is based upon an analysis of theories and models for (physics) teachers' professional development and the theory of basic models of teaching and learning by

Oser, modified for physics instruction. This frame of reference is used to develop the category system of the video analysis as well as for constructing questionnaires and interviews (using the repertory grid method by Kelly). Development, use and results of a low and high inferent video analysis, as applied for the analysis of the teachers' and students' activities in the physics classroom will be explained in de-

– 42 –

tail. Since field tests of the different tools and the teacher training will be finished in summer 2005 we will report also about the reliability of the different tools and about structure and implementation of the teacher training seminar stressing different facets of the used video analysis.

A 5 23 August 2005 14:30 - 16:30 Room E005

Symposium

Mathematics Education

THE MATHEMATICS EDUCATION TRADITIONS OF EUROPE (METE)

PROJECT: PRINCIPLES AND OUTCOMES

Chair: Paul Andrews, University of Cambridge, United Kingdom

Organiser: Paul Andrews, University of Cambridge, United Kingdom

Discussant: Peter Op 't Eynde, University of Leuven, Belgium

Kurt Reusser, University of Zurich, Switzerland

The mathematics education traditions of Europe (METE) project is a European Union funded study of the teaching of mathematics in the age range 10-14 in Flemish

Belgium, England, Finland, Hungary and Spain. The main data set has been video recordings of sequences of lessons taught on topics common to all curricula and representative of different forms of mathematical thinking.This symposium outlines the processes by which the project coding schedule was developed and, in respect of the topics covered, presents the analyses of both qualitative and quantitative data in respect of the teaching of percentages and polygons in primary classrooms and linear equations and polygons in secondary classrooms.The coding schedule, the outcomes of a grounded process, offers a unique perspective in comparative education while the topic analyses offer alternative perspectives on standard content for mathematics educators irrespective of national location.

The mathematics education traditions of europe (mete) project: methodological perspectives and instrument development

Paul Andrews, University of Cambridge, United Kingdom

This paper outlines the processes by which a multinational team of researchers developed a descriptive schedule for use with video-recordings of mathematics lessons in five European countries: Flemish Belgium, England, Finland, Hungary and

– 43 –

Spain. Located within a constructivist perspective on learning as the negotiation of meaning, we discuss the resolution of initial problems concerning false assumptions about shared understanding of educational terminology. We discuss, also, drawing on grounded theory, an iterative process of classroom observation and schedule development which was undertaken in all countries by teams of researchers from each country. The schedule thus developed enabled all project colleagues, irrespective of nationality, to code mathematics lessons with confidence and understanding while highlighting both similarities and differences between the lessons taught by teachers in project countries. The coding schedule comprised three sections plus a timeline.

The first section, named the mathematical focus, comprised seven codes related to observable general learning objectives of mathematics. These were mathematics as conceptual, structural, derivational, procedural, efficiency, problem solving and reasoning. The second section, called the mathematical context, comprised four codes related to the conception of mathematics underpinning the tasks on which learners were invited to work. Each was related to some sense of a real world and the integrity of the data used. They were real-world genuine data, real-world fabricated data, not real-world genuine data and not real-world fabricated data. The third section, called the mathematical didactics, comprised ten categories concerned with the observable didactic strategies project teachers employed. These were activating prior knowledge, exercising prior knowledge, explaining, sharing, exploring, coaching, assessing or evaluating, motivating, questioning and differentiating. These three sets of categories, when applied to the episodes of a lesson, allowed for the diverse nature of project lessons to be exposed and critically examined. Some implications for comparative research in general are discussed.

Mathematics education traditions of europe (METE) project: A comparative study of the teaching of linear equations in five european countries

Paul Andrews, University of Cambridge, United Kingdom

Judy Sayers, University College Northampton, United Kingdom

This paper reports on one aspect of the work of the Mathematics Education Traditions of Europe (METE) project. Drawing on a socio-cultural perspective on teaching and learning and a grounded theory approach to research, the project has explored the teaching of mathematics to learners in the age range 10-14 in Flemish

Belgium, England, Finland, Hungary and Spain. Its main objective was to examine the ways in which learning is structured over time by examining videotaped sequences of lessons taught on standard topics representative of all project curricula and forms of mathematical thinking. This paper reports on the teaching of linear

– 44 –

equations and discusses both quantitative analyses, derived from codes applied to the videotaped lessons, and qualitative analyses focused on the ways in which teachers structure the topic and the forms of activities used to develop it. The development of the coding schedule for videotaped lessons, undertaken the previous year of the project, involved a week of live observations in each project country. Each lesson was observed by at least one colleague from each participating country and subsequent negotiative processes enabled, in a grounded manner, the development of a coding schedule which all colleagues understood and were able to implement.

The outcomes of the quantitative analysis indicated several significant differences, determined by nationality, in respect of the observable learning objectives, the conception of mathematics in which learners' tasks were embedded, and the didactic strategies employed. The qualitative analysis indicated, for example, the balance metaphor for equations was introduced, in varying degrees of explicitness, in four of the five countries while equations with the unknown on both sides were introduced early in three of the five countries to justify the learning of the procedures that followed. The outcomes are discussed in relation to the framework developed earlier and the implications considered.

Teaching percentages in the primary school: A five country comparative study

Fien Depaepe, University of Leuven, Belgium

Erik De Corte, University of Leuven, Belgium

Lieven Verschaffel, University of Leuven, Belgium

Peter Op 't Eynde, University of Leuven, Belgium

The paper deals with a small-scale videobased comparative study of the teaching of percentages at the primary school level in five European countries: England,

Belgium-Flanders, Hungary, Spain, and Finland. In each country a sequence of four or five consecutive percentage lessons was videotaped in a fifth grade mathematics class. All lessons were analysed quantitatively and qualitatively. The quantitative analyses covered the following categories: mathematical focus (referring to the underlying objectives of teacher's actions and decision-making), mathematical context

(referring to the conception of mathematics underlying the tasks posed in a lesson), didactics (referring to the didactic strategies teachers use in different contexts), and the concrete materials used by the teacher and the students to support the teaching-learning process. A more qualitative comparison of the different approaches of teaching percentages focused on the different pedagogic and social activities in which students were involved. All lessons were framed within a perspective on the teaching and learning of percentages that was based on findings from the recent

– 45 –

literature, and related to the objectives of percentage instruction, conceptual aspects of percentages, and some didactic tools. The aim of the study was to identify distinctive features of the percentage education traditions in the five countries. It did not aim at the evaluation and generalisation of the different educational practices, but attempted to understand some good mathematical practices within their specific educational context. The paper discusses the outcomes of those analyses. It presents the frequencies and patterns of the subcategories of the different dimensions of our analyses, and describes differences as well as similarities of the percentage lessons with regard to the perspective on teaching and learning percentages.

The mathematics education traditions of Europe (METE) project

Jose Carrillo, University of Huelva, Spain

Nuria Climent, University of Huelva, Spain

This paper reports on an analysis of sequences of lessons on polygons taught to primary-aged learners by four teachers in Flanders (Belgium), England, Hungary and Spain. Our objective, as part of the mathematics education traditions of Europe

(METE) project, was to compare how these four teachers, considered to be good practitioners by their local communities, approach the teaching of the topic. To achieve this objective we adopted a double methodological perspective. Firstly, each lesson is analysed in terms of the variables defined in the theoretical/methodological paper (Andrews, 2005) and the instruments described therein. Secondly, a qualitative analysis identified teachers' underlying perspectives on the topic and the activities they used to attain them. Importantly, the two analyses described above allowed us to identify both similar and different features of these teachers' practice and, in so doing, enabled us to infer some nationally-informed characteristic features in respect of the teaching of polygons.Our work draws on a number of perspectives on the teaching and learning of polygons derived from literature, both in terms of the mathematics itself and the didactics as they impact on the topic. Our analysis acknowledges the context in which each teacher works and then presents both an individual and comparative account of the ways in which teachers in project countries manage the teaching of polygons. The implications of the analyses are discussed within the literature-based framework derived earlier.

– 46 –

Teaching polygons in the secondary school: a comparative study in five European countries

Eva Szeredi, Eotvos Lorand University Budapest, Hungary

Judit Torok, Eotvos Lorand University Budapest, Hungary

The paper deals in details with one aspect of the mathematics education traditions of Europe (METE) project concerning the teaching and learning polygons in grades

7 or 8. The general theoretical framework, the methodology and the design of the research are discussed in details in the introductory paper. In this particular paper we present a particular theoretical framework for the teaching of geometry in order to provide a reference for our analysis, to facilitate the interpretation of our results and the posing of new questions. We present an analysis of both qualitative and quantitative data derived from 18 lessons, in four sequences, taught on secondary polygons and videotaped in four European countries. The quantitative analysis is based on codes applied according to a schedule developed earlier in the project and described in full in the first paper of this symposium. Mean scores, calculated for each code across all lessons allowed for the development of a composite European lesson against which others can be compared. Moreover, composite lessons for each country were also constructed to facilitate comparison. Statistical analyses indicated substantial nationally-located variation while correlations identified interrelationships between variables. The qualitative analysis showed interesting variations in the ways in which teachers conceptualise and present polygons to learners.

For example, some teachers emphasised classification and vocabulary while others focused attention on definitions and theorems. We relate our findings to the existing literature and, in so doing, consider the implications for the ways in which geometrical learning might be better managed according to locally defined curricular expectations.

– 47 –

A 6 23 August 2005 14:30 - 16:30 Room E102

Symposium

Peer Interaction

QUALITY AND CHARACTERISTICS OF PEER COLLABORATION

Chair: Sylvia Rojas-Drummond, National Autonomous Univerity of

Mexico, Mexico

Organiser: Ed Elbers, Utrecht University, Netherlands

Sylvia Rojas-Drummond, National Autonomous Univerity of

Mexico, Mexico

Discussant: Karen Littleton, Open University, United Kingdom

The papers in this symposium share an interest not only in the quality but also in the conditions and characteristics of peer collaboration in the classroom. All papers are based on observations of collaboration of primary school children and apply in-depth analyses of verbal interaction. The studies in this symposium address two related issues:

1. The relationship between global and micro-analytic measures for the characterization of the collaboration as well as for the assessment of its quality.

2. The extent to which the quality of collaboration should be assessed in relationship to the conditions of the collaboration, such as the goal, the educational programme of which it is a part, and the nature of the task.

The papers show that global measures (such as Mercer's concept of exploratory talk) are insufficient for characterizing peer collaboration in some circumstances.

The description of the quality of collaboration in terms of explicit and accountable reasoning does often not correspond with the reality of actual interactions. Students' attempts to mutually coordinate and integrate contributions go hand in hand with interruptions, overlaps and conflicts (Vass, Haan/Elbers). An increase of students' communicative acts does not always imply an improvement of exploratory talk (Fernandez). The papers share the emphasis on the conditions of peer collaboration. Three contributions (by Vass, Fernandez, and Rojas-Drummond) point to the importance of the task (creative writing compared to a logical reasoning task), whereas De Haan and Elbers draw attention to the role of language skills and the cultural norms for reasoning in a multi-ethnic classroom. Rojas-Drummond et al. argue that the creation of learning communities provides an appropriate context for

Children's collaboration.

– 48 –

The co-construction of oral and written texts in children’s collaborative learning using ICT

Sylvia Rojas-Drummond, National Autonomous Univerity of Mexico, Mexico

Daniel Albarran, National Autonomous Univerity of Mexico, Mexico

Guadalupe Vega, National Autonomous Univerity of Mexico, Mexico

Mariana Zuniga, National Autonomous Univerity of Mexico, Mexico

Maricela Velez, National Pedagogical University, Mexico

The purpose of this paper is to describe how children co-construct knowledge as they work in a project which involves discussing together to create a written text using ICT. The study is based on a sociocultural perspective. To achieve this account, the paper is divided into two parts. In a first part, we work at a micro level by analysing in detail the quality of the interaction and discourse taking place in these collaborative groups of children as they work in their project. In a second part of the paper, we move to a macro level by describing the broader context in which this collaborative groups work, which corresponds to a learning community.

Concerning the first, micro level, the analysis is based on examples of dialogues produced by 4th grade children (9 to 10 years old) as they plan, share ideas, argue their points of view, negotiate, elaborate and integrate their different perspectives to create a multimedia story. With these examples we illustrate how children participate and coordinate efforts in the co-construction of knowledge. In relation to the second, macro level, we describe the broader sociocultural context in which children develop their projects. Children participate in an educational programme which has been implemented in public primary schools in Mexico called ìLearning

Together. The purpose of the programme is to form learning communities with the active participation of primary students, teachers, as well as university researchers.

This programme seeks to develop functional social, cognitive, psycholinguistic and technological abilities in primary students. In conclusion, in our paper we illustrate how the creation of learning communities provides appropriate contexts for Children's joining efforts in the co-construction of knowledge.

– 49 –

Oracy for literacy: Exploratory talk and co-construction in chldren’s collaboration

Juan Manuel Fernandez Cardenas, Comite Norte de Cooperacion con la UNESCO,

Mexico

Sylvia Rojas-Drummond, National Autonomous Univerity of Mexico, Mexico

Nancy Mazon, National Autonomous Univerity of Mexico, Mexico

Rupert Wegerif, University of Southampton, United Kingdom

The present report has two main purposes. The first is to analyse the quality of the strategies children used for discussing in small groups to solve two tasks: a) a modified version of the Raven's Standard Test of Progressive Matrices (RSPM), and b) a test where children discussed and wrote an integrated summary of three related texts. The second purpose is to test two approaches to the analysis of discourse: a) a scheme for analysing exploratory talk (ET) and b) the ethnography of communication. We report two related studies with 6th grade primary school Mexican children (11 to 12 y.o.).They solved the two tasks mentioned as pre- and post tests before and after they were trained in the use of Exploratory Talk (ET) as well as in strategies for producing summaries. We found that, after training, children improved substantially in the use of ET to think together and argue when solving the

RSPM test but not the summary task. However, when we used the ethnography of communication to analyse their talk for the second task, we found that the number and quality of communicative events and acts increased importantly. Also, the quality of the summaries children produced improved significantly. In conclusion, the overall discursive analyses carried out for both tasks revealed that, in spite of some differences, there were also many commonalities in Children's talk. These included: asking for and providing opinions, generating alternatives, elaborating on the information shared, coordinating efforts and seeking agreement. To account for these similarities, we propose to adopt the concept of ìco-construction as an inclusive term to characterise the efforts of collaboration and coordination of activities used by the children in a variety of educational contexts.

The discourse of collaborative creative writing. Peer collaboration as a context for mutual inspiration

Eva Vass, University of Southampton, United Kingdom

Drawing on socio-cultural theory, this paper focuses on Children's classroom-based collaborative creative writing. The central aim of the reported research was to contribute to our understanding of young Children's creativity, and describe ways in

– 50 –

which peer collaboration can resource, stimulate and enhance their creative writing activities. The study drew on longitudinal observations of ongoing classroom activities in Year 3 and Year 4 classrooms (children aged 7-9) in England. Selected pairs' collaborative creative writing activities were observed and recorded using video and audio equipment in the literacy classroom and in the ICT suite (13 pairs, about

2-4 occasions each). The research built on the contextualised, qualitative analysis of the social and cognitive processes linked to shared creative text composition via the in-depth study of verbal interaction. For the analysis a functional model was developed. By linking cognitive processes associated with writing (engagement and reflection) to observed collaborative and discursive features, the research has identified discourse patterns and collaborative strategies which facilitate sharedness and thus support joint creative writing activities. A central finding was the centrality of emotions in the observed creative writing activities. Also, the analysis revealed Children's reliance on collaborative floor (Coates, 1996) a discourse building on interruptions and overlaps, often without clear-cut turn-taking. Challenging traditional descriptions of effective collaboration, the use of collaborative floor was found to support the joint generation of creative content. By enabling the partners to mutually inspire each other, it facilitated sharedness. These findings have implications to both educational research and practice, contributing to our understanding of how peer interaction can be used to resource joint creative activities.

Transactivity in peer collaboration in a multi-ethnic primary school

Mariette De Haan, Utrecht University, Netherlands

Ed Elbers, Utrecht University, Netherlands

In this paper we present an analysis of peer collaboration in a multi-ethnic school in the Netherlands. The school encourages collaboration and students are accustomed to work in small groups of four or five students. Our observations focus on students in a 7th grade (children between 11 and 13 years of age) during mathematics lessons. We recorded the conversations between the children during the collaborative activities with audio recorders and transcribed them. The analysis was based on

20 hours of transcribed recordings. Studies of collaboration typically use one or more predefined concepts for assessing the quality of collaboration. In this study, rather than evaluating the collaborative activities with pre-defined norms, we use the analysis of the actual talk for figuring out how students structure their collaboration from moment to moment. We distinguished transactive episodes (TE's), which we define as episodes where minimally one student attempts to create a common problem solving situation that is answered by minimally one other student. For

– 51 –

each TE, we established which activity students were undertaking or what part of the problem solving process students were addressing. These TE's could, for instance, encompass the exchange of information about the task, the definition of the mathematical procedure, or the actual performance of a mathematical procedure.

Analysing the students' conversations in this way, we could understand the contributions of students more easily as sensible forms of participation in the process of collective reasoning. In our interpretation of the students' forms of reasoning we will pay attention to how language skills, cultural norms of reasoning, and particular interpretations of the task contribute to the development of the particular forms of argumentation found in this classroom.

A 7 23 August 2005 14:30 - 16:30 Room A007

Symposium

Student Learning in Higher Education

THRESHOLD CONCEPTS AND VARIATION IN STUDENTS' UNDER-

STANDING

Chair: David Galbraith, Staffordshire University, United Kingdom

Organiser: Jean Mangan, Staffordshire University, United Kingdom

Peter Davies, Staffordshire University, United Kingdom

Discussant: Glynis Cousin, Higher Education Academy, United Kingdom

This symposium is concerned with the development and use of two approaches, threshold concepts and learning study, that may help teachers to understand and address the difficulties that students face in learning troublesome knowledge within a discipline. These approaches share the assumptions that teaching should aim to develop students' deep, transformative, learning and that teachers must possess a rich knowledge of typical variations in students' subject understanding. Whilst the threshold concept approach focuses attention on integrating ways of thinking that subject experts take for granted, learning study focuses on the processes by which teachers can be helped to identify and respond to variation in students' understanding of particular phenomena. The first paper discusses a synthesis of how threshold concepts have been embraced across different disciplines, using a case study approach. It provides perspectives on constructively aligning courses to the experience and understanding of students, conceptualising how students approach, engage in and emerge from potentially transformative and troublesome episodes

– 52 –

of learning and the variation within student learning. The embedding of threshold concepts in the curriculum and assessment practice is explored. The other papers discuss new empirical work, applying the approaches within the discipline of economics. The first considers the testing of learning theory in the classroom using a control group for comparison. Substantial improvement in understanding is reported. The two other papers use the approach of threshold concepts. In the first a wide range of methods for the identification of thresholds within a discipline are discussed and the effectiveness of a number of these is assessed empirically. The second tests the hypothesis that peer group ability in one's study group may impact on the acquisition of a threshold concepts using a research design involving both the randomly-assignment and self-selection of students to study groups.

Overcoming barriers to student learning: Threshold concepts and troublesome knowledge

Jan H.F. Meyer, University of South Australia, Durham University, United Kingdom

Ray Land, Coventry University, United Kingdom

The current interest in threshold concepts' and troublesome knowledge has arisen from work undertaken within an ESRC funded research project (2000-2005) Enhancing Teaching and Learning Environments in Undergraduate Courses. Threshold Concepts' has since been recognised by the Higher Education Funding Council for England (HEFCE) as an area meriting further study in its own right, and from autumn 2004 HEFCE is funding a national FDTL 5 consortium project on Developing first year undergraduates' acquisition of threshold concepts in economics led by Professor Peter Davies at Staffordshire University. The topic of threshold concepts is arousing the interests of academics and educational researchers in a growing number of countries. The work, to our knowledge, is already being cited in Australia, Hong Kong, Sweden, Greece, South Africa and the USA. The basic idea underpinning this field of enquiry is that in certain disciplines there are conceptual gateways' or portals' that lead to previously inaccessible, and initially perhaps troublesome ways of thinking about something. A new way of understanding, interpreting, or viewing something may thus emerge ó a transformed internal view of subject matter, subject landscape, or even world view. In attempting to characterise such conceptual gateways it was suggested in the earlier work that they are transformative (occasioning a significant shift in the perception of a subject), and may also be irreversible (unlikely to be forgotten, or unlearned only through considerable effort), integrative (exposing the previously hidden interrelatedness

– 53 –

of something) and troublesome, for a variety of reasons. This paper draws on the research of both these major projects, as well as the work of an international range of scholars who have become interested in and contributed to this new field of enquiry. It presents a synthesis of how threshold concepts have been embraced across different disciplines.

Recognising threshold concepts: An exploration of different approaches

Peter Davies, Staffordshire University, United Kingdom

Jean Mangan, United Kingdom, United Kingdom

Meyer and Land (2003) have proposed that there exist in many disciplines threshold concepts, which can be considered akin to passing through a portal, or conceptual gateway, thus opening up a new and previously inaccessible way of thinking about something. Such concepts lead to a transformed way of understanding, or viewing something that may represent how people think in a particular discipline, or how they perceive, apprehend or experience particular phenomena within a discipline.

If threshold concepts are to be embedded in the curriculum as a means of improving student understanding, the first step within any discipline is the identification of such concepts. This paper considers how such concepts can be recognised, why particular methods may be appropriate and discusses the strengths and weakness of the various approaches. The methods considered include dialogue with lecturers, analysis of students' writing and oral expression, analysis of mark schemes and comparison of the treatment of a phenomenon by members of different disciplines.

The discussion may be applied to any field, but the discipline of economics is used as the example in our exposition. Data has been collected on a variety of these approaches with different groups of students (both specialist and non-specialists) and staff. The results are presented of their effectiveness in revealing threshold concepts, consistency with other approaches and ease of data collection and analysis.

The paper concludes with recommendations as to the more fruitful approaches to consider in future work.

Peer effects on learning threshold concepts in economics

Martin Shanahan, University of South Australia, Australia

Jennifer Foster, University of South Australia, Australia

Jan H.F. Meyer, University of South Australia, Durham University, United Kingdom

One metaphor for the idea of the threshold concept is that it is a portal, which, when

– 54 –

crossed by the learner, grants access to a previously inaccessible or transformed way of thinking. Within the discipline of economics, the idea of opportunity cost has been identified as a threshold concept (Meyer and Land 2003; Shanahan and

Meyer 2003). In economics, recent attention has been paid to the role of peer ability in education. These explorations have attempted to isolate the effect of average peer ability on overall achievement measures from the potentially confounding factor of self-selection into student groups. Empirical results have been mixed, reflecting in part the gap in understanding regarding the causal process by which peers affect education. One such causal process may be peer influence on the acquisition of threshold concepts. In this study, we test the hypothesis that peer group ability in one's small study group, measured by external measures of peers' educational attainment (university entrance scores and prior economics knowledge) and internal learning processes (measured by an economics-specific reflection on learning inventory), may impact on students' understanding of threshold concepts in introductory economics. We use a unique research design where some students are randomly assigned to study groups. Both randomly-assigned and self-selected students are surveyed at various points in the course to ascertain their understanding of opportunity cost. This design allows us to circumvent student selection in our estimate of the raw peer effect, and also allows us to explore whether students who are able to choose their study group acquire these key threshold concepts more completely than those who are randomly assigned. The results of the study cast new light onto the nature of peer influence in economics education, the factors that influence the acquisition of threshold concepts, and the impact of self-selection in education.

Learning theory as teaching resource: Another example of the radical enhancement of students' understanding of the economic aspects of the world around them

Ming-Fai Pang, University of Hong Kong, China, Peoples Republic of

Ference Marton, Gothenburg University, Sweden

A group of experienced secondary school teachers used a novel learning theory as a resource for planning and carrying out their teaching of a difficult economic concept. Their students' mastery of this concept after a series of three lessons was compared with the mastery of the same concept by students who were taught by another group of teachers under the same conditions except for the use of the theory. The difference in learning outcomes was extreme. Observations of what was happening in the classrooms showed subtle but decisive differences correlated with the differences in outcome. These differences were interpreted in terms of the theory used by the first group, and the results seem to give support to the theory.

– 55 –

A 8 23 August 2005 14:30 - 16:30 Room A108

Symposium

Systems in Education

MULTIPLE PERSPECTIVES ON CLASSROOMS AROUND THE WORLD:

THE LEARNER'S PERSPECTIVE STUDY

Chair: David Clarke, University of Melbourne, Australia

Organiser: David Clarke, University of Melbourne, Australia

Discussant: Jeremy Kilpatrick, University of Georgia - Athens, United States

Since its inception in 1999, the Learner's Perspective Study (LPS) has documented the practices of 'well-taught' eighth-grade mathematics classrooms for sequences of ten or more consecutive lessons using a three-camera approach, supplemented by post-lesson video-stimulated interviews with teacher and students. The move to employ multi-camera and multi-audio in researching educational settings is not primarily technology-driven, but rather motivated by the recent shifts in theories of learning, from a view of learning as transfer to a view of learning as constructed in action. The LPS research community now includes 14 countries and as the body of data has grown, so has the capacity for fine-grained comparative analyses of the practices and associated learning in classrooms as geographically distant as

Uppsala, Melbourne, Tel Aviv, Durban and San Diego, or as close as Hong Kong,

Shanghai and Tokyo. The five presentations in this symposium offer different perspectives on classroom practice: revisiting the teacher-centred, student-centred dichotomy from the perspective of Asian classrooms; utilising complementary theoretical frames to reveal tensions between participation and content-coverage; exploiting the contrasts of international classroom practice to offer novel analyses related to time and flow; utlising particular lesson events as units for comparative analysis of classroom practices and meanings across six cultures; and juxtaposing teacher and student perceptions of those events. The result is a rich, multi-faceted portrayal of the practices of competent teachers in classrooms around the world.

Taken individually, the presentations both confirm and challenge accepted wisdom in relation to valued classroom practice and in several instances problematise elements of the reform agenda being pursued in 'Western' cultures. The result should be sufficiently confronting to stimulate lively discussion.

– 56 –

Teacher-dominating lessons in Shanghai: A triangulation from three perspectives

Ida Ah Chee Mok, University of Hong Kong, Hong Kong

Students learning in teacher-dominating classrooms in Confucian-heritage cultures give very good performances in comparative studies such as TIMSS. These results create a myth regarding the value of the teacher-dominating lessons. In the LPS data from different countries it is not difficult to find episodes with different levels of control or intervention from the teachers. We argue here that what really matters can be perceived from three perspectives: (1) what the teacher wants his students to learn, (2) what the students capture their lessons, and (3) how the teacher's input creates a possibility of understanding the mathematics. The data in this paper come from 15 consecutive (grade 7) mathematics lessons by a teacher in Shanghai, including the teacher's and students' interviews. In this attempt to triangulate the picture of a lesson from three different angles, we see that the teacher's and the students' characteristics are essential elements contributing to the success of the lesson. It is worth noting that like his counterparts in other parts of the world the Shanghai teacher values his students' thinking and participation very much and agrees that knowledge should not be acquired by transmission. However, in the implementation the teacher shows a version of student-centredness different from that in the western world. In spite of the similarity in the belief of what a good learning environment should embrace, the image of the lesson is very much content-oriented and teacher-controlled. Referring to a detailed analysis of the content of a specific lesson via the framework of variation, it is found that the teacher is very skilful in using variation even though he may not be aware of it. The teacher demonstrates a personalized pedagogical theory which he demonstrates successfully in his own lesson.

The price of participation - how interaction affords and constrains the classroom learning of mathematics

Jonas Emanuelsson, Gothenburg University, Sweden

Fritjof Fritjof Sahlstr?m, Uppsala University, Sweden

This presentation aims to demonstrate empirically the interdependency of the content of learning and the organisation of interaction in the mathematics classroom.

Drawing upon two different theoretical positions, we have analysed videotaped interaction in the naturalistic classroom. Materials analysed are collected within the

Learner's Perspective Study, called LPS (Clarke, 2003) and data from several countries are compared and contrasted in the presentation. The general aim of this paper

– 57 –

is to further the understanding of how mathematical content is taught and learned in classrooms, using two different and complementary theoretical perspectives for the analysis. The same empirical material, consisting of whole-class teaching, and seat interaction among peers, from two different mathematics classrooms in the

United States and Sweden, is analysed both within a conversation analysis framework (Sacks, Schegloff & Jefferson, 1974; Schegloff, 1996) and from the point of view of variation theory (Marton & Booth, 1997; Marton et. al., 2004). In the article, the tension between participation and mathematical content is a recurring theme. Results show that extensive possibilities for the students to participate in interaction are associated with a price. The price to be paid is the consistency in the presentation of the mathematical content and hence more limited possibilities for the students to learn mathematics. However, under certain circumstances the "content - cost" is less than under others. We describe and discuss these circumstances.

The unfolding of lessons and their pedagogical rationale as a criterion for international comparisons: The case of teaching systems of linear equations

Michael Fried, University of the Negev, Israel

Miriam Amit, University of the Negev, Israel

International studies such as the TIMSS have taught us, among other things, that international comparisons are devilishly difficult to make (e.g. Keitel & Kilpatrick,

1999). Even where curricular complexities may be put aside and a common subject agreed upon, Stigler and Hiebert (1999) and others have shown that lesson structure and presentation can vary greatly from country to country, culture to culture. The present paper adduces further evidence for this fact and underlines a crucial aspect of the presentation of mathematical subjects to be taken into account in international studies, namely, the manner in which lessons unfold in time. Aspects of time in mathematics teaching and learning have been studied by various scholars such as Chevallard (1985), Brousseau (1999) and Arzarello, Bartolini Bussi, & Robutti

(2002). In this paper, we shall examine how algebra lessons on systems of linear equations flow in time and how their pedagogical rationale unfolds. The lessons, which are the focus of this paper, were observed in an eighth grade classroom in the Negev city of Beer Sheva. These are compared with lessons in Shanghai and

Hong Kong on the same subject matter, as described by Mok, Leung, Lopez-Real, and Marton (Mok et al., 2002). The comparison of our findings with those of Mok, et al. showed that while there were some differences between the lessons in Hong

Kong and Shanghai, those lessons were far more comparable with one another than they were with the lessons we observed in Beer Sheva. Despite the concurrence of

– 58 –

subject matter in all three classrooms, the way the Beer Sheva lessons flowed and developed the material differed so strikingly from either Hong Kong or Shanghai that the very possibility of comparison became no longer obvious. This finding suggests that criteria for international comparison need considerable and critical investigation.

Exploring international classroom practice through a fine-grained analysis of one lesson event in the classrooms of six cultures: Kikan-Shido

David Clarke, University of Melbourne, Australia

Catherine O'Keefe, University of Melbourne, Australia

If we approach social settings (and the situations they frame) as multiply-constructed and open to multiple construal, then the methodology employed in their study must offer a voice to the several participants in these settings. This is particularly true when our interest is the cultural-specificity of practice. In this study, we examined classroom practices in Asian classrooms (Tokyo, Hong Kong and Shanghai) and contrast these with practices in San Diego, Berlin, and Melbourne by applying a layered analytical approach the utilised the 'lesson event' as its basic unit of analysis and identified patterns of participation characteristic of the classrooms studied. Our analysis suggests that there are distinct differences in these patterns of participation even among 'Asian classrooms'. In this presentation, the lesson event 'Kikan-Shido'

(Between desks instruction) is used as the unit of analysis to compare the patterns of participation operating in classrooms in Shanghai, Hong Kong, Melbourne, San Diego, Tokyo and Berlin are identified and compared. In addition, the reconstructive accounts elicited from teachers and students in those classrooms provide insight into the motivations and meanings that underlie classroom participants' collaborative structuring of these patterns of participation. In particular, differences are identified in teachers' enactment of 'Kikan-Shido' and these differences are related to specific pedagogical principles that appear to underlie the teachers' practice. The simplest example of such a difference is the distinction between 'monitoring' and

'guiding' student mathematical work during Kikan-Shido. Our analysis identifies the proportion of time spent on each of these activities as a signature characteristic of each teacher's practice. Another of these key differences relates to the distribution of responsibility for knowledge generation. Our analyses confirm the parallel work of Mok and her colleagues in demonstrating that the teacher-centred/studentcentred dichotomy seriously misrepresents the subtleties of practice in both Asian and Western classrooms.

– 59 –

Perceptions of lesson events and associated values held by the teacher and the students

Yoshinori Shimizu, Tokyo-Gakugei University, Japan

This paper reports the analysis of post-lesson video-stimulated interviews with the teachers and students in three eighth-grade grade mathematics classrooms in Tokyo.

The methodology employed in the Learner's Perspective Study offered the teachers and the students the opportunity in post-lesson video-stimulated interviews to

ìparse the lesson they had just experienced. That is, to identify for the interviewer those events in the lesson that the participant felt to be significant, the teacher and the students were given control of the video replay and asked to identify and comment upon classroom events of personal importance. It is clearly possible that students identify as significant classroom events that are quite different from those intended by the teacher. By juxtaposing their perceptions of classroom events, both within one particular lesson and in the sequence of consecutive lessons, discrepancies and agreements between teacher and the students were identified. Associated values held by the teachers and the students were analyzed in relation to such lesson events as "Kikan-Shido" (instruction between desks) and "Matome" (summing up). The analysis described in this paper suggests that, while they share views on

ìsignificant lesson events, they construct different meanings associated with the same event. Then, the analysis raises the question of the extent to which teacher and learner practices are in a mutually supportive relationship. The results indicate that teacher and learner practices appear to be both conflicting and mutually sustaining.

The analysis also reveals the richness and potential of the collected data, as well as strength of the methodology, in the Learner's Perspective Study.

– 60 –

A 9 23 August 2005 14:30 - 16:30 Room E112

Symposium

Comprehension of Text and Graphics

LEARNING THROUGH CONSTRUCTING EXTERNAL REPRESENTA-

TIONS

Chair: Erica de Vries, University of Grenoble II, France

Shaaron Ainsworth, University of Nottingham, United Kingdom

Organiser: Erica de Vries, University of Grenoble II, France

Shaaron Ainsworth, University of Nottingham, United Kingdom

Discussant: Elsbeth Stern, Max Planck Institute for Human Development,

Germany

Many learning situations offer multiple forms of representations for learners to interpret, but there is increasing interest in asking learners to construct their own external representations. In addition to editing text, learners may construct graphs, build models, draw pictures or write equations. Despite the large amount of research on interpreting representations, much research needs to be done as to the role of construction activities in learning with computer-based representations or with more traditional paper-and-pencil representations. A number of questions should be addressed if we are to understand learning through construction. These include:

What forms of representation are suggested, used and/or invented by learners? How does collaborative construction differ from that of individual construction? What roles does construction play in the processes and products in learning? When is it advantageous to construct a representation versus interpret a given representation?

What support do learners need to construct representations? The papers address a wide range of learning topics which vary in the number of permissible solutions to problems (single correct versus multiple permissible answers) and tools used to support representation construction, i.e. from pen and paper to on-line discussion and simulation environments. Consequently, the main aim of the symposium is to address these questions by bringing together researchers with varied approaches to the study of learners' construction of external representations.

– 61 –

Co-elaborating knowledge with external representations

Marije van Amelsfoort, Utrecht University, Netherlands

Jerry Andriessen, Utrecht University, Netherlands

Gellof Kanselaar, Utrecht University, Netherlands

This study focuses on the collaborative construction of diagrams in argumentationbased learning. In collaborative argumentation-based learning (CABLE), students discuss an issue together in order to better understand it. Diagrams are said to be beneficial for CABLE for many reasons, including forcing students to make opinions and arguments explicit, or keeping focus on the discussion. We are mostly interested in diagrams as tools to structure argumentation and relate argumentative knowledge. We want to know whether the collaborative construction of a diagram supports students in relating knowledge in a space of debate, enabling a better understanding of the issue under discussion. Earlier studies found that students have difficulties in the construction and use of diagrams in CABLE. They focus on the content of the boxes, but not on the relations between the boxes, (or on the overall picture). This is perhaps not a surprising finding. Students are unused to constructing diagrams for argumentation, let alone in collaboration. The construction of an argumentative structure might be too hard for students who are used to narrative structures. There might also be differences in individual preferences for verbal or visual learning. In this study, pairs of students discuss the issue of Genetic Modification. In condition one, students construct a diagram in which focus is placed on relations by labeling them. In condition two, students construct a diagram in which focus is placed on arguments by labeling them. In the third condition, students can only use chat. We investigate the amount and nature of relations students

(co-)construct. We expect understanding of the space of debate to be highest in the relation condition, and lowest in the chat-only condition. Differences in preference for verbal or visual learning will be taken into account. Results of the study will be discussed in Cyprus.

Dynamic modeling, the added value of simulating a representation

Wouter van Joolingen, University of Twente, Netherlands

Simone Lohner, University of Amsterdam, Netherlands

Elwin Savelsbergh, Utrecht University, Netherlands

The central question in this presentation is to what extent simulating the dynamic properties of dynamic models improves inquiry and understanding of the domain that is being modeled. In dynamic modeling tasks, students build a representation

– 62 –

of the domain using a system-dynamics representation as found in tools such as

STELLA. In the study presented, 43 students created a dynamic model of the earth's greenhouse effect, one group could simulate the model constructed; a second group could not simulate the model. In pre- and post tests, learner's knowledge and reasoning was assessed with items in which learners had to expose reasoning ability by predicting the change in the model as result of a particular event, such as removing the earth's atmosphere. A coding scheme for reasoning ability with dynamic models is introduced, based on dimensions of knowledge type and knowledge subject. It is expected that students in the dynamic modeling group are better in generating graphs. With respect to the reasoning with the model, one expects students using the conceptual model to generate more elaborate verbal explanations of their thoughts, as they had the need to reason explicitly with their model. On the conference, the results of analysis, which is currently in progress, will be presented.

Learning by constructing self-explanation diagrams

Shaaron Ainsworth, University of Nottingham, United Kingdom

Ioanna Iacovides, University of Nottingham, United Kingdom

The self-explanation effect (whereby students generate explanations to themselves as they are studying) has been shown to enhance learning in many domains. Recent research has demonstrated that the way that material is presented influences the self-explanation effect. Ainsworth & Loizou (2003) presented students with information about the circulatory system in either text or diagrams and prompted them to self-explain. Diagrams students outperformed text students at post-test, generated more self-explanations and their learning was more dependent on self-explaining.

The current study sought to explore if these same benefits would ensue if students constructed self-explanations in diagrammatic form. Consequently, twenty-four subjects were given information about the human circulatory system to learn. Half of them were given the information in the form of diagrams and asked to write down their self-explanations. The other half were given the information in the form of text and asked to construct their own self-explanation diagrams as they self-explained. The results showed that students in both conditions learnt and at post-test performed identically on every measure of learning. They also generated the same number and quality of explanations. The only ways these two groups differed is in the amount of information they chose to translate across representations. Text students included almost twice as much information in their pictures as diagram students in their summaries. Furthermore, the amount of information translated predicted learning outcomes whereas the number of self-explanations did not. Overall,

– 63 –

these results showed that by generating their own diagrammatic self-explanations while studying, students can overcome the previously reported text disadvantage. It also suggests that some of the some of the benefits of self-explanation may be due to translating information over representations of different forms.

Analysing design problem solving products as external representations

Erica de Vries, University of Grenoble II, France

This contribution examines external representations as problem solutions in design tasks for technology education. Design tasks are introduced to put students in learning situations that resemble professional situations and to make students use the physical and semiotic tools of the trade. Design tasks involve the construction of a plan of a future artefact, and one of its defining characteristics is the fact that multiple solutions exists which can be rated on their feasibility and technical quality, cf. rating technological aspects of solutions such as manufacturing, use, and maintenance of the future artefact. Design solutions are not simple drawings; they can be characterized as composite representations that incorporate geometrical and functional information using different semiotic codes. The current study investigated drawings and textual comments produced by students from different backgrounds for a student residence problem in order to give answers to questions about the strategies and representational formats used. Two groups of students participated in the study: students from a secondary school for vocational training and fourth year university students in educational science. The analysis scheme takes into account information in graphical and textual form and a number of different representational aspects. Results are presented and interpreted in terms of both affordance of the situation, cf. aspects inducing students' choice of a particular representational format, and vicariance, cf. the interchangeability of cognitive processes for determining action in a given situation.

– 64 –

A 10 23 August 2005 14:30 - 16:30 Room A010

Symposium

Motivational, Social and Affective Processes

EXPANDING MOTIVATION AND SELF-REGULATION IN MULTIPLE

SOCIAL LEARNING AND TEACHING CONTEXTS

Chair: Hanna Jarvenoja, University of Oulu, Finland

Sanna Jarvela, University of Oulu, Finland

Simone Volet, Murdoch University, Australia

Anastasia Efklides, University of Thessaloniki, Greece

Susan Bobbitt Nolen, University of Washington, United States

Marja Vauras, University of Turku, Finland

Erkki Olkinuora, University of Turku, Finland

Jarkko Makinen, University of Turku, Finland

Pekka Salonen, University of Turku, Finland

Organiser: Marja Vauras, University of Turku, Finland

Simone Volet, Murdoch University, Australia

The session focuses on challenges for research on motivation, emotion, and selfregulation in complex learning and teaching environments typical in today's knowledge- and competence-oriented society. Formal and in-work learning environments in all educational levels are, increasingly, greatly variable, rapidly changing, and less structured, mixed-motive situations. Productive engagement in learning activities as well as professional development has to be understood in relation to the real contexts in which it is embedded, since context is assumed to give meaning to taskoriented actions. Theoretical perspectives to teaching-learning processes taking place in various contexts, captured in terms like contextualized, socially mediated, co-regulated, have more recently extended to research on motivation, emotion and self-regulation. Researching situated motivation, emotion and regulation implies locating goals, engagement and affect in the dynamic activities of social systems.

Social interaction with dynamic, relational regulation and negotiation patterns become accentuated and, thus, shift research towards multi-dimensional aspects of learning and development. However, despite some rather sophisticated attempts to describe socio-cognitive interaction dynamics, there is still lack of concepts and research on macro-level social structures (e.g., participation or relational control structures) shaping interaction processes and subsequent developmental pathways.

– 65 –

In five presentations, we discuss challenges for motivation, emotion and self-regulation research and ideas for a more expanded, holistic framework for the analysis of these constructs. The focus is on dynamic, developmental and situational interactions between motivation, affect, self-regulation, personal growth and learning processes. Some new concepts, better depicting the social nature of these processes, will be proposed. Empirical observations and data clarifying the arguments will be presented, although the focus is on the state-of-the-art literature or meta- analysis.

Also, we draw attention to efficient methodological tools to examine and assess dynamic motivational, affective and regulatory processes, and of finding models for good practice and instructional design.

Emotional and motivation control in collaborative learning situations - how students are able to regulate emotions evoked from these challenges?

Hanna Jarvenoja, University of Oulu, Finland

Sanna Jarvela, University of Oulu, Finland

The aim of the study is to investigate the interplay of the regulation processes between an individual and a group, as well as dynamics between emotion and motivation control. The social learning situations where individuals' characteristics, goals and demands meet can arouse emotions and create motivational conflicts.

For example, the idea of collaborative learning models is that the students aim for construction and attainment of a shared goal so that the responsibility of the learning process is equally shared. This requires constant negotiation, argumentation and sharing ideas among the group members and during these processes social conflicts can emerge. Socially challenging situations are critical for successful collaboration and they therefore invite the need to control both individuals' own motivation and emotions and the other group members' emotional arousal and motivation.

This paper aims to specify what kind of emotional challenges social aspects of collaborative learning bring to the situation and as well as students self-, other- and shared-regulation in these situations. The data consists of educational psychology students' (n=35) self-reports collected with a specific ìdynamic questionnaire. The questionnaire is designed to reach the dynamic nature of collaborative learning. In the analysis the experienced social challenges are linked to the reported activity of the students' use of regulation strategies. Also student groups' internal dynamics are analysed by creating groups' goal and regulation activity profiles. The results indicate that certain types of social challenges are linked to more active use of regulation strategies than the others. The data of the group dynamics gave multi-faceted view on how individual students' goals and interpretations of the social challenges

– 66 –

are linked to the use of regulation strategies. It also showed how the students interpret the group's role for their achievement.

Self-regulation and self-in-context regulation: pathways to knowledge construction in social learning environments

Simone Volet, Murdoch University, Australia

Marja Vauras, University of Turku, Finland

Pekka Salonen, University of Turku, Finland

This paper discusses an expanded, holistic framework for the analysis of regulatory constructs (self-, other-, co- and shared-regulation). The framework aims at clarifying the diversity of regulatory models through screening the individually- and socially-oriented regulatory constructs, and relating them to the basic theoretical assumptions underlying the socio-cognitive and socio-cultural approaches. A social shared regulation space is conceptualized, where knowledge is socially constructed through participation in activities that vary in locus of regulation. A range of established and hypothetical paths link these forms of shared regulation with self-regulation. The developmental paths have a sound theoretical grounding supported by extensive empirical work, and some have also been turned into instructional paths as they have inspired the development of instructional strategies, e.g., reciprocal teaching, joint problem-solving, computer conferencing and various forms of collaborative learning. Some hypothetical paths have been added, leading to a tentative full model of regulation in learning. This model questions whether self-regulation should be considered as the single ultimate form of regulation. In light of the second wave of cognitive theory, which has emphasized the necessity to focus on the

"individual-in-context", we propose a new concept, "self-in-context regulation", in order to explain regulation in social learning context. SCR integrates self-regulation and shared regulation in a single process. The concept of SCR will be used to explain how the forms of shared regulation lead to knowledge construction in real learning situations, and how it may work best as an instructional tool. Illustrations from real-life learning at micro-level will be presented.

– 67 –

Self- and co-regulation in instructional context: restraining and facilitative interaction patterns

Pekka Salonen, University of Turku, Finland

Anastasia Efklides, University of Thessaloniki, Greece

Marja Vauras, University of Turku, Finland

This paper attempts to link learning-related cognitive-metacognitive, motivational and affective regulatory processes to patterns of interpersonal transaction. Most models of socially guided learning presuppose a gradual progress from other- to self-regulation, often described through the scaffolding metaphor. In optimal scaffolding, the teacher, e.g., sensitively regulates task difficulty, assists the student to articulate the essential features, and contingently doses and fades help-giving. Such scaffolding acts are often discussed in a somewhat idealized manner. Based on the findings from interpersonal and relational control theories (pertaining e.g. to therapy and familial transactions), we propose that instructional interaction is governed by inter-behavioral macro-structures, such as recursive patterns of relational control and affect coordination These patterns may importantly co-determine the ways the teacher and student respond to each other, and the ways the teacher modulates the student's cognitive, metacognitive, motivational, and affective activity. For example, complementary relational patterns may be 'stabilized' and, particularly under stress, become escalatory with increased rigid domineering-submissive behaviors and negativity. Although there is some empirical evidence that recurrent interpersonal control and affective patterns contribute to instructional processes, there are no studies focusing on the interplay of interpersonal relational control, affectivity, and teacher's regulatory activity. Our pilot studies suggest that recurrent relational control and affective patterns may essentially co-determine the successfulness of teacher's cognitive, metacognitive and motivational regulatory activity during scaffolding processes. Thus, it is important to understand how interpersonal motivational, affective and relational control mechanisms and socio-cognitive processes interact and contribute to the development of a learner's regulatory processes.

– 68 –

When worlds collide: negotiating competing views of teaching across social contexts and the effect on student teachers’ motivation to learn

Susan Bobbitt Nolen, University of Washington, United States

Ilana Seidel Horn, University of Washington, United States

Christopher J. Ward, University of Washington, United States

Reed R. Stevens, University of Washington, United States

Katherine Estacio, University of Washington, United States

Researchers have recently examined the role of multiple contexts in students' developing interest and motivation to learn. In the US and other countries, individuals who wish to be teachers learn by participating in multiple contexts: both formal education in colleges or universities and apprenticeships as student teachers or interns in school settings. Taking a socio-cultural perspective, these contexts could be studied as places of identity development rather than as contexts for learning. Holland, Lachicotte, Skinner & Cain (1998) claim that motivation and will are formed as the individual develops identities in the socially-constructed roles, meaning-systems, and symbols of the cultural contexts in which they live, so-called figured worlds'. In this perspective, motivation to learn or interest in learning particular aspects of teaching could be framed as the individual choosing tools that ìfit his or her identity as a teacher. But pre-service teachers may be faced with one set of practices endorsed by university instructors, while a conflicting set of practices may be endorsed by school staff. The same strategy may have different significance in the two worlds. The students might be motivated to learn a strategy that is most congruent with their developing ideal teacher self, but that self develops dialectically in the sometimes conflicting meaning systems of different educational contexts. Multiple contexts might give rise to multiple, competing teacher identities that students must negotiate in their development as novice teachers. This presentation will discuss the theoretical implications of framing motivation and interest in terms of their relationship to identity development. We will illustrate the framework with data from our study of pre-service teacher development. However, we will argue that an identity-development framework could also shed light more generally on the development of stable individual interests in other social contexts.

– 69 –

Meta-analysis of general study orientations

Erkki Olkinuora, University of Turku, Finland

Jarkko Makinen, University of Washington, Finland

In this presentation, we take a look at research focusing on students' general study orientations. General study orientations are regarded as one central viewpoint concerning the motivation and goals of degree studies. The scope of the study extends from early measurements in the late 90's to very recent and even ongoing investigations. The main goal of the presentation is to portray the most common structure of study orientations throughout different higher education institutions representing multiple study fields and various student groups. Findings based on several independent surveys (orientations measured by the Inventory of General

Study Orientations, IGSO) and a comprehensive interview study shows that general study orientations typically take a seven dimensional structure. These dimensions are deep orientation, anxious surface orientation, achievement orientation, systematic orientation, work-life orientation, social orientation, and non-commitment. The meta-analysis of all quantitative data available on the issue as well as the results of the interview study confirms the seven dimensional structure of general study orientations. These orientations are then related to close concepts presented by well-known researchers (like, Entwistle, Vermunt, Marton etc.) in the area of higher education.

A 11 23 August 2005 14:30 - 16:30 Room E113

Symposium

Teacher Professional Development

PROFESSIONAL DEVELOPMENT FOR SUSTAINABLE EDUCATIONAL

CHANGE: ON THE INTERRELATIONSHIP BETWEEN INSTIT

Chair: Michael Eraut, University of Sussex, United Kingdom

Organiser: Philip Adey, King's College London, United Kingdom

Discussant: Michael Eraut, University of Sussex, United Kingdom

Sten Ludvigsen, University of Oslo, Norway

This symposium will address the issue of sustainable change in schools through teacher professional development, from a variety of perspectives. We will investigate the two-way interaction between the school as an institiution, represented by

– 70 –

its senior management and the departmental structure, and teachers as individuals.

Imants et al. study this process over 14 years in one school as a process of bureaucratisation. Adey et al. study the relationship of coaching to sustainable change in schools.

Teachers' professional development and sustained change

Jeroen Imants, Radboud University of Nijmegen, Netherlands

J Luttenberg, Radboud University of Nijmegen, Netherlands

T Carpay, Radboud University of Nijmegen, Netherlands

In this paper we focus on the problem of sustainability of change, as it is related to teachers' professional development. To this end we analyze the developments in one secondary school over a 14 year period, both at the school level and the level of individual teachers. Most insights in educational change and teachers' professional development are based on cross-sectional research and this research has a restricted focus on early implementation. Research on educational change ignores the dynamics in the processes of sense making as well as the problems of institutionalisation and sustained change. By means of a case study of one reforming secondary school for a 14 year period we gain deeper understanding in how these individual and organizational sense making processes contribute to (the absence of) sustained change. In our conceptual framework we build on Giddens' notion of structuration and Mintzbergs' notion of bureaucracy. Six key elements for implementation of innovations in schoolsas identified by Fullan are used to describe the change process in the school. School development was analysed by means of document analysis, followed by repeated interviews with key persons. To study teacher development three teachers who worked in the school during the research period with diverging positions towards the school reform were interviewed twice. Our conclusion is that the process in the school can be regarded as a process of bureaucratisation. in which opportunities for innovative teachers' professional learning and sustained educational change in the direction of self-directed learning structurally are reduced.

– 71 –

The nature of ìcoaching in professional development

Philip Adey, King's College London, United Kingdom

Gwen Hewitt, freelance, United Kingdom

John Hewitt, freelance, United Kingdom

In designing a comprehensive long-term professional development programme intended to introduce radical changes in pedagogy, for the development of higher order thinking in school students, coaching played a central role. This paper will describe the various types of coaching used, including demonstrations, observation and feedback, and team teaching. It will offer some data from interviews conducted with teachers who were coached and with teacher-tutors who became coaches, in a local authority context where an attempt was being made to make professional development systemic. The main aim, however, is to relate the practice of coaching teachers to three theoretical perspectives on pedagogical change: teacher conceptualisations (with implications for necessary conceptual change), reflection on practice (with implications for the role of the tutor as expert), and teachers' intuitive knowledge which develops when new explicit techniques become internalised and automated. It will be argued that these three complementary perspectives provide theoretical justification for the practice of coaching (whose empirical worth has been well established) and also give guidance for forms of coaching which should be most effective (and cost-effective).

Implementing content-focused coaching in large urban school districts: Challenges and design strategies

Fritz Staub, University of Zurich, Switzerland

Donna DiPrima Bickel, University of Pittsburgh, United States

A pivotal question for collaborations between educational researchers and practitioners is how abstract research-based knowledge can be infused into large systems in ways that transform current practice and enhance student learning. The problem of how abstracted knowledge, such as general principles of learning, can be made useful for teachers in classrooms is at the heart what leads to the development and conceptualization of Content-Focused CoachingSM (CFC) (West & Staub, 2003).

CFC is a professional development model for the advancement of student learning and teaching by having a coach and a teacher or a group of teachers jointly plan, enact and reflect on lessons. The model consists of a specific activity setting and a set of theory-based conceptual tools assisting coaches and teachers in conducting content-focused coaching conversations. This study documents and analyzes the

– 72 –

implementation of Content-Focused CoachingSM for the teaching of elementary literacy in three large urban US school districts, which began in the years 2000,

2001, and 2002 respectively. The case studies are based on interviews with staff developers and district personnel, as well as written formative evaluations on the implementation of CFC from coaches, principals and other district personnel. The paper will present and discusses the challenges met, the kind of design problems solved, and begin to make explicit the expertise required to bring an innovative professional development model such as CFC to large school districts.

A 12 23 August 2005 14:30 - 16:30 Room E009

Symposium

Multimedia and Hypermedia Learning

MULTIMEDIA AND PROBLEM-BASED LEARNING

Chair: Jos Beishuizen, Vrije Universiteit Amsterdam, Netherlands

Organiser: Jos Beishuizen, Vrije Universiteit Amsterdam, Netherlands

Discussant: Abbas Darabi, Florida State University, United States

Recently, the development of theories about learning from multiple media, like verbal and pictorial information, or examples and theory descriptions, has received increased attention. Multimedia enhanced learning environments enable learners to make active use of available information by a simple mouse click. In this symposium, questions related to the value of so many multimodal information sources for learning processes and learning outcomes are addressed by reporting experimental research. When both text and pictures are available, how should these modes by presented in order to help the learner to create an integrated semantic representation of both concepts and similarities? When a knowledge domain has to be explored by students preparing for a simulation experiment, do they make use of examples or do they focus on theory descriptions? To what extent are their preferences determined by their learning style? When novice and expert students are working in a problem based learning environment, do they differ in their need of information? And can they be supported to determine their information requirements by conceptual and procedural help? When advanced internet users have to carry out an information search task in a well-known domain, related to their own expertise, or in an unknown domain, do they differ in search strategies? Contributions to this symposium will be discussed in order to contribute to existing theories about multimedia

– 73 –

learning.

Integrating information from text and pictures: How can learners be supported?

Tobias Bartholome, University of Muenster, Germany

Rainer Bromme, University of Muenster, Germany

One of the main cognitive demands in learning from text and pictures is mapping the different representations onto each other and integrating the information from the different sources into a mental model. In this research we examine the effects of two kinds of instructional measures for supporting this conceptual integration process in a computer-based setting: support for mapping and prompts supposed to structure the integration process. Participants (n = 84) learned concepts from the domain of botany with a combination of texts and line drawings in one of four groups (mapping support [numerical labels vs. interactive highlighting of corresponding elements] x structuring prompts [given vs. not given]). Subsequently, cognitive load and confidence in learning were rated. Post-tests included different text- as well as picture-oriented measures of conceptual botanic knowledge (e.g., categorising plant exemplars or giving definitions). In addition, the impact of the learner-related factors self-efficacy, spatial abilities and working memory span on knowledge acquisition was assessed. Main results yielded an interaction between the two kinds of supportive measures as well as a main effect of mapping support.

Results are discussed with regard to prominent resource-oriented models of learning with text and pictures (e.g., generative theory of multimedia learning, Mayer,

2001; cognitive load theory, Sweller, van Merrienboer, & Paas, 1998). Our results call for complementing these resource-oriented models with a conceptualisation of the cognitive and metacognitive processes involved. We propose a new notion of the integration of text and pictures as construction of hybrid knowledge structures, incorporating similarity-based and rule-based components (e.g., Keil, Smith, Simons, & Levin, 1998).

Supporting information interaction during problem solving in powerful e-learning environments

Paul Kirschner, Open University, Netherlands

Liesbeth Kester, Open University, Netherlands

Jeroen van Merrienboer, Open University, Netherlands

Modern curricula are increasingly making use of powerful electronic learning environments (pELEs) to facilitate complex cognitive skill acquisition that contain

– 74 –

realistic practice problems (e.g., simulations etc.), and varied information resources

(e.g., texts, auditory fragments, animation). When confronted with such a realistic practice problem, all learners need certain information to help them construct their own knowledge. They need to determine and act upon their information requirements by choosing and using the right information at the right time. However, novices in a domain do not have a good impression of what there is to know about a particular problem (Ormrod, 2004) and therefore cannot determine which information might help them to solve it. The decisions made by learners on what information to use can result in misconceptions especially when dealing with more complex problems or tasks (Hannafin, Land & Oliver, 1999). In addition, research shows that properly perceiving the problem demands is often problematic for domainnovices (see Broekkamp, van Hout-Wolters, van den Bergh & Rijlaarsdam, 2004 and Broekkamp, van Hout-Wolters, Rijlaarsdam, & van den Bergh, 2002 for task demands in relation to test expectations; Luyten, Lowyck, & Tuerlinckx, 2001 for task perception; and Zumbach & Reimann, 2001 for goal orientation). This impedes novices in acting upon their information requirements.

Learning styles, experience and information seeking

Nelleke van Wouwe, Leiden University, Netherlands

Jos Beishuizen, Vrije Universiteit Amsterdam, Netherlands

Previous studies have shown that learning style and experience affect learning and information seeking on the WWW. The investigators are inconclusive though, if certain learning styles improve search performance more than others. This study tested the effect of learning styles and experience on efficiency, systematic searching and learning on the WWW. Furthermore it characterized different information seeking strategies (by the frequency of behavioural movements and search patterns) per learning style and experience. Forty-two IBM employees were assigned to the experimental conditions by their scores on the Inventory of Learning Styles

(ILS; Vermunt, 1998) and experience. Each participant received two knowledge tests and search tasks. Information seeking behaviour was logged and afterwards categorized on the basis of a behavioural model. Results showed no main effects of learning style or experience on learning, efficiency and systematic searching. An interaction effect of experience and learning style on systematic searching and efficiency was observed though. The influence of experience was more prominent with the undirected learning style than with the meaning directed learning style, which was hypothesized. Unexpected was the high efficiency and systematic searching by experts with an undirected learning style which might be due to absence of nega-

– 75 –

tive learner characteristics in a non-educational setting. The application of the ILS on the work floor has to be further investigated. Information seeking behaviour observed in this study was comparable to the normative model used. The behaviour shown by participants with a meaning directed and an undirected learning style were respectively comparable to expert and novice strategies that resulted from research on problem solving tasks.

Theory-based or examples-based designing and conducting experiments

Herman Jonker, Vrije Universiteit Amsterdam, Netherlands

Jos Beishuizen, Vrije Universiteit Amsterdam, Netherlands

Students differ in the extent to which they spontaneously generate examples when they encounter concepts and principles presented in a study text. Vermunt (1992) called this habit "concrete elaboration". Is it the case that more clear-cut differences emerge between high and low concretisers when students study expository texts in order to prepare for conducting experiments rather than for preparing for a comprehension text? According to the adaptation versus accumulation explanation, students high on concrete elaboration are expected to focus on the theory paragraphs of the hypertext, and are expected to quickly engage in conducting experiments.

A computerized learning environment (Flexible Inquiry Learning Environment;

FILE) was used to allow students to conduct experiments. Feedback was used to further explore the theory paragraphs of the text. Students were expected to seek evidence confirming their initial ideas about the relationships between independent and dependent variables, as collected in the initial theory interview. According to the adaptation versus accumulation explanation, students low on concrete elaboration were assumed to focus on the examples paragraphs of the expository text, and were likely to study the text extensively, before engaging into the experimenting part of the task. According to the deductive versus inductive learning explanation, students low on concrete elaboration were expected to quickly start to experiment and use feedback to further explore the examples in the text. Data will be reported during the EARLI meeting.

– 76 –

A 13 23 August 2005 14:30 - 16:30 Room E111

Symposium

Metacognition

METACOGNITION IN YOUNG CHILDREN

Chair: David Whitebread, University of Cambridge, United Kingdom

Organiser: David Whitebread, University of Cambridge, United Kingdom

Wolfgang Schneider, University of W?rzburg, Germany

Shirley Larkin, King's College, Univeristy of London, United

Kingdom

Marcus Hasselhorn, University of Gottingen, Germany

Discussant: Marcel Veenman, Leiden University, Netherlands

Chair: David Whitebread, University of Cambridge, UK

The development of metacognitive skillfulness is now recognised as being a significant determinant of educational achievement. This symposium is concerned with the development of metacognition in young children.There is increasing evidence that much of the early work on metacognition, which used laboratory-based and self-report methodologies, may have seriously under-estimated the metacognitive abilities of young children. The papers in this symposium present significant recent

European research concerned with the encouragment, identification and analysis of the development of metacognitive abilities in children within elementary education.

Findings include evidence of the early emergence in 3 year olds of an impressive range of metacognitive abilities, evidence that theory of mind in 3 year olds plays an important role in facilitating later metamemory, evidence that teachers' own metacognitive experiences and their views of teaching and learning can affect the metacognitive environment of the classroom for 5-6 year old children, and evidence that task specific rather than domain general metamemory significantly contributes to the emergence of memory strategies during the late elementary school years.

– 77 –

Metacognition in young children: Evidence from a naturalistic study of 3-5 year olds

David Whitebread, University of Cambridge, United Kingdom

Penny Coltman, University of Cambridge, United Kingdom

Holly Anderson, University of Cambridge, United Kingdom

Sanjana Mehta, University of Cambridge, United Kingdom

Deborah Pino Pasternak, University of Cambridge, United Kingdom

There is increasing evidence that self-report and laboratory-based methodologies may seriously under-estimate the metacognitive abilities of young children. Recent work related to metacognition has demonstrated the advantages of naturalistic, rather than laboratory-based, studies with young children (Perry, 1998) and of the advantages of using observation schedules and techniques to evaluate metacognitive learning in young children (Winne & Perry, 2000). This paper presents findings from a 2 year study exploring the development of self-regulatory and metacognitive abilities in young children (aged 3-5 years) in educational naturalistic settings in the UK (English Nursery and Reception classrooms). 32 early years educators collected evidence of metacognitive abilities evidenced by children in their classes during learning activities which were constructed to be 'meaningful' for the children and in other ways most likely to facilitate children's articulation of their metacognitive knowledge and self-regulation of their performance. This evidence consisted of metacognitive 'events' recorded by means of field observations, supported by digital photographs, video of children engaged in learning activities and periodic assessments of children against an observational checklist. Altogether over 1000 such events have been collected and documented, with a detailed analysis of the metacognitive processes evidenced within them being conducted. This analysis enabled the construction of an observational schedule and model of metacognitive abilities which appear to be observable within this age group. The paper will present evidence of the reliability and validity of observations of events using the schedule of metacognitive abilities (CHILD 3-5: Checklist of Independent Learning 3-5). Evidence of the relative incidence of these different metacognitive abilities in the 3-5 age range will also be presented, together with examples of the detailed protocol analysis of particular events. These examples will illustrate the ways in which children in this age range evidence emergent metacognitive abilities during meaningful learning activities.

– 78 –

Knowledge about the mind: Predicting metamemory from earlier theory-of-mind

Kathryn Lockl, University of Wurzberg, Germany

Wolfgang Schneider, University of Wurzberg, Germany

This longitudinal study aims to combine aspects of research on theory of mind and metamemory. Despite the fact that both areas aim to investigate the development of

Children's knowledge and cognition about mental phenomena, the two research literatures have been surprisingly distinct and unconnected. Therefore, the main goal of the study is to examine longitudinal relations between Children's performances on tests of earlier theory of mind and later metamemory. In total, 183 3-year-old children were recruited for this longitudinal study. Children were tested at three time points, separated by a testing interval of approximately one year. At each time of testing, children completed a set of theory of mind tasks including false-belief, appearance-reality and second-order belief tasks. At Time 3, Children's declarative metamemory was assessed in an interview which contained examples from everyday memory tasks as well as from laboratory-like situations. In order to take into account that the relation between theory-of-mind competencies and later metamemory could be mediated by individual differences in general language or nonverbal abilities we also included the assessment of language competencies and nonverbal

IQ. The results demonstrate strong relationships between the domains theory of mind, metamemory, language and nonverbal intelligence. Remarkably, the correlation between theory of mind and metamemory remained significant when the scores for language and nonverbal intelligence were partialled out. Hierarchical regression analyses showed that theory of mind competencies at Time 1 and Time 2 made independent contributions to the prediction of metamemory at Time 3. Overall, the findings indicate that theory of mind plays an important role in facilitating later metamemory.

Creating metacognitive environments in year one and the impact of teachers’ beliefs, opinions and knowledge

Shirley Larkin, King's College, University of London, United Kingdom

Metacognition, the ability to reflect on one's own thinking and monitor ongoing thinking developes with age and through metacognitive experiences. Some environments may facilitate metacognitive experiences more than others. In year one classrooms (age 5-6 years) the teacher is in a powerful position to create an environment that facilitates metacognitive experiences for the children. The specific research questions explored here are: Does teachers' knowledge of and beliefs about

– 79 –

metacognition impact on the development of metacognition in the classroom? Do teachers' beliefs and opinions about learning and teaching affect the way that they facilitate metacognition in the classroom? Can teachers' ability to promote metacognition be developed through professional development? This research was undertaken during a larger cognitive acceleration project ([email protected]), one aspect of which was the promotion of metacognition. The quasi-experimental design of the main project meant that experimental school teachers undertook a professional development programme, which included the theory and practice of facilitating metacognition. Five experimental school teachers and three control school teachers were interviewed at the beginning and end of the intervention year. Through thematic analysis teachers were grouped into descriptive categories named: ìprotective, ìchild centred and ìdisciplined. Over 50 classroom observations of case and non-case lessons were coded for both teacher and student metacognitive behaviours. The results showed that teachers in the ìchild centred category employed more metacognitive behaviours in the case lessons than did other experimental school teachers. However, there were other individual differences and these are described and explored.

In general, the results showed that teachers' own metacognitive experiences along with their views of teaching and learning can affect the metacognitive environment of the classroom. The research highlighted methodological issues, which are also discussed.

Developmental relationships between metamemory and memory strategy use: Task specific or domain general

Marcus Hasselhorn, University of Gottingen, Germany

Maren Richter, University of Gottingen, Germany

Michael Lingen, University of Gottingen, Germany

In a longitudinal study the developmental relationships between metamemorial knowledge about organizational strategies and the usage of category organization as well as between metamemorial knowledge about cumulative rehearsal and children?s rehearsal set size were investigated. In total, 80 9-year-old children were recruited and tested at four measurement points, separated by a testing interval of six months. At each time of testing, children answered a couple of questions aimed to assess either task specific metamemory regarding category organization or task specific metamemory regarding cumulative rehearsal. In addition, children were given two sort-recall tasks and three free recall lists. The mean level of using categorical relatedness during sorting was chosen as a measure of using category organization as a memory strategy. The average rehearsal set size across the three

– 80 –

free recall lists was taken to assess the usage of cumulative rehearsal strategy. The results demonstrate a strong relationship between both kinds of task specific metamemory not before the fourth measurement point. Remarkably, multiple regression analyses for both of the two strategies showed that at the age of 10 years

(measurement points 3 and 4) only the related task specific metamemory but not metamemory with regard to the other strategy significantly contributes to predict later strategy use. Only at the fourth measurement point task specific metamemory regarding category organization and task specific metamemory regarding cumulative rehearsal were substantially correlated. Overall, the findings indicate that it is more task specific than domain general metamemory that significantly contributes to the emergence of memory strategies during the late elementary school years.

A 14 23 August 2005 14:30 - 16:30 Room A110

Symposium

Lifelong Learning and Professional Development

INITIATING CHANGES OF TEACHERS' PERCEPTIONS, KNOWLEDGE

AND ACTIONS IN VET

Chair: Fritz Oser, University of Fribourg, Switzerland

Frank Achtenhagen, University of Gottingen, Germany

Organiser: Frank Achtenhagen, University of Gottingen, Germany

Discussant: Els Boshuizen, Open University, Netherlands

Frank Achtenhagen, University of Gottingen, Germany

Megatrends the global economic, social and political changes as well as increasing heterogeneity of apprentices and students especially caused by the figure of migrants provide new challenges for teaching/training and learning/working processes in the fields of vocational and occupational education and training (VET).

This development has also consequences for the teacher training programs in the field. It seems to be necessary to promote teachers perceptions, beliefs, knowledge and actions with regard to the broad field of teaching and training. With proposed symposium is dealing with this overarching issue by especially focussing on three topics:

(1) Conceptual focus: Teacher schemas of perceptions, knowledge and actions as well as teachers beliefs and epistemologies are discussed with regard to their shape and possibilities of change, especially in direction of expertise (Stevenson and Gru-

– 81 –

ber).

(2) Intervention focus: Two central topics of teaching/training and learning/working in VET are thematized: how to cope with the promotion of motivation and how to initiate intercultural learning? (Winther & Achtenhagen and Weber).

(3) Training focus: By especially conceptualized training firms VET teachers shall be trained with regard to sets of competences needed; above all instruments of diagnosing learning (Oser, Schwaller, Steiner & Renold).The three foci stand exemplarily for patterns of revised teacher training programs as one possible answer to the new global challenges and there consequences for VET: We need a revision of concepts, have to develop new and effectives interventions strategies and should also develop new ways of supporting teacher training by thoroughly constructed cases which has to be systematized punctually and permanently within the teacher training curriculum. This symposium shall open the discussion about further necessary steps of the professional development of teachers for VET and, put emphasis and proposals which can be used within the practice of teacher training.

Targeting vocational teacher schemas through analysis of the objects and instruments of inter-related activity systems

John Stevenson, Griffith University, Australia

This paper proposes an inter-disciplinary approach for illuminating the cognitive schemas that need to be developed in vocational teacher education, and for designing appropriate learning experiences. The cognitive schemas are analysed in terms of the objects and instruments of three inter-related activity systems: vocational practice, vocational education learning settings and vocational teacher education.

Using cultural historical activity theory, the cognitive schemas needed as instruments in vocational activity are derived from considerations of the objects of contemporary working practice. These instruments, in turn, are assumed to be the objects of vocational education preparing vocational learners for the realities of contemporary and future workplaces. Further, the teaching schemas (instruments) that vocational teachers need in vocational educational practice are suggested as an important object of teacher education, preparing teachers for the challenges of vocational education. The paper uses this analysis to argue that an important object of vocational teacher education learning experiences should be the kinds of declarative, specific procedural and metacognitive procedural knowledge that the teacher can, in turn, use in engaging vocational learners in activities appropriate as preparation for work. The required learning activities in teacher education are mainly examined in terms of Stevenson's (2004) concept of memorable activity. In particular, it is

– 82 –

argued that teacher education in this field needs to involve consequential transitions that interconnect the different forms of cognitive artefacts of vocational and vocational educational theory and practice, including personal knowledge of workplace activity, discipline-based (theoretical) knowledge, encapsulated and scripted expert knowledge and the informal knowledge that new learners bring to the setting. Suggestions are made for engaging trainee teachers in appropriate kinds of memorable activity through consequential transitions. These include explicitly designed experiences of eliciting, rendering and articulating different kinds of tacit and codified, personal and public meanings, and the relationships among them.

How to support vocational teachers to become experts?

Hans Gruber, University of Regensburg, Germany

Teaching in vocational schools can be considered as a complex professional domain in which substantial individual differences in the level of expertise exist. However, research on teacher training so far has paid little attention to research on expertise.

In this contribution it is aimed to show both theoretically and empirically that research on expertise provides fruitful ideas how to support vocational teachers. The use of the term "expert" indicates a modern use of the concept of teaching competence (Terhart, 2000). It includes (1) a rich variety of different knowledge forms,

(2) smoothly running routines, and (3) the availability of working episodes that can form the base of subjects' experiences. The interplay of these three components helps to integrate knowledge and action. It is argued that a number of principles of vocational student learning can be transferred to vocational teacher training, in particular the combination of theoretical and practical components as in the German "dual system". Supporting professional learning of vocational teachers needs contributions from three theoretical perspectives (DeCorte, 1997): (1) Knowledge, skills, and epistemological beliefs of teachers of different levels of expertise have to be analysed in order to determine the direction of professional development. (2)

Mechanisms of successful learning from experience have to be identified. (3) Didactical assumptions have to be developed how to design instruction and practice so that professional learning is fostered. In this contribution, empirical evidence from a number of studies is presented that shows how research on expertise can direct future vocational teacher training.

– 83 –

Measurement of motivation: State-perspective of motivation in learning tasks

Esther Winther, University of Goettingen, Germany

Frank Achtenhagen, University of Goettingen, Germany

A central question of teacher training is how can motivation affect learning? There is need for models of motivation in learning tasks that include motivational as well as cognitive processes and that allow us to study learning as a process and not only the output. To focus on the process dimension of learning is necessary in order to explain differences in the prevalence of learning motivation states associated with vocational schools as compared with learning at the workplace. Data portends a negative trend in the frequencies of self-determined learning at school (Prenzel,

Kramer & Drechsel, 2002). On that score motivational influences on the acquisition and application of knowledge in learning tasks should be measured. At the same time it is important not to interfere to much into the learning process. Vollmeyer

& Rheinberg (2000; 2003) as well as Boekaerts (2002) proposed instruments that measure students' motivation by indicating their current motivation in a quasi online procedure. That presupposes that learners can regulate their motivation. We took up this idea and put up a measuring instrument for discussion that (1) makes a forecast of students' performance and (2) gives a description of cognitive, metacognitive and motivational interdependences. The aim of our research is to disentangle these different processes and to show how it is possible to measure simultaneously each of these processes. We analyse the motivational potential of learning tasks by contrasting motivational factors and motivational behaviour. 385 students in schools with a focus on business administration participated in this study. Teacher students as well as experienced vocational teachers co-operate in this project for transforming the procedure and the results into an intervention program for VET teachers.

Initiating intercultural learning using a Design experiment"

Susanne Weber, University of Goettingen, Germany

The necessity of intercultural learning is emphasized in lots of contexts. This is especially the case within the field of vocational education and further education.

Although at a first glance it seems to be intuitively quite clear what is meant, this construct as well as the corresponding terms and objects of this area get vague and hardly tangible. Thus, it does not wonder that concrete measures and concepts for fostering intercultural learning do not follow a theory rather than practical necessities, normative principles or isolated single concepts. This causes many problems for efficient and effective teaching. But these show only minor effectiveness in

– 84 –

intercultural behavior in practice, as diverse studies make overt. Therefore, in this study the concepts of ìculture, ìintercultural communication as well as ìinterculturality should be illuminate and discussed. On the basis of Engestrom's considerations about ìactivity theory and ìexpansive learning as well as those of Ting-Toomey's

ìmindful identity negotiation approach a theoretical framework for ìintercultural learning and development is (re-)conceptualized. In this paper it will be shown how this theoretically reformulated concept guiding the implemented design experiment

- can support the change of individual perceptions and believes, but also changes in intercultural behavior. As the research and practice field of intercultural learning is that diffuse this study is not to be understood as a solution for all intercultural problems. But having the necessity of intercultural learning in mind it is the intention in this study to raise and to continue an intensive (inter-) disciplinary discourse for developing a stringent theoretical model as a basis for intervention to foster intercultural learning and development as well as to initiate changes.

Instructional vignettes as basis for diagnosing teachingcompetencies

Fritz Oser, University of Fribourg, Switzerland

Cyril Schwaller, University of Fribourg, Switzerland

Christine Steiner, University of Fribourg, Switzerland

Ursula Renold, BBT, Bern, Switzerland

In our project we developed standards for teachers behavior, especially for diagnosing teachers competence. In a first step we tried to get a consensus between experts. Criteria are: a) Wholeness of intended action (f.i. to organize and conduct group work as an exercise with a transfer task), b) immediate understanding, c) remembering of similar situations, d) easy handling with respect to memorizing details for judging the quality. A first starting point was filming with professional film teams and actors playing school situations; with should solve the problem of instructional ambiguity with respect to understanding of ongoing learning processes. Finally we decided to not use this path and to give more weight to the validity of the situatedness. Thus we decided to work with real teachers and real students. In this presentation we would like to show exemplarily two film vignettes and discuss their exemplarity and rootedness in the daily process of teaching. Secondly we will present the transformation process of these vignettes into a computer based test instrument for measuring the quality of teacher's competence. The items for the teacher in front of such a vignette are: a) descriptive (what happens in this scene, etc.?), b) diagnostic (what will be the problem for the continuation, what kind of difficulties will arrive?), c) oriented towards quality judgments, d) oriented towards

– 85 –

a professional transferability (how many similar situations the person can produce in a given time slot), e) as professional creativity check (how could we reach the same result with a different methods?), and f) as remembering of similar situations.

- The third step will be the testing of differences in relationship to the judgment: 40 teaching experts will be compared with 40 non-teachers (laymen). First data will be presented.

A 15 23 August 2005 14:30 - 16:30 Room E003

Symposium

Social Aspects of Learning

COGNITIVE AND SOCIO-MOTIVATIONAL ASPECTS IN LEARNING

FROM DYNAMIC VISUALIZATIONS (PART II)

Chair: Peter Gerjets, Knowledge Media Research Center, Germany

Organiser: Katharina Scheiter, University of Tuebingen, Germany

Peter Gerjets, Knowledge Media Research Center, Germany

Discussant: Stephan Schwan, Knowledge Media Research Center, Germany

Contrary to the widespread use of dynamic visualizations in educational software, only very little is known with regard to their design and their instructional functions. This lack of insight may also explain why current literature reviews often fail to report consistent positive results in favor of learning from animations.

In Part 1 of the symposium we address this issue by presenting empirical evidence for the importance of specific design features of animations which influence the cognitive processing of visualizations. Scheiter et al. will first discuss whether animations should entail concrete objects or should be designed in a more abstract way. Schuh et al. demonstrate how animations can foster problem-solving transfer by supporting abstraction across contextually and structurally variable instances.

Huk et al. focus on the impact of three-dimensional animations and visual cues on multimedia learning, whereas Tabbers and de Koeijer investigates how learner control affects the trade-off between effectiveness and efficiency of learning from animations. The second part of the symposium extends this cognitive perspective by incorporating socio-motivational aspects that may additionally influence the effectiveness of learning from animations Linek et al. discuss the design of audio materials for animations from a socio-cognitive perspective by considering the gender of the learner as well as of the presenting voice. Merrill and Atkinson show that

– 86 –

incorporating pedagogical agents as well as animating solution procedures both increase problem-solving transfer. Moreno investigates cognitive and motivational consequences of dynamic visualizations in case-based teacher education. Her results support the claim that more knowledge on moderating factors is needed before animations can be put into practice effectively. Finally, Betrancourt et al. extend the research on animations to collaborative learning scenarios and show that the effectiveness of animations is moderated by the instructional setting in that animations produced positive outcomes only when students learn in a collaborative setting.

Gender-specific design of narrated animations? Effects of using male versus female voices for auditory text

Stephanie Linek, Knowledge Media Research Center, Germany

Peter Gerjets, Knowledge Media Research Center, Germany

Katharina Scheiter, University of Tuebingen, Germany

According to the modality principle advocated by the cognitive theory of multimedia learning (Mayer, 2001) animations with explanatory text are best presented as narrated animations. However with respect to the design of narrations several questions arise, especially which kind of voice should be used. The presented study addressed the impact of male versus female voices in combination with the learner's gender. As experimental material we used a hypermedia learning environment on probability theory with worked-out examples. The worked-out examples were presented as narrations that accompanied dynamic visualizations illustrating the problem statement and the solution procedure. The animations were constructed in a way that made use of the concrete objects mentioned in the example. For every example there was always one animation for the problem statement as well as one for every solution step. Independent variables were the learner's gender and the speaker's gender resulting in a 2x2-design. We measured affective-motivational variables including perception of and motivation towards the speaker as well as cognitive variables including learning success. Results demonstrated that male learners were superior with regard to learning time, cognitive load, and learning success. With respect to the speaker's gender data clearly revealed that learners preferred female speakers. Female voices were rated as being nicer and friendlier than male voices. Learners showed a higher motivation and performed better when listening to female voices. However, no interactions between the learner's gender and the speaker's gender were found. This study demonstrated the importance of voice features for the design of narrated animations. According to our data, we propose to use female voices irrespectively of a learner's gender. Several interpretations based

– 87 –

on the social agency theory will be discussed.

Testing the efficacy of animated versus static worked examples and pedagogical agents

Mary Margaret Merrill, Louisiana State University - Shreveport, United States

Robert Atkinson, Arizona State University, United States

The purpose of this study was to examine the effectiveness of animating workedout examples and providing animated pedagogical agents in a multimedia learning environment involving proportional reasoning word problems. The animated agents included were designed to direct learner attention to appropriate problem states and consisted of: (a) a fully embodied agent, who assisted learners via both verbal and nonverbal modes of communication, (b) a minimally embodied agent, who provided only verbal instructions in the absence of any nonverbal communication cues, and (c) a no agent condition, in which learners received verbal instructions identical to the former two conditions simply without the presence of an on-screen animated agent. An additional goal of this study was to investigate specific types of worked examples incorporated into the computer-based learning environment. The proportional reasoning worked examples implemented in this study were: (a) animated, where the solution steps were gradually unfolded until the example was presented in its entirety, or (b) static, where the solution steps were all presented at the onset of the worked example. Results indicated that participants exposed to the animated agent conditions outperformed students in the no-agent condition. This experiment provides modest evidence to support the claim that multimedia learning environments encompassing animated agents as virtual learning assistants are superior to environments that only provide verbal instructions. The lack of a significant difference between the FE and ME conditions suggests that while the visual presence of an animated agent fosters learning, the agent's mobility is a less important factor.

Additionally, students receiving animated solution steps displayed better transfer performance than their peers presented with static solution steps. This supports the existence of a sequential principle, that is, that animated examples that contain sequentially presented subgoals are superior to static examples that consist of simultaneously presented subgoals.

– 88 –

Using dynamic classroom visualizations for teacher education

Roxana Moreno, University of New Mexico, United States

Ludmila Ortegano, University of New Mexico, United States

How can we help teachers in preparation to effectively apply principles of educational psychology to classroom experiences? One promising technique consists of using cases in teacher education: the presentation of a classroom situation to examine and clarify the concepts involved in teaching practice. In this presentation, we will report the findings of a study which compared the learning and affective outcomes of students who were asked to learn about an educational psychology topic either with or without the presentation of a classroom case. More specifically, we were interested in examining the role of dynamic visual materials when learning from cases. Based on a cognitive-affective theory of learning with media (CATLM;

Moreno, in press) we hypothesized that, compared to reading about a classroom case, watching a video or a computer simulation of a classroom case would help student learning by decreasing cognitive load and increasing their motivation. One of the underlying frameworks incorporated in CATLM is cognitive load theory

(CLT). According to CLT, dynamic visualizations provide an external visualization of the depicted scenario, therefore leaving students more cognitive resources to connect the materials with their prior knowledge than if they had to create a visual representation of their own. In addition, according to CATLM, a way of increasing learning is to promote students' interest and engagement in the learning task.

Similar to other findings on visual aid effects (Moreno, Estrada, & Ahonnen, 2004), we expected students to report investing higher effort and perceiving the learning experience as more interesting and motivating when visualizations are presented in the lesson. To test our hypothesis, we used retention and transfer tests to measure students' learning and a program ratings questionnaire to measure students' affective reactions to each learning condition. The results have theoretical and practical implications for the use of visualizations in case-base learning.

– 89 –

Why did learners in collaborative situation benefit more from animations than individual learners?

Mireille Betrancourt, University of Geneva, Switzerland

Pierre Dillenbourg, Swiss Federal Institute of Technology, Switzerland

Cyril Rebetez, University of Geneva, Switzerland

Mirweis Sangin, University of Geneva, Switzerland

With recent technology advances, computers now offer animated graphic devices, which seem attractive and efficient to instructional designers. However, the research carried out so far failed to establish the advantages of using animated graphics over static ones on learning. Among several problems, animations seem to increase the learners' cognitive load, hence reducing the cognitive resources available for learning. Nevertheless, we believe that, beyond these shortcomings, animations offer unique opportunities to understand dynamic systems. To bypass these shortcomings, we need to deepen our understanding of the cognitive benefits that can be expected from animations in order to turn this understanding into design principles.

The use of animations is not limited to user-system communication but is also often used in computer-supported collaborative learning. In these settings as well, the empirical studies have not confirmed the benefits that one could intuitively expect from the use of animations. This lack of positive results may be explained either in terms of cognitive load, as in user-system interactions, or may be due to the fact that peers use external representation to ground their mutual understanding.

We designed an experiment to investigate the effects of learning from a series of animated sequences or a series of static graphics, in individual or collaborative

(pair) situation. The material was a set of two explanative multimedia documents on geological and astronomical phenomena. We found a significant main effect of animation on retention and transfer performances, with participants in the animated condition outperforming participants in the static condition. Additionally, there was a significant interaction between collaboration and animation for the transfer test, in the sense that only the participants form the collaborative group benefited from animation. These results contradictory with previous findings in the literature are discussed in terms of deepness of processing of an animated material.

– 90 –

A 16 23 August 2005 14:30 - 16:30 Room E002

Symposium

Learning and Cognitive Science

CHILDREN'S DRAWING: ITS RELATION TO LEARNING AND IN-

STRUCTION IN KINDERGARTEN AND PRIMARY EDUCATION

Chair: Nora Scheuer, CONICET - Comahue University, Argentina

Organiser: Eva Teubal, Hebrew University of Jerusalem and David Yellin Te,

Israel

Nora Scheuer, CONICET-Comahue University, Argentina

Discussant: Giyoo Hatano, University of the Air, Japan

The aim of this Symposium is to analyze from different theoretical and methodological perspectives Children's use of drawing, drawing strategies and drawing conceptions with relation to learning and teaching processes. Drawing has been a key object of psychological and educational study along most of the 20th century, but it lost importance in the past 30 years. The decline of interest in drawing as a topic of psychological research might be associated with the loss of vitality of the traditional approaches dealing with it (maturational, psychometric and projective) in the current psychological and educational agenda, concerned with themes as the intervention of metacognitive processes in notational production; processes whereby a child appropriates cultural ways and uses of representation or the ways in which the production of external representations contributes to key sociocognitive processes such as memory, referential strategies, meaning elaboration, explicitation of various components (specific domain knowledge, mental attitudes, self) or representational redescription.Lately, there appears to be a revival of interest in

Children's drawing. This is probably due to two factors. First, the fact that drawing is a typical activity which children are highly motivated to engage in. Second, the new forms of cultural production and communication are increasingly characterized by the use of a great variety of graphic texts. However, learners are most often regarded as readers rather than as producers of pictorial images.In sum, there is a need to generate new theoretical and methodological frameworks so as to permit a fruitful re-incorporation of drawing into psycho-educational investigation. This symposium attempts to contribute in such a direction by bringing together five different studies dealing with Children's drawing strategies and conceptions. In order to gain in depth of analysis we will focus on early and middle childhood / Kinder-

– 91 –

garten and primary education.

Developmental Process of Drawing Three Dimensional Objects

Kiyomi Akita, Graduate School of Education, University of Tokyo, Japan

Koike Wakaba, Department of Psychology, Atomi Women's University, Japan

This study focuses on the developmental process of learning mapping rules in drawing. Brenneman, Massey and Gelman (1996) stated that young Children's ideas about drawing and writing are constrained by domain-specific knowledge about words and objects. Representational drawings relate to the perceptual features of objects. Features of margin, color, and number of sides of an object are mapped to the drawings. Young children learn how to represent and map three-dimensional entities on two-dimensional paper. How do children learn these mapping rules?

Twenty-three 4-year-old children, nineteen 5-year-old children, nineteen 5-yearand-a-half-old children participated in drawing tasks. They were asked to draw a cylinder. Their drawing processes were recorded. One year later, the same children asked to draw the same cylinder and draw a cube task. Their spatial-cognitive ability was also tested, using WIPPSI. The analysis of developmental changes and the comparison between two drawing tasks showed the following major findings. 1)

Most four-year-old children could map the contour figure of the salient sides and color of it on the drawing. But four-year-old children tended to draw only contour, while the older children tended to color the surface area. 2) As for the numbers of sides, Children's preference for realism is seen. Four-year olds and early five year old children tended to draw more sides that could be perceived but the late fiveyear-old children drew the correct number of sides. 3) Four-year-old children invent how to draw the relation between sides of objects in various ways (unit-division strategy, part-whole strategy, etc), but five-year-old children tended to use authentic drawing, vertical linking figures strategy. 4) An influence of language (naming figures) on drawing strategies was evidenced.

Young children’s development of the ability to discriminate between different notational domains- drawing, writing and numerical notation

Esti Klein, Hebrew University of Jerusalem, Israel

Eva Teubal, Hebrew University of Jerusalem and David Yellin Te, Israel

Anat Ninio, Hebrew University of Jerusalem, Israel

Young children have an increasing ability to use symbol systems to interpret, manipulate, and express meanings. Notations are permanent external symbol systems

– 92 –

that fulfil the above functions in particularly powerful and characteristic ways.

Studies of the development of Children's notational abilities are currently viewed as both a way of examining the development of competence with external representational systems and as a means of identifying some of the mechanisms that influence developmental trajectories. There are quite a number of interesting questions related both to Children's ìrecognition and ìproduction of notational systems. The present paper deals with the topic of notation within the framework of ecologically valid tasks: subjects were confronted with tasks which "make sense" to them within different notational domains. Namely, drawing, writing and numerical notation.The questions we asked are: 1)How early do children discriminate between different notational domains as manifested in their production; 2) What is the impact of function of notation upon notational production; 3) When do children become able to use their own notational production in a communicative- referential manner.

Ninety-four children from middle class background participated in the study (50 girls and 44 boys). They were divided into three age groups: I) 33 children, average age 44 months; II) 31 children, average age 59 months; III) 29 children, average age 69 months. Children's production and use of notations were assessed across 3 different tasks: 1) production of a birthday card; 2) production of a shopping list

(including item names and their prices); 3) use of the self-produced shopping list.

The documents produced by the children were coded by eight independent adult judges: four of them were aware of the task context and four were not. We analyze the results from three perspectives: developmental changes, task based changes, and consistency of notations across tasks.

The relation between language, drawing, and sculpturing: A representational view

Esther Adi-Japha, School of Education and Gonda Center for Brain Res, Israel

Tehia Hagoel, School of Education, Bar-Ilan University, Israel

Pnina S. Klein, School of Education, Bar-Ilan University, Israel

Symbolic representation is considered as part of human cognition, originating in the brain. It is agreed among researchers that in different domains a same skill may be expressed differently.In this work we study the links among Children's graphic, language, and sculpture production relating to symbolic development. Two aspects will be studied within these domains: symbolic meaning and form of symbols.

Our working hypothesis is that there is a correlation among representational skills.

However, we expect individual differences. Within those children with advanced linguistic skills and within those with lower abilities we expect that language and other representational skills will be negatively correlated, because of compensation

– 93 –

mechanisms (Lewis & Beard, 1993).Based on previous research (Adi-Japha, Levin

& Solomon, 1998) for eliciting representational meaning of drawings, we study the relation between symbolic representation in drawing, sculpturing and speech. Forty preschool children aged 4; 0-5; 0 years participated in this study and presented with several representational tests. Tests of language include receptive and expressive skills. Tests of drawing and sculpturing rate symbolic meaning (as indicated by the child and by objective judges), and clarity of form. Preliminary results: Using a computational model we have studied the similarity between linguistic and graphic representation in preschool children. Results suggest that individual differences exist: children that draw and speak about the same subjects have one dominant mode of expression. In line with previous research, their total representational score is correlated with language skills.

Children's ideas about different sources for their drawing activity: Looking at the natural world, at a ready-made drawing and into their own mind

Monica Echenique, Comahue University, Argentina

Nora Scheuer, CONICET - Comahue University, Argentina

With the aim of exploring the development of epistemological aspects involved in implicit theories of learning to draw during middle childhood, we interviewed

32 children in first and fourth grade at a public elementary school in Neuquen,

Argentina. The main task tapped Children's appreciation of three different drawing sources, depicted on graphic cards: natural world, a culturally validated, readymade drawing and learner's mental world. Category analysis considered positive and negative choices and the corresponding justifications, in terms of how the informational value of the source and the learner's mental activity were referred to.

A Multiple Factorial Correspondence Analysis was applied to study associations among choices, justifications, school grade and sex. School grade achieved test value; sex did not. Three response patterns were distinguished. Pattern 1 reveals the positive choice of the ready-made drawing and the natural source, on the basis of their informational simplicity/stability and of the mental activity of immediate capture of such external sources. First-graders are associated to this pattern. Pattern

2 reveals the positive choice of all three sources, on the basis of the learner's both reproductive and productive mental activity. Pattern 3 selects both the natural and mental sources as the most helpful, due to their potential to promote productive mental activity. The ready-made drawing is rejected, precisely because it does not promote mental production. Fourth-graders lie in between response patterns 2 and

3. We interpret these patterns as indicating different epistemological perspectives

– 94 –

that in turn underlie different implicit theories of learning of drawing. We propose that a better understanding of Children's view about learning to draw may be a fruitful road to enrichen psychologists' and educators' view of this process and hence enable them to promote it in a more comprehensive way.

Children’s models in scientific knowledge construction

Maria Arca, Molecular Biology and Pathology Institute, Italy

Andres Acher, University of Barcelona, Spain

This research concerns a long-term study in twenty classes of Kindergarten (ages three to six, Modena) and Primary School (ages six to eleven, Rome, Venice and

Turin) in different socio-economical environments. The first aim of the work is to explore how children, involved in scientific modelling activities, represent in their drawings how they imagine what they cant see, about different biological or physical processes. The second step concerns the ways in which teachers can be trained to promote the use of models in their didactical activities. We promoted and analysed pupils' productions collected throughout their complete scholastic period

(3 years for Kindergarten, 5 years for primary school). On the basis of action-research methodology, we engaged pupils in long-term, in-depth projects where they were presented with opportunities to communicate and compare their ideas orally verbally and graphically, as well as to modify and exploit the graphic models they had formulated individually or in small groups. We found that in a learning context where ìto think is a value, children learn to freely express their scientific imagination, going beyond the stereotypical images typical of textbooks: they do not imitate adult schemes, but rather construct and express their own ideas about phenomena.

We understand models as metaphorical interpretations of reality that show at the same time aspects of facts and of Children's ways of thinking about them. Hence, working with models enables teachers to observe changes in Children's knowledge, their representation of new relations among facts and the emergence of different cognitive strategies. Contrasts among individual models provide the teachers with rich data to reflect on difficulties and acquisitions in Children's thinking and help them in the continuous re-structuring of their didactical project.

– 95 –

A 17 23August 2005 14:30 - 16:30 Room E004

SIG Invited Symposium

Metacognition

ACTIVATING METACOGNITIVE PROCESSES IN THE CLASSROOM

Chair: Zemira Mevarech, Bar-Ilan University, Israel

Organiser: Zemira Mevarech, Bar-Ilan University, Israel

Discussant: Zemira Mevarech, Bar-Ilan University, Israel

The purpose of the present symposium is to examine how teachers and students activate various kinds of meta-cognitive processes during the solution of complex problems. The symposium includes four presenters and a discussant. All four presenters will focus on meta-cognitive processes that take place in the classrooms, and all will refer to mathematics complex problems. Yet, each presentation is based on a different methodology, as follows. Bracha Kramarski from Israel examines the development of two groups of teachers, each exposed to a different kind of professional development course: one implements meta-cognitive guidance, and the other introduces the notion of meta-cognition without really training teachers to use it. Kramarski reports substantial differences between the two groups on teachers' pedagogical content knowledge. Marcel Veenman and his colleagues from the

Netherlands analyze the roles of meta-cognitive skills in different kinds of learning episodes. Using protocol analysis of high school students solving a series of probability-calculus tasks, Veenman et al show a differentiation in meta-cognitive activities for each kind of Instructional Learning Episodes, along with differences in the contribution to learning outcomes. Eckhard Klieme from Germany examines meta-cognitive processes that high-school students activate during the solution of

PISA problems. Klieme focuses particularly on self-regulation as well as the relationships between item features and item difficulty. He finds that complexity of reasoning processes highly correlate with self-regulation. Finally, Ruhama Even from

Israel follows the professional development of mathematics during three years. Her study indicates that only when teachers reflect on what they are doing, they improve their pedagogical-content knowledge. The discussant will focus on the theoretical, methodological, and practical implications of all presentations.

– 96 –

Integrating knowledge and practice as a means to raise PD providers' awareness of their own conceptions and beliefs

Ruhama Even, Weizmann Institute of Science, Israel

The aim of this research is to examine the potential of integrating knowledge and practice as a means to raise awareness of providers of PD (Professional Development) for teachers of their own conceptions and beliefs regarding the following two aspects: (1) the nature and quality of school mathematical tasks, and (2) the nature of students' ways of learning and knowing mathematics. Thirty PD providers who participated in the MANOR Program participated in this study. Analysis focused on data related to two multi-stage activities on the above mentioned aspects. Main data sources include questionnaires, interviews, documentation of, and reflection on, Program meetings and PD activities conducted by the participants, and their yearly portfolios. Data analysis indicates that getting acquainted with theoretical background by reading and discussing research papers helped to increase participants' interest in, and knowledge about, the issues. But only when participants were requested to use the theoretical ideas in connection with practice and to reflect on these experiences, did they become aware of their own conceptions and beliefs, and were even able to modify and expand them. Still, analysis of attempts to enact knowledge in practice shows that learning new practices and reflecting on this learning should accompany the more remote from practice work in order to connect the new awareness to a more complex modification of conceptions and beliefs, knowledge and practice. The results of this research point to promising ways of raising PD provides' as well as teachers' awareness of their own conceptions and beliefs, so that knowledge can be enacted in practice.

The role of metacognitive skills in different types of learning tasks in the domain of math

Marcel V. J. Veenman, University of Amsterdam, Netherlands

Marleen Spaans, Leiden University, Netherlands

Bernadette H. A. M. van Hout-Wolters, University of Amsterdam, Netherlands

Marianne Elshout-Mohr, University of Amsterdam, Netherlands

Instruction Learning Episodes (ILEPs) are different types of learning tasks, characterized by being either productive or reproductive of nature, by being either knowledge or skill based, by being metacognitive or not, and by aiming at near or far transfer. This study focuses on the role of metacognitive skills in different ILEPs within the discipline of math. Twenty 14-15 yrs old secondary-school students com-

– 97 –

pleted a series of probability-calculus tasks while thinking aloud, with each task representing a different ILEP. Metacognitive activities were assessed for each ILEP task through protocol analyses. Separate posttests were administered for each ILEP.

Results show a differentiation in metacognitive activity between ILEPs, along with differences in the contribution to learning outcomes for ILEPs.

Effects of general vs specific metacognitive training on teachers' mathematical professional development

Bracha Kramarski, Bar-Ilan University, Israel

Tali Revach, Bar-Ilan University, Israel

The study investigates the effects of general vs. specific metacognitive training on teachers' mathematical professional development. Professional development was measured by teachers' mathematical knowledge, pedagogical-mathematical knowledge and self-regulated learning (e.g., Boekaerts, 1997). Participants were

64 primary school teachers who were exposed to metacognitive professional development programs during one year in Israel. Thirty teachers were assigned to the general meta-cognitive training and thirty four teachers were assigned to the specific meta-cognitive training. The training was based on the IMPROVE metacognitive questioning approach that emphasizes the use of four main questions directed to

ìeffective learning/teaching model: Understanding questions; connection questions; strategy use questions; and reflection questions (Kramarski & Mevarech, 2003).

The teachers in general meta-cognitive training were exposed to the importance of metacognitive questioning in general, while the teachers in specific meta-cognitive training practiced the metacognitive questioning explicitly in their mathematical discourse. Three measures were used in the study: A pre/post real-life task (OECD,

2003) that assessed teachers skills on algebraic reasoning, mathematical argumentation, and transfer knowledge, a pedagogical task that asked the teachers to plan a lesson for their students on the real-life task and a pre/post self-regulated learning

(SRL) questionnaire (Kramarski & Mevarech, 2003) that assessed teachers' strategy use in teaching problem solving, argumentation, planning and reflection on the lesson. Results indicated that the specific meta-cognition teachers outperformed the general meta-cognition teachers on various skills of solving mathematical real-life tasks, mathematical argumentation, transfer and mathematical pedagogical knowledge regarding planning a lesson on real-life task. Findings further indicated that at the end of the study all teachers (specific and general meta-cognition training) improved their self-regulated learning skills but no differences were found on SRL between the two groups. Educational and practical implications will be discussed

– 98 –

at the conference.

Metacognitive regulation and levels of problem solving competence

Eckhard Klieme, German Institute for International Educational Res, Germany

Problem solving is highly popular in the human resources literature as well as in educational policy, pedagogy and even in curriculum development. In recent years, several attempts have been made to invent measures of problem solving competence. In order to understand the meaning of the scale, test developers do in-depth task analyses and try to explain item difficulty by task demands. Within the ALL pilot study (N=2102 adults from five countries) most important features were (a) the number and connectivity of information elements and (b) the cognitive complexity of the reasoning, both rated by experts. Within PISA 2003, the number of constraints or variables as well as the number of representations used were identified as most important task demands. In many of the most difficult tasks, students had to consider a rather large number of aspects (conditions to be fulfilled, variables, temporal and other restrictions to be taken into account). Constructing a solution for those items will most probably require the student to work back and forth between his solution and the conditions laid out in the problem description. Students have to organize and monitor their thinking while working out their solution. Thus, the degree of mental load and the complexity of self regulation during problem solving contribute to the difficulty of the task. To sum up: Independent studies came to the conclusion that the amount of metacognitive regulation needed to solve an item has a large effect on item difficulty in the assessment of cross-curricular problem solving competence. Especially, the top level of problem solving competence in each of these assessments is characterized by the ability to coordinate a great number of task components (variables, constraints, representations). Metacognition seems to be crucial for problem solving competence as it is measured in these large scale assessments.

– 99 –

A 18 23 August 2005 14:30 - 16:30 Room A111

Symposium

Education for Citizenship

TEACHING AND LEARNING IN CITIZENSHIP EDUCATION

Chair: Nava Maslovaty, Bar Ilan University, Israel

Organiser: Nava Maslovaty, Bar Ilan University, Israel

Discussant: Cees Klaassen, Radboud University Nijmegen, Netherlands

Citizenship education is an important issue in many countries. Different theoretical concepts are used and different types of practices are emerging. Theories of moral development and democratic education are part of these concepts. Research into processes of teaching and learning in citizenship education is necessary to support these practices. In this symposium we will present four papers from different parts of the world.

The inclusion of roma children from the belgrade deponija enclave into mainstream education

Macura Milovanovic Suncica, University of Belgrade, Yugoslavia

This paper describes an action study that investigates the inclusion process of Roma children from ìDeponija enclave (Belgrade, Serbia) into mainstream education system. The aim of a study is revealing, defining and explaining the key points/obstacles in the Roma children inclusion in the education system; understanding the meanings that main research participants attribute to the phenomena that go along with the inclusion process. The study method is an action research. Research participants are 60 Roma children of primary school age, their families, the school headmaster and teachers, members of the research team.

The design of action research includes four cycles, carried out during two school years, for a period of 32 months (Oct. 2000 - June 2003). Each cycle consists of defining the general problem or the question, creation of the action plan, action implementation and monitoring, outcomes evaluation, outcomes analyses and interpretation. Based on the reflections of the previous cycle, in the next cycle the problem is being redefined, as well as the action plan, followed by its implementation, evaluation, etc. The study describes an application and evaluation of the holistic model of inclusion, with a concept of linking all relevant actors of inclusion within

– 100 –

the process; and their educational and psychological empowerment, in order to motivate them for participating in finding solutions for the problem of Roma children education exclusion. Upon the third research cycle, the results indicate that Roma community have acquired motivation for their children to be included in schooling, but that within the social context of teaching/learning yields one of greatest obstacles for integration.

Upon the fourth research cycle, the results reveal teacher's awareness of Roma Children's improvement in school achievements, but unchanged Roma Children's lowest sociometrical status indicate that their teachers lack interpersonal competencies needed for creating safe and accepting environment in multicultural classroom.

Teacher approaches to civic education in Mexican primary schools: an ethnographic case study

Romali Rosales Chavarria, Department of Educational Researches, Mexico

Education for democracy is high on the international agenda but also has direct implications for teachers and student in the classroom. The structure of the Mexican education system constrains the way in which teachers approach civic education in their everyday work. This paper presents and analyses how teaching of civic education contents is organized in public primary schools from an ethnographic perspective. Teachers put their practical knowledge into practice in order to cope with some of the conditions of civic education. Civic education can hardly compete with the sheer weight of subjects like Language and Mathematics, especially when there is a lack of time available for teaching purposes and overburdened teachers.

There are no textbooks or teaching materials provided to the teachers, so they look for materials they can use for this subject with their students. During the classes, children and teachers interact with a variety of texts in the constructions of school knowledge. Knowing more about the way in which civic education takes place in specific school contexts can contribute to the understanding the way in which the content being taught define some teaching decisions and strategies. It can also contribute to the understanding of teacher thinking, instructional strategies, classroom discourse and social aspects of teaching.

– 101 –

The contribution of professional background, perceived ideal student and school context in explaining teachers’ innovative and democratic teaching strategies

Nava Maslovaty, Bar Ilan University, Israel

Yinon Zadik, Bar Ilan University, Israel

This study aims to examine the contribution of teachers' professional background, professional beliefs and the school context in explaining their choice of teaching strategies. Two approaches to teaching are examined: the traditional-authoritative and the innovative-democratic. Teachers' belief systems affect their modes of thinking and hence their teaching strategies. Teachers' perceptions of the ideal student provide an opportunity to examine how teachers' beliefs are manifested in their work. The effect of the subject discipline on teacher's thinking and teaching processes show that, mastery of knowledge, beliefs concerning the nature of the knowledge structure, and the way knowledge is constructed by students are of particular importance in choosing teaching strategies. The school context affects teachers' beliefs and teaching strategies. A questionnaire was administered to 170 teachers in 9 Israeli high schools. Findings: 1. Low to moderate correlations between the four ideal student indices and innovative teaching strategies. Positive correlations between higher-order thinking, social, and emotional competencies and innovative strategies. 2. Differences were found between teachers who taught biology, chemistry and physics and those who taught computer science, electronics and mathematics with respect to innovative strategies. 3. The four school context indices: teacher community, care for students, professional commitment, satisfaction, were positively correlated with innovative strategies. Correlations were found between technical culture, high expectations and professional commitment, and conservative strategies. 4. The indices that predict innovative strategies are: care for students, higher-order thinking, and professional commitment. It would be more accurate to view teaching strategies as located on two parallel continua, innovative- democratic and traditional-authoritative that are neither dependent on nor related to each other.

– 102 –

A 19 23August 2005 14:30 - 16:30 Room A107

Interactive Poster

Computer-supported Learning Environments

ASYNCHRONOUS MEDIATED COMMUNICATION IN LEARNING CON-

TEXTS: COMPARING METHODS

Chair: Maria Beatrice Ligorio, University of Bari, Italy

Organiser: Maria Beatrice Ligorio, University of Bari, Italy

Stefano Cacciamani, University of Valle dAosta, Italy

Donatella Cesareni, University of Rome, Italy

Discussant: Robert-Jan Simon, IVLOS, Netherlands

Asynchronous mediated communication has many potentialities for educational contexts. Environments, such as forums, seem to suit schools needs and constrains and well serve educational aims, such as collaborative learning and knowledge building. Several methods can be used to analyse forum based discussions, depending on features of the software, participants' goals, nature of the task, and so on.

This symposium aims at comparing different methods and at discussing critical points concerning the relationship between methods, aims, and nature of the data.

The papers here included are looking at different aspects of the learning process and for each of them a specific method is proposed. Cacciamani's et al. paper focuses on a set of methods inquiring metacognitive skills, awareness of knowledge building in university students, and relationship between virtual identity and learning. Veermans and Lallimo's contribution presents three case-studies of students with different learning profiles. Traditional instruments (such as questionnaires) are combined to the analysis of on-line performances. The third proposal (Chan and

Aalst) describes the embedded and concurrent assessment, a complex strategy able to analyse individual and collective advances and to mix computer-based tools and in-classroom pedagogical principles. Mamalougos and Kollias's paper proposes a method to deeply inquire cognitive and conceptual changes on the domain of psychics. All the papers intent to capture the complexity of discussions fostered by sophisticated asynchronous tools. In fact, two common points can be outlined: a) All papers inquiry about web-based platforms designed upon strong theoretically principles. The implicit assumption is that, when virtual space is well designed and fully integrated into a well organized setting, which includes also face-to-face tools and materials, than discussion occurring in the forum is rich, deep, and valuable; b) Discussions triggered through and within asynchronous spaces promote new di-

– 103 –

mensions of the knowledge building process. These dimensions call for new methods.

Complex methods to analyze asynchronous interactions: The CKBG research experience

Stefano Cacciamani, University of Valle dAosta, Italy

Donatella Cesareni, University of Rome, Italy

Maria Beatrice Ligorio, University of Bari, Italy

The Collaborative Knowledge Building Group (CKBG) is an Italian community of researchers from different fields (sociology, educational psychology, pedagogy, economy, and political study) and teachers (from kindergarten up to university professors) interested on studying the use of technologies in educational and training fields. This group meets into the Knowledge Forum (KF), a discussion forum designed by Scardamalia and Bereiter (1999). KF is a tool with specific features designed upon socio-constructive principles, such as for example introducing scaffolds into the notes to guide writing and reading. The CKBG uses KF as virtual space to meet but at the same time as a tool upon which reflect collectively. This paper is aimed in particular at giving a survey of the three main directions through which the research activity of this community developed during the last two years: a) use of the metacognitive reflection to study personal representation of the activity undertaken in a Knowledge Building Community, b) on-line collaborative knowledge construction at University level with reference to the Progressive Inquiry Model

(Hakkarainen, 2003) and to the argumentation strategies (Pontecorvo and Girardet,

1993), and c) the relationship between digital identities and learning, including the perspective of dialogical self and mediated identities (Hermans, 2004). For each of these research directions the method of analyzing web-forum interactions will be presented and the main methodological issues will be discussed. From the methodology array presented, activities taking place into a web-forum environment emerge as a specific way of discussing and the relationship between research questions and methodology used will be out-lined.

– 104 –

Multimethod approach for analyzing students' motivational profiles and their participation in virtual collaboration

Marjaana Veermans, University of Turku, Finland

Jiri Lallimo, University of Helsinki, Finland

In the new complex learning settings, the conventional research methods are not entirely applicable, and new means for investigation are needed. Conducting research that combines qualitative and quantitative methods has been characterized as the multimethod approach. The aim of this study was to explore how (49) students with different motivational and individual profiles participate in a distance-learning environment. Collaborative learning occurred asynchronously, and two teachers were tutoring the students throughout the entire course. Three types of questionnaires were used, in addition, a selection of learners' postings for the group assignments were analyzed. A cluster analysis was used to classify students on the basis of the questionnaire variables. The results clearly showed that each of the three groups had a rather distinct motivational profile. A more detailed analysis of the nature of students' participation was conducted for three cases, each of them representing different motivational profile. The individual cases showed that students with different motivational profiles had different participation patterns, different learning paths, which ended up at same grading level of course performance. This may indicate that problem-based/inquiry-based learning makes possible for the students to choose different types of learning paths. However, in order to understand the dynamics of context-sensitive constructs, such as interest and motivation, application of multi-method approaches may not be enough as such, but they should be used in a holistic way. With the holistic view we mean treating the concept/phenomenon under investigation as one whole that is comprised by interrelated, and changing dynamics between parts; an event or action is explained by identifying its place in a pattern that characterizes the ongoing processes of change in the whole system.

Assessing knowledge building using embedded and concurrent assessments

Carol K.K. Chan, Faculty of Education, China, Peoples Republic of

Eddy Lee, Raimondi College, China, Peoples Republic of

Jan van Aalst, Simon Fraser University, Canada

This study examined the assessment of knowledge building in the context of asking students to assess their own knowledge advances on Knowledge Forum (Scardamalia & Bereiter, 2002). Specifically, we sought (a) To design an embedded and concurrent assessment approach to characterize knowledge building in classroom

– 105 –

context, and (b) To investigate the roles of the assessment approach in fostering students' knowledge building and conceptual understanding. 120 students studying in three Grade-Nine classes in a regular high school in Hong Kong participated to the study. An embedded and concurrent assessment approach, examining multiple aspects of knowledge building, was designed. Three components were included:

(1)Participation, Engagement, and Collaboration. A software called Analytic Toolkit

(ATK) showing quantitative indices of database usage (e.g., number of notes read, written, revised etc.) was used by the teacher to monitor student participation and growth, (2) Individual Knowledge Advances. Students contributed computer notes rated for depth of inquiry and discussed different models in class, and (3) Collective Knowledge Advances. Students produced an electronic portfolio identifying and documenting how clusters of notes reflected collective knowledge advances in the computer discourse. To investigate the assessment approach, different design conditions were examined in three classrooms using: (a) Knowledge Forum, (b)

Knowledge Forum with portfolios; and (c) Knowledge Forum with portfolios guided with knowledge building principles. Results showed different patterns of collective knowledge advances as reflected in students' portfolios - ideas were viewed as artifacts that were examined and improved. Students in the Grade 9 classroom using knowledge building portfolios outperformed other students working on participation, depth of inquiry, and conceptual understanding. These findings suggested that giving students the agency to assess their community's knowledge advances may scaffold their metacognitive understanding of the knowledge-building process and provide an alternative approach for assessing and scaffolding knowledge building in computer networked environments.

Design a methodology to analyse structured discussions stored in the asynchronous space of SYNERGEIA

Nektarios Mamalougos, University of Athens, Greece

Vassilios Kollias, University of Thessaly, Greece

In this study, knowledge building inquiry and discourse among 8th grade students are examined. The Web based CSCL software ´Synergeiaª was used and students dealt with scientific questions included into school courses on electric circuits. During the design of the analysis we looked at: knowledge building, collaboration, and development of metacognitive skills. Mainly, attention was given to conceptual understanding of scientific concepts, such as models for the flow of electricity, role of battery in a simple electric circuit, and relevant energy transformations. Main features of the collaborative computational environment, which was extendedly

– 106 –

used, are: the asynchronous discussion tools, that allowed storing the dialogues; and participants labelling of each of their contributions by using a ìthinking type categorization. Beside the discussion data bases, tests were administrated before and after the proposed activities. Also, some classes participated as control groups.

Results so far indicate that progresses were made on conceptual understanding and collaboration. Methods used to analyse the data will be here presented and discussed. We found out that some new indicators were needed when deeply analyzing the data bases of discussions. For the type of activity we proposed, a special teacher/technology structure was necessary to monitor and analyze a learning environment for collaborative knowledge building. This structure allowed assessing the depth of the investigation and the strategies found to resolve difficulties; allowed monitoring students' role in searching interaction sources. Discourse occurring between pairs working at the computer was recorded, transcript, segmented and categorized. A segment can take on many purposes: it could provide explanations of phenomena, especially causal explanations that refer to possible mechanisms; it could also include arguments, and finally it could also function as monitor, referring to a procedure of controlling one's own or others' understanding. A segment could also provide many other functions socially oriented.

A 20 23 August 2005 14:30 - 16:30 Room A018

SIG Invited Symposium

Research Methodology

QUALITATIVE AND QUANTITATIVE APPROACHES TO LEARNING

AND INSTRUCTION - OPENING A NEW SIG

Chair: Heinz Mandl, University of Munich, Germany

Organiser: Guenter Huber, University of Tuebingen, Germany

Philipp Mayring, University of Klagenfurt, Austria

Discussant: Mechthild Kiegelmann, University of Tuebingen, Germany

This is the opening symposium for the newly installed EARLI-SIG "Qualitative and quantitative approaches to learning and instruction". The aim of the symposium is to discuss the role of research methods in the study of learning and instruction.

Modern instructional sciences are strongly oriented in their empirical work toward quantitative approaches (experimental designs of intervention research, correlational studies, large scale questionnaire studies). On the other hand the quantitative

– 107 –

approach often had been criticized and more exploratory and descriptive studies are claimed (qualitative turn and interpretive turn of social sciences). Many researchers argue that a strict dichotomy between qualitative and quantitative approaches is misleading and unproductive and are looking for combinations and integrations of qualitative and quantitative research. The symposium tries to balance the possibilities and limitations of qualitative and quantitative approaches to learning and instruction and to discuss the role research methods.

Learning from learners

Ference Marton, Goeteburg University, Sweden

The descriptive approach of Phenomenography has in recent years been developed to a theory with explanatory aims. This theory has been tried out in about 150 so called Learning studies in Hong Kong. The model of the Learning study builds - in addition to the theory - on the idea of the Japanese Lesson study and the methodology of Design experiments. A Learning study is carried out by a group of teachers, who teach the same subject at the same level,and one or more researchers. First, a topic is chosen that is important, difficult and can be dealt with in one lesson-or in a few lessons. A pretest consisting of open-ended questions is administered to the participating students. From their answers, their difficulties are identified in form of "critical features", distinguishing between mastering and failing to master what is expected to be learned (the object of learning). A lesson design is developed, aiming at creating the necessary conditions for appropriating the object of learning.

These conditions are defined- in accordance with the theory- in terms of patterns of variation and invariance. After the lesson the test that was previously used as pretest is administered again as posttest. On the ground of the pre- and posttest differences and the course of the lesson, the group of teachers and researchers may revise the lesson plan and the previous sequence of pretest-lesson-posttest-discussion is repeated. Up to 4-6 such sequences might be carried out before the study is finished by being documented. The results are so far truly remarkable.

Processes and products as objects of research on learning and instruction

Guenter Huber, University of Tuebingen, Germany

Since decades, on the one hand the majority of teaching/learning studies is focused on products of learning, and on the other hand critics demand that learning processes should be considered additionally. This paper outlines the E-L-I-T-E model of learning research, which tries to reconcile research routine and critique. The cycle

– 108 –

of questions addresses expectations of teachers, who determine learning situations, which - depending on individual differences of learners (competencies, preferences, orientations, etc.) - allow various transactional processes between learners and subject matter that lead to learning effects, which in turn may modify teachers' expectations. Based on concrete empirical examples, the presentation will concentrate on a discussion of methods matching the stages of this cycle and promising answers to the questions linked to the model's components. Some of these answers can be found most probably by quantitative methods, while others demand a qualitative approach. This rises a pivotal further question, whether findings of mixed-methods studies can be generalized at all or which of these findings may be generalized and thus applied to resolve practical problems of learning and instruction. The presentation will approach this problem by analyzing, what "generalization" means within various types of practical problems, that is, questions of learning and instruction investigated in a range of research designs.

The problems and potentials of mixed methodology in the study of learning and instruction

Philipp Mayring, University of Klagenfurt, Austria

The central aim of the conference is integrating multiple perspectives and reflecting different approaches to enhancing our knowledge of learning processes. This paper tries to contribute to that aim by discussing the possibilities on a methodological level. The potentials of the mixed methodology approach in the study of learning and instruction is elaborated. First the theory of science background of the qualitative-quantitative debate is elaborated, analyzing the different paradigm conflicts in social sciences (natural versus human science approach, positivism debate, constructivism, interpretive turn). We conclude, that an overcoming of the strict qualitative-quantitative dichotomy is necessary, but a theoretical background for mixed methods approaches is needed (pragmatic, dialectical or constructivist theory). In the next part we try to give an overview about mixed methods models. We can differentiate between combination models (e.g. sequence models, pilot study models, generalization models), which put together methodological approaches in an additional, complementary sense and integration models, which try to put the different qualitative and quantitative approaches under one general study design (e.g. triangulation approaches). We can see, that a broad range of different models of mixed methodology had been developed within the last years. The potentials and problems of mixed methods models are demonstrated by examples in the field of learning and instruction (e.g. PISA studies, Qualitative Content Analysis). Mixed methodology

– 109 –

seems to be a promising solution of paradigm conflicts in social sciences. We have to collect examples and exchange experiences in such mixed methodology projects to come to methodological knowledge and technical rules of such combination and integration models.

Forms of learning in the rhythm of education

Hannu Soini, University of Oulu, Finland

Mark Flynn, University of Saskatchewan, Canada

In this presentation the authors propose that the critical incident technique offers a promising means of studying learning from the perspective of students, particularly in higher education. The potential contribution of the critical incident technique is significant because it can illuminate the authentic aspects of students' personal learning experiences. Authentic learning experiences are defined as the concrete personal experiences of students as described by themselves, independent of the abstract theoretical presuppositions of researchers. The second aim of our presentation is to consider how these authentic descriptions of personal learning experiences might be consistent with Whitehead's (1942) concept of the rhythm of mental growth. Whitehead has pointed out that, what is learned should be grounded in the experience and involvement of the learner. Moreover, there is a rhythm in the learning process where the learner freely contemplates experience, disciplines their thinking, and then freely and creatively applies what has been learned. There appears to be a natural symmetry between the descriptions of learning provided by students in the study and Whitehead's description of the rhythm of mental growth.

For example, the fact that emotional significance constituted the most important single aspect of learning for students seems to be consistent with Whiteheads idea that romance - emotion - is always present in any real learning experience. We also found the importance of autonomous reflection - freedom - in the student's description of learning very interesting. Our analysis leads us to conclude that Whitehead's theory of mental growth has merit and could inform research in psychology and the formulation of new theories of learning that take greater account of: (i) human aesthetic experiences such as emotions; and (ii) the importance of extended periods of freedom or autonomy in the learning process.

– 110 –

A 21 23 August 2005 14:30 - 16:30 Room A019

SIG Invited Symposium

Reading

ASSESSING READING COMPREHENSION SKILLS AND STRATEGIES:

TOWARDS AN EUROPEAN FRAMEWORK

Chair: Eduardo Vidal-Abarca, University of Valencia, Spain

Organiser: Eduardo Vidal-Abarca, University of Valencia, Spain

Jean-Francois Rouet, University of Poitiers - CNRS, France

Discussant: Bernadette van Hout-Wolters, University of Amsterdam,

Netherlands

There is a great concern in all European countries about students' reading comprehension difficulties. Those difficulties are more apparent now than before because of the great distance between the skills that the Knowledge Society demands and what the education system is able to provide with. A specific problem in this regard is that classical comprehension assessment tools are not helpful to orient educators about which comprehension skills and strategies should be assessed, and how it should be done. Reading comprehension research has produced advances in the last ten years to contribute to remediate this situation. There is now data to provide the educators with guidelines about new reading comprehension skills and strategies and about how they could be assessed. This symposium is aimed at dealing with this issue. More specifically, three are the main topics:

1) Methods to asses reading comprehension skills and strategies, e. g., questionanswering, on-line measures, etc. More specifically, possibilities, limitations, and compatibility among new and classical methods will be discussed.

2) New reading comprehension skills and strategies for new tasks and settings (e. g., constructing spatial and interpersonal mental models, detecting inconsistencies, finding a hierarchical order in the text, searching information, etc), and new measures to assess these new comprehension skills and strategies.

3) From research to practice. Could the new methods and tasks inspire the development of new assessment tools to be used by practitioners? How could it be done?

Researchers from four different European countries are exchanging research results and ideas about these three topics. It is the first step to elaborate a general framework to asses students' reading comprehension difficulties at the European level.

– 111 –

Assessing different sub-abilities in reading comprehension: An Italian project

Pazzaglia DeBeni, University of Padova, Italy

Research in the last decades has demonstrated that reading comprehension is a complex task which requires both cognitive and metacognitive abilities. Reading comprehension develops throughout life with the acquisition of new competencies, knowledge and strategies. One of the primary tasks of school education is to enhance reading comprehension from the first phase of learning to the acquisition of advanced competencies. In pursuing this goal, the assessment of different abilities involved in comprehension is crucial. The present paper presents an Italian project

- New Guide to Reading Comprehension, by Rossana De Ben et al.- that aimed to improve reading comprehension in 8- to 14-year old children. In this project the reading comprehension's unitary process was subdivided in 10 different sub-abilities, shown by research to be relevant to comprehension and varying from the more traditional (i.e. finding text's characters, places and times) to more innovative tasks

(i.e. constructing spatial and interpersonal mental models, detecting inconsistencies, finding a hierarchical order in the text). Achievement in each sub-ability was assessed by means of specific reading comprehension tests, which differed for content and difficulty level, so that the entire project had 10 comprehension tests (one for each specific sub-ability), each proposed in 2 levels of difficulty, for a total of

20 tests. They were administered to a sample of 1155 children from schools situated in different geographical areas within Italy. Participants were divided into two groups: 782 (399 m., 383 f.) aged 8-11, and 373 (130 m., 143 f.) aged 12-14. Tests were analysed for their psychometric characteristics (developmental trend, reliability, external validity, factorial dimension) and criteria of reference were given for each level. Results evidenced that the tests' reliability and external validity were satisfactory and that their use was particularly fruitful within educational interventions for improving reading comprehension.

– 112 –

LIRALEC: A Web-based resource for the assessment and training of reading comprehension skills

Jean-Francois Rouet, University of Poitiers - CNRS, France

Antonine Goumi, University of Valencia, France

Audrey Maniez, University of Poitiers - CNRS, France

Anthony Raud, University of Poitiers - CNRS, France

We discuss the skills that underlie reading literacy, considered from a functional perspective. We introduce LIRALEC, a set of Web-based resources aimed at assessing and training reading comprehension skills in middle school students. LIRALEC provides facilities for the creation and the performance of text-based tasks according to various scenarios. Exercises can be assembled in series tailored to the need of students. A full size field study is presently being conducted over a school year with the participation of about 200 students. The purposes are (a) to find out if consistent patterns of difficulties emerge over a large number of trials and (b) if individualized training series result in significant improvement of literacy skills by the end of the school year.

Processing of diagnostic text information as a component of reading comprehension ability

Stephan Dutke, University of Kaiserslautern, Germany

Christiane Baadte, University of Kaiserslautern, Germany

Andrea Hohnel, Technical University of Dresden, Germany

Ulrich von Hecker, Cardiff University, United Kingdom

Mike Rinck, University of Maastricht, Netherlands

The mental model of the situation described in a text gradually unfolds as incoming text information augments the propositional text base and activates background knowledge. Some times the model is ambiguous. An efficient comprehension strategy is to search the text input preferably for information disambiguating the situation model (diagnostic text information) as opposed to information that simply adds to the initial situation model without providing any opportunity to check its global consistency (nondiagnostic text information). Three experiments investigate how readers recognize and process diagnostic text information and to what extent these processes were related to working memory capacity. The experimental procedure required eight narrative texts to be read. Each text consisted of about 50 sentences and described a social situation involving six individuals. This situation was ambiguous with regard to whether this group of individuals consisted of two or three

– 113 –

cliques. Later in the text, a critical pair relation was presented that was compatible with either (a) the two-clique situation, (b) the three-clique situation, or (c) both situations (nondiagnostic relation). Analyses of subjects' reading times demonstrated that sentences presenting diagnostic relations were read more slowly than sentences presenting nondiagnostic relations (Experiments 1-3), irrespective of whether the diagnostic information confirmed the initial situation model or required the initial model to revised (Experiments 1 and 2). After reading each text, the participants indicated the number of cliques (Experiments 1 and 2) or evaluated six different clique structures (Experiment 3). The results showed that the situation model was updated according to the diagnostic information found in the text. Using diagnostic text information in updating the initial model seemed to be related to working memory resources. The recognition and processing of diagnostic text information are discussed as a component of reading comprehension ability as well as of new methods for its assessment.

Behavioural on-line measurements to assess reading comprehension skills and strategies: From research to practice

Eduardo Vidal-Abarca, University of Valencia, Spain

Tomas Martinez, University of Valencia, Spain

Ramiro Gilabert, University of Valencia, Spain

Pilar Selles, University of Valencia, Spain

Behavioural on-line measurements record data of reading and rereading times while reading the text, and also on the sequence of reading, at the same time that they preserve the natural course of reading. Those data give an account of the reader's strategies. Researchers have extensively used computerized tools that record behavioural on-line data to analyze the reader's skills and strategies in a great variety of comprehension tasks. A computerized test to assess school Children's skills and strategies called TEC-e (Test on-line de Estrategias de Comprensi?n, On-line Test of Comprehension Strategies) based on a research tool that records behavioural online data called Read&Answer is presented. TEC-e has been used to capture reading comprehension strategies of 5th, 7th and 9th graders when they read a text and answer questions. TEC-e automatically generates a report which includes: (a) comprehension scores, (b) on-line measures, and (c) pattern of reading comprehension strategies. Results show that TEC-e is useful to assess the student's comprehension skills and strategies.

– 114 –

Analysing think-aloud protocols of beginning readers:a two layered system

Gonny Schellings, Radboud University Nijmegen, Netherlands

Cor Aarnoutse, Radboud University Nijmegen, Netherlands

Jan van Leeuwe, Radboud University Nijmegen, Netherlands

In the present study, we examined the reading behaviors of young readers, while reading an expository text. A total of 24 third graders was administered a thinkaloud task on two occasions. Their protocols were analysed by a coding system that captured two levels of the reading process: the word identification level and the reading comprehension level. Three indices reflecting three different reading behaviors were discerned: reading errors, reproduction, and activities referring to reading strategies. Correlational analyses showed the reading strategy index to be related to reading comprehension as measured by standardized tests. The thinkaloud task constitutes a valuable instrument for examining strategic reading among young readers.

A 22 23 August 2005 14:30 - 16:30 Room B107

Expert Panel

INTERDISCIPLINARY RESEARCH IN EDUCATION: PROBLEMS, PROS-

PECTS, AND PROMISES

Chair: Andreas Demetriou, University of Cyprus, Cyprus

Organiser: Andreas Demetriou, University of Cyprus, Cyprus

Participants: Denis C. Phillips, Stanford University, USA

Monique Boekaerts, Leiden University, Belgium

Erik De Corte, University of Leuven, Belgium

Fritz Oser, University of Fribourg, Switzerland

Roger Saljo, Gothenburg University, Sweden

Stella Vosniadou, University of Athens, Greece

In the recent years research and theorizing crosses the boundaries between fields and disciplines much easier than in the past. This approach is reflected in the emergence of new terms, such as cognitive science, neuroscience, personality science, etc., which indicate the shift of emphasis from a particular field of study to a class of phenomena and problems that are approached from the perspective of all disciplines concerned. This epistemological and methodological shift already started to

– 115 –

shed line on complex phenomena that cannot be understood from the limited perspective of a particular discipline. Education is a vast space where practically every phenomenon is by definition complex, multidimensional, and variable, thereby necessitating a multidisciplinary approach. This round table discussion will focus on interdisciplinary research in education with the aim to highlight its present status, point to the weaknesses that must be removed if will be able to have an educational science as successful as cognitive science or neuroscience, and point to the directions that must be taken en route to this successful educational science. The discussants, coming from different fields, such as philosophy of education, educational psychology, cognitive science, developmental science, will attempt to open the discussion that EARLI will have to foster and cultivate for many years to come, in its will to function as a catalyst for new developments in research, theory, and practice in education.

This expert panel is organized under the auspices of the International Academy of

Education.

B 1 23 August 2005 17:00 - 18:20

Paper Presentation

Assessment and Evaluation

STUDENT ASSESSMENT

Chair: Leonidas Kyriakides, University of Cyprus, Cyprus

Room A109

Developing and validating a questionnaire on Student Perceptions of Assessment

Bruce Waldrip, University of Southern Queensland, Australia

Darrell Fisher, Curtin University, Australia

Jeff Dorman, Australian Catholic University, Australia

The overall aim of this study was to investigate relationships among students' perceptions of their assessment tasks, teacher-student interactions and attitude towards science in middle school science classes. There has been a substantial amount of research into types of assessment but very little research into students' perceptions of assessment. By including students in the teaching - testing - grading cycle, the validity of the assessment processes can be enhanced and invalid assessment in-

– 116 –

struments that result in very high failure rates can be avoided. This phase of the project involved the development and validation of an instrument, the Students'

Perceptions of Assessment (SPA) questionnaire to assess students' perceptions of their assessment tasks in science classes. After examination of relevant literature, the following scales were adopted: Congruence with Planned Learning, Diverse

Methods, Authenticity, Student Consultation, Transparency, and Accommodation of Student Diversity. Factor analysis of 94 middle school classrooms confirmed the presence of six scales. The internal consistency/reliability and scale item mean of each of the SPA scales provided alpha coefficients that ranged from 0.50 to 0.84 confirming that each SPA scale has acceptable reliability, especially for scales containing a relatively small number of items. Correlational analysis shows distinct but somewhat overlapping scales. The higher year levels perceived less Student Diversity, and Diverse Methods. The student interviews therefore examined the veracity of the students' perceptions. When we interviewed the students about the extent of degree of Congruence with Planned Learning, the students perceived that it was important that the assessment tasks align with the goals, objectives and activities of the learning program. Overall, this study has shown that identifying the components of assessment using students' perceptions of their assessment is worthwhile and that further study employing the SPA would be valuable.

Development and validation of a questionnaire for measuring students’ capability for effective learning

Angela Ho, Hong Kong Polytechnic University, Hong Kong

Kenneth S.H. Tam, Hong Kong Polytechnic University, Hong Kong

Kam Por Kwan, Hong Kong Polytechnic University, Hong Kong

This paper reports the development and validation of the Capability for Effective

Learning Questionnaire (CELQ). The CELQ uses an empirically emerged model of learning to learn as its framework and attempts to measure students' capability for effective learning in the following domains: (1) contextualised goal-setting, (2) cognitive strategies, (3) metacognitive strategies, and (4) collaborative learning.

Questionnaire items were derived from the interview data from an earlier research on students' learning-to-learn needs (Ho, Chan, Sun, & Yan, 2004) and with reference to other relevant instruments from the literature. The questionnaire was piloted with a sample of over 700 students from 16 academic departments from a university in Hong Kong. Data was factor analysed and reliability for each scale was estimated. Items with low structural coefficients and poor internal consistency were either removed or rewritten. Two iterations of piloting were carried out and

– 117 –

81 items remained in the final questionnaire. The resulting questionnaire obtained a high reliability for all four scales with alpha values between 0.84 and 0.93. Concurrent validity of the questionnaire was established by correlating its scores with that of a theoretically comparable validated instrument (Study Process Questionnaire) (Biggs, Kember, & Leung, 2001) and through a known group comparison procedure. It was concluded that the CELQ possesses satisfactory psychometric properties for a reliable and valid measurement of students' capability for effective learning, but further work needs to be done to enhance the sub-scales to provide more detailed diagnostic information on student learning capabilities. The theoretical underpinnings of the CELQ and its applications are discussed and recommendations for further work proposed.

Student assessment: Analysis of teachers' understanding and interpretation of what is said and demonstrated by students

Tali Wallach, Weizmann institute of Science, Israel

Ruhama Even, Weizmann institute of Science, Israel

This study aims to examine the structure and content of teachers' understanding and interpretation of students' talk and action while engaged in mathematical problem solving. Participants are 12 elementary school teachers in an in-service workshop on mathematics teaching. Data collected include the workshop leader's journal, written work prepared by the teachers, and video-tapes of the following: all workshop sessions, the pairs of students' problem-solving sessions, individual interviews with each teacher that centered on episodes that the teacher chose from the videotape of her students, two focus-group interviews. Data analysis is based on the

Phenomenological Hermeneutics method and the Grounded Theory method. Structural analysis of the individual interviews data centers on two dimensions: types of interpretation and focus of interpretation. A two dimensional analysis of the teachers' interviews reveals four structural profiles of interpretation, defined as follows:

(1) local-cognitive interpretation - focus on the event (the students' work) with little connections to external contexts, together with an emphasis on cognitive aspects of the students' work, (2) local-varied - focus on the event with reference to different aspects of the students' work, (3) connected-cognitive - using external contexts to explain the event or inferring new insights, together with an emphasis on cognitive aspects of the students' work, and (4) connected-varied - using external contexts with reference to the different aspects. Content analysis of the data suggests two categories of characteristics related to teacher hearing (i.e., understanding students' talk and action). One category is defined as compatibility of teacher hearing with

– 118 –

what the students are saying/doing. In this category the following characteristics were identified: over-hearing, under-hearing, no-hearing, and compatible-hearing.

The second category is defined as the complexity of teacher hearing, and includes the following characteristics: flexible-hearing, firmed-hearing, complex-hearing.

Students' assessment preferences in relation to their study-results in new learning environments.

Gerard van de Watering, University of Maastricht, Netherlands

David Gijbels, University of Antwerp, Belgium

Janine van der Rijt, University of Maastricht, Netherlands

Filip Dochy, University of Leuven, Belgium

In the present study students' assessment preferences are topic of research. Two questions are central in this study: (1) Which assessment and item format do students prefer and which cognitive processes students prefer to be assessed within a new learning environment? (2) In which way are assessment preferences related to students' study results? Students' assessment preferences were measured by means of the Assessment Preference Inventory. Students' study-results were measured by means of their final exam, which consisted of both multiple-choice and open-ended questions. The new learning environment used in this study concerned problembased learning (PBL). 208 first-year Law students following a PBL-course on the topic of public law participated in the study. Results were analysed by means of descriptive statistics for the measures used in the present study and analysis of variance were conducted to probe into the relationships between students' assessment preferences and their study results. Results indicated that students preferred written tests wherein they are allowed to use supporting materials like notes and books.

Oral tests, computerised tests and portfolios were not preferred by the students as a way to assess their knowledge and skills in a problem-based learning course. Students' assessment preferences were related to their final (written) exam results: On the one hand students who preferred written assessments obtained lower marks for the total assessment than students who did not prefer written assessments at all or were neutral towards written assessments. On the other hand students who did not prefer alternative assessments like portfolios and papers or projects obtained higher scores on the part of the (traditional) assessment with multiple choice questions.

Several explanations are discussed in the paper.

– 119 –

B 2 23 August 2005 17:00 - 18:20 Room A108

Paper Presentation

Assessment methods

ENHANCING TEACHING AND LEARNING

Chair: Constantintos Papanastasiou, University of Cyprus, Cyprus

Peer assessment as a tool for learning: a meta-analysis

Nanine A.E. Van Gennip, Leiden University, Netherlands

Mien Segers, Leiden University, Netherlands

Harm Tillema, Leiden University, Netherlands

Changes in the current view of learning have led to changes in the nature of assessment. The implementation of new modes of assessment in order to increase learning outcomes in new and dynamic learning environments appears more and more.

These changes in assessment are moving from testing to multiple and integrated assessment. Peer assessment is an example of such a new mode of assessment.

This meta-analysis examines the relation between peer assessment and learning, or the development of professional competences. This relation is of great importance, because of the learning benefits assessment can offer. Additionally, the conditions that affect this relation, such as social factors, the perceived validity of peer assessment and prior experiences with peer assessment, are examined. The results of this meta-analysis showed that the studies examining peer assessment and learning can be categorized in at least three sub-categories: (1) qualitative versus quantitative studies; (2) experimental versus non-experimental studies; and (3) subjective and objective measures of learning. Especially the relation between peer assessment and subjective learning appeared to be a strong one. The sub-categories and the conditions that appeared to affect the relation between peer assessment and learning will be discussed.

Design-based multi-level assessment for enhancing discourse

Gita Taasoobshirazi, University of Georgia, United States

Dan Hickey, University of Georgia, United States

Dionne Cross, University of Georgia, United States

Accountability-oriented reforms have highligted enduring tensions for classroom

– 120 –

instruction in all domains, including mathematics. In many countries, schools are under pressure to deliver continually increased scores on criterion-referenced tests.

Particularly in the US, these gains are expected for all students, but are also not expected to come at the expense of other valued outcomes that are not assessed on those particular tests (e.g., classroom discourse, deeper understanding, performance on other tests, graduation rates, etc). This paper describes an approach that extends existing research on assessment to attain these seemingly conflicting goals for educational reform. The approach features at least three levels of assessment (i.e., close, proximal, distal), each using an increasingly formal representation of knowledge (ìcultural, ìcognitive, and ìbehavioral,). Formative and summative functions within and across each level are iteratively refined across three increasingly formal cycles of design research (implementation, experimentation, and evaluation). This framework extends existing research on assessment and testing in several ways.

At the close level, innovative formative feedback rubrics for each quiz item are used along with ìfeedback conversations that build on students shared knowledge of specific curricular activities. These conversations provide feedback to teachers that is ideal for informally guiding student discouse and understanding. At the proximal level, formal classroom assessments provide evidence of the impact of the close level activity on individual understanding, and provide teachers with feedback useful for remediating specific students or specific topics, and formally refining the curriculum. The distal-level assessments provide valid evidence of the impact of these refinments, and can be adminstered to comparison groups. Results from a pilot project involving seven hours of third-grade fractions curriculum are presented. A three-year study recently funded by the US National Science Foundation is introduced.

Enhancing teaching and learning by using an interactive assessment file for professional development

Christian Fantoli, HEP-VD University of Education, Switzerland

George Hoefflin, HEP-VD University of Education, Switzerland

According to the Bologna declaration (BA, MA process), implementation of systematic summative assessment in the Swiss Universities of Education (HEP) requires new methods for the evaluation of teaching. This paper specifically investigates further steps of implementation of a portfolio for the assessment of teaching provided by the Special Education section of the University of Education of the canton of Vaud in Switzerland (HEP-VD). This portfolio has developed into an ìinteractive assessment file that shows important effects on the professional develop-

– 121 –

ment of trainees working in the field of special needs (Hoefflin & Allal, 2004). Further exploratory research investigates whether teaching is qualitatively enhanced by increased interactivity between trainers and trainees. In Switzerland, special needs training is provided to former ordinary teachers (BA) who start teaching in special needs school classes and simultaneously follow a specific academic curriculum of three years. After analysing the advantages and disadvantages of using formative assessment to evaluate professional development in the field of special needs, this paper will consider qualitative improvement of teaching ìdidactic and disciplinary skills (French and mathematics).

Critical thinking dispositions inventory as the tool for evaluating instructional alignment

Patrick Lai, Educational Development Centre, Hong Kong

Alex Wong, The Hong Kong Polytechnic University, Hong Kong

Priscilla Chan, The Hong Kong Polytechnic University, Hong Kong

Constructive alignment has been the guiding principle on instructional design. Basically it calls for an integration between assessment, learning objectives and the learning-teaching activities. For the elicitation of effective learning, the assessment methods so employed have to address the learning outcome. Taking into consideration of the importance of this principle in curriculum design and assessment, some means has to be developed to evaluate whether a certain program and subject has fulfilled the alignment principle. This study employs the presage-process-product model as the theoretical framework to evaluate whether the assessment methods employed by a university subject has addressed elements of the critical thinking.

The Critical Thinking Disposition Inventory (CCTDI) was administered to students at the start of the subject. Students were classified into positive, ambivalent, or negative categories in the different sub-scales of the inventory. Assessment scores of students in the different sub-groups were compared to find out whether there was any significant difference in group means. Two focus group interviews were conducted to investigate students' perception of the learning, in particular the assessment environment. Results showed that the positive sub-group of the open-mindedness subscale of CCTDI scored significantly higher than their negative counterpart in the project performance. Interviews also indicated that students agreed that projects served as means to help them learn and interact. As a conclusion, the results indicated that the project assessment method has facilitated students to engage in an open-minded thinking mode. This study is the first of its kind to use the CCTDI as a tool to evaluate whether instructional alignment has occurred. It might well serve

– 122 –

as a springboard for similar studies to be conducted so as to further develop quality assurance procedures for curriculum development.

B 3 23 August 2005 17:00 - 18:20 Room A110

Paper Presentation

Language Education

APPROACHES TO ENHANCING LANGUAGE DEVELOPMENT

Chair: Jonas Emanuelsson, Gothenburg university, Sweden

Validity of indicators to study the alignment equivalence of two Language Arts curricula

Dany Laveault, University of Ottawa, Canada

Carol Miles, Carleton University, Canada

The purpose of this study is to compare Francophone and Anglophone students of

Ontario on how they use rubrics to assess writing. Equal opportunities to understand rubrics are crucial in making sure that each respective Language Arts curriculum of Ontario, the French as well as the English one, are aligned properly (La Marca,

2001). If it can be shown that, despite separate curricula, the ability of Francophone and Anglophone students to develop good writing skills is not influenced by different ways of interpreting rubrics, we may then assume some degree of ìalignment equivalence between both curricula. Three indicators were developed to measure this construct: (1) the discrepancy between a student's own rating and the province's established achievement level of a series of writing exemplars, (2) the severity and

(3) the student's confidence level. Our hypothesis is that these variables may be used to compare the two Arts programs, in terms of their respective alignment and of in terms of their impact on students' achievement. Results show that the curricula for the Francophone and Anglophone students may be considered as aligned in approximately equivalent ways. A multiple regression of the three indicators on the school marks in writing shows that the most significant predictor is the discrepancy score. Multiple Rs for each group are statistically significant and both equations of prediction include the same variables with comparable weights. This study shows that the indicators used for this research, specially the D score, may be used as early indicators of differential test functioning. It may help bring up the attention of teachers and school boards to differential understanding of rubrics and to their

– 123 –

potential impact on students' achievement.

Modes of talk and their relation to interaction patterns in language lessons in the seventh grade

Hayuta Yinon, Haifa University, Israel

Lily Orland-Barak, Haifa University, Israel

Classroom discourse is a unique kind of discourse, which is shaped through interaction patterns, namely, through reciprocal communication systems amomgst participants. These interaction patterns, and mainly the IRE/F sequence, shape rule systems, which, in turn, define and regulate classroom activity, and determine how to talk in the classroom. Despite this understanding, research has paid little attention to the relations between interaction patterns and modes of talk This study aims at filling this gap in the specific and unique context of frontal language lessons in one seventh grade classroom in northern Israel, by exploring three questions:

1. What characterize existing modes of talk in the classroom?

2. What are the rule systems, that define and regulate modes of talk in the classroom, and how are they shaped through interaction patterns?

3. What interaction patterns contribute / impede to the preservation of rule systems, which define modes of talk in the classroom?

The research paradigm of the study is qualitative, naturalist and interpretive, and it integrates ethnography and classroom discourse analysis. The data was collected through three sources: interviews, observations of lessons and collection of school documents. Data analysis followed by grounded theory procedures.

The findings of this study lead to several understandings. First, the study extends the potential of the use of the IRE/F sequence to the exmination of modes of talk in the classroom. Second, the study points to the existence of more flexible discourse patterns in frontal lessons in the Israeli classroom, challenging the common belief that pupils seldom take an active part in the lesson. Third, the study reveals that investing effort in determining modes of talk in the classroom, in creating rules to sustain it and in "talking about it" with pupils, does not necessarily lead to the accomplishment of classroom management.

– 124 –

Following phonemic awareness traces in Spanish

Francisca Serrano Chica, Univesidad de Granada, Spain

Sylvia Defior-Citoler, Univesidad de Granada, Spain

Carmen Gonzalez Trujillo, Univesidad de Granada, Spain

This study aims to evaluate the development of phonemic awareness skills, the deepest level of Phonological awareness (PA), before and after reading instruction.

Participants were 105 Spanish children, followed in a three year longitudinal study

(from preschool age until first year of primary school). Children were evaluated twice in each academic year (November and May), using a phonemic awareness task, where the linguistic complexity of the items was manipulated (number of shared phonetic features and initial consonant cluster versus single phoneme). Results show emergent phonemic awareness skills at an early age of development.

However, in the first year of primary school these abilities improve considerably.

The linguistic complexity of the items influence children performance. Implications for the predictive value of phonemic awareness skills are considered.

A somatically enhanced approach to creating a language-learning environment

Felicia Zhang, University of Canberra, Australia

Research has shown that traditional methods of teaching pronunciation in foreign language learning have not been as effective as might be desired. This is largely because many of the teaching methodologies are designed along the idea of working from a database of errors that learner typically make and then trying to find solutions to correct those errors. The teaching pedagogy described in this paper starts from the opposite end and deals with training students' perceptual mechanisms so as to enable them to have better pronunciation in an L2 from the beginning. This paper describes a significant innovation in language teaching pedagogy. Language should be learnt not through a process of manufacture but through discovery. In effecting this change, curriculum and learning environment design should consider learner needs and desires as primary without sacrificing the objectives of producing proficient speakers of the target language. The primary technology that contributes to the learning process in this environment is the 'body'. Together with the body,

CD-ROMs, a speech processing tool which enables voice comparison mechanism for audio files, and the internet constitute 'machines' in the learning environment

(which are waiting to be activated by learners). Finally, the paper briefly reports on an experiment for teaching Mandarin to English speaking learners using this unique teaching methodology in an Australian university. Experimental results demon-

– 125 –

strate that Mandarin Chinese L2 learners, after a mere 65 face-to-face contact hours in this environment, not only achieve a faster rate of acquisition of the segmental system of Mandarin Chinese, they also exhibit a higher level of motivation, initiate their own learning activities and organize their own process of learning. They also develop a positive attitude towards Chinese culture and become able to cope in a wide range of Chinese speaking contexts.

B 4 23 August 2005 17:00 - 18:20 Room A111

Paper Presentation

Science Education

CLASSROOM STRATEGIES

Chair: Theodora Kyratsi, University of Cyprus, Cyprus

Group work and conceptual growth in science: Taking account of post-group effects

Christine Howe, University of Strathclyde, United Kingdom

Research has repeatedly shown that conceptual understanding in science can be promoted by group work between learners. However, the research is inconclusive over the mechanisms by which group work impacts upon individual understanding.

One popular account sees learners as constructing collective insights during group work that are superior to their starting points, and then assimilating these insights into their individual knowledge. However, assimilation cannot account for conceptual growth that is stimulated by group work but: a) occurs despite collective views that are inferior to individual starting points and that are unrelated to the extent of progress; b) requires a post-group interval of several weeks to become apparent. Delayed conceptual growth of this kind has been reported on many occasions, and the project to be summarised in this paper explains why it occurs. The project involved studies with primary school children aged nine to twelve, and addressed mastery, subsequent to group work, of the factors relevant to floating and sinking.

The group task used in the studies required children to predict which, from sets of small objects, would float and which would sink, test their predictions by immersion in a tank of water, and interpret outcomes. Mastery of factors was established by paper-and-pencil tests that were administered to whole classes as pre-tests prior to the group tasks and as post-tests eight weeks afterwards. The results show that

– 126 –

delayed conceptual growth occurs when group work 'primes' children to make productive use of relevant events that are experienced during the post-group interval.

In addition to their theoretical implications, these results have practical implications for how teachers interpret group interaction, and for how they support its consequences over the succeeding weeks. The latter will be discussed in detail.

Examining the Reliability of Using Science Notebook as Assessment Tools

Min Li, College of Education, University of Washington, United States

Maria Araceli Ruiz-Primo, Stanford University, United States

Student notebooks have been proposed and investigated as an instructional and unobtrusive assessment tool (e.g., Aschbacher & Alonzo, 2004; Harmelink, 1998;

Ruiz-Primo & Li, 2004; Shepardson & Britsch, 1997). Furthermore, policymakers and researchers have started considering the inclusion of different sources of assessment information into accountability system, such as, curriculum-embedded assessments, class projects, notebooks (e.g., Shavelson, 1999; Wilson, 2004). To evaluate the technical soundness of using notebooks as assessment tools for both instruction and accountability, this paper focuses on using Generalizability (G) theory to examine how notebooks can be analyzed consistently and efficiently for supporting credible interpretations of student performance. G theory allows to estimate the amount of score variance due to the inconsistency between notebook entries or raters and to determine scoring procedures of optimizing the number of raters or entries needed to minimize measurement error (Shavelson & Webb, 1991). A random stratified sample of thirty-six notebooks selected from twelve science classrooms was independently scored by four well-trained raters. All the raters were teachers with several years of experience in teaching the curriculum and using notebooks in their classrooms. For each notebook, four out of the 12 investigations in physical science required by the curriculum were scored. Student x entry x rater G studies were performed to examine the reliability of the notebook scoring. The analyses will address the following questions: (1) Can well-trained raters reliably score a sample of students' notebooks entries? (2) Is there a difference in notebook scores between two types of scorers, own teachers versus outside scorers? And (3) how many entries and raters are necessary to obtain reliable notebook scores for student performance? Preliminary analyses found that scoring several entries could provide reliable interpretation of student performance though reliability coefficients varied across the investigation sub-aspects and across the dimensions focused by the scoring procedure.

– 127 –

The impact of integrated illustrated instructions on cognitive load and understanding within secondary school science practical work

Carolyn Barr, University of Auckland, New Zealand

Richard Hamilton, University of Auckland, New Zealand

This study looks primarily at improving the understanding and learning from practical work actually performed by students in secondary schools by modifying the presentation of the instructions students are required to follow to complete these practical tasks which involve integrating information from two different mediums

(written instructions and electrical equipment). The instructions were modified using guidelines from Cognitive Load Theory and Theory of Multimedia Learning.

This study also looked at the effect of gender on understanding and learning in a practical context in science. 96 junior secondary school students who were unfamiliar with content knowledge and the equipment commonly used in practical work on electricity took part. One group was given modified instructions containing integrated text and illustrations and the other group was given conventional instructions containing text only. A series of MANOVAs found that modified instructions produced significantly higher levels of performance on task, lower time to completion, lower perceived cognitive load and task difficulty and higher post test scores than conventional instructions. Significant gender effects were found for the time to completion and on all performance measures. No significant treatment X gender interactions were found on any measure. This data suggests that for practical work in science, when learners are inexperienced and the information is complex that physically integrating mutually referring sources of information reduces split attention and hence cognitive load, and therefore makes instructions easier to understand.

The treatment effects were also robust, i.e., not moderated by gender. These results indicate that subject's understanding was facilitated to a higher degree in the modified instructional treatment and are consistent with the notion that the annotated, integrated illustrations made the instructions easier to understand and were successful in eliminating any split attention effect. In addition, the modified instructional materials facilitated greater learning as indicated by the post test scores.

Expertise in the Domain of Science Discuss Processes of Change in Their Thinking

Hava Greensfeld, Michlalah Jerusalem College, Israel

Ilana Elkad-Lehman, Levinsky College of Education, Israel

This study is a part of a wide-scale research which examines the reasons underlying

– 128 –

changes in the way teacher educators in specific domain (science or literature) perceive their thinking about their discipline, their manner of instruction, or both. This presentation will focus on the processes of change in the thinking of science teacher educators, which were examined against recent changes in the education system, the development processes of teachers and of teachers in academia, and processes of change in science and in science education. We used qualitative methods to examine the extent of teacher educators' awareness of the processes of change in their thinking. The main research tool was a semi-structured interview that stimulates reflective thinking. Seven teacher educators were selected, and the narratives of some of them will be presented. The data was processed using qualitative research tools and methodologies from the fields of discourse study and literary research.

Our study reveals a complex, multi-varied picture regarding the factors involved in the processes of change in thinking. We found that motivation and metacognitive abilities are an indispensable condition for change in thinking processes. These abilities propel change in the domain-specific knowledge of science and in teaching, while cognitive conflict, which had been presented as a central cause of change, provides only a partial explanation for changes in thinking as these related to science teacher educators. More over, from the perspective of examining changes in thinking through introspection, our study reopens the question of the fruitfulness of

"conceptual change" as a term to describe changes in learning processes of teacher educators. The research findings may contribute new knowledge and insights regarding the thinking process of science teacher educators when relating to their domain-specific knowledge.

– 129 –

B 5 23 August 2005 17:00 - 18:20 Room A112

Paper Presentation

Early Childhood Education

LANGUAGE AND CREATIVE THINKING

Chair: Chrystalla Papademetri, University of Cyprus, Cyprus

The creative thinking of young children in a community of enquiry

Hanneke Jones, Newcastle University, United Kingdom

The aim of this study was to investigate to what extent the Community of Enquiry could stimulate and support creative thinking in children aged 5-7. The main research questions for this research were: 1. How could aspects of creative thinking, expressed by children during Community of Enquiry sessions, be identified and investigated? 2. How did that creative thinking develop in the course of a school year? 3. How did that development in creative thinking compare to the Children's creative thinking as expressed in two Torrance Tests of Creative Thinking? Methodologically, analysis was carried out of both quantitative data: The Torrance Tests of Creative Thinking, carried out in October and July. 17 transcribed Community of Enquiry sessions. ïEach of the responses was categorised as either reproductive, tangential, original or reasoning, and Qualitative data: ïPersonal observations; Children's opinions; General classroom observations; Microgenetic analysis of several of the session transcripts. Main findings: On average, a higher percentage of the responses was productive in the second half of the year, than in the first half of the year. The response category indicating the highest level of creative thought (reasoning) had risen the most. A high correlation was found between disagreement and reasoning. Patterns indicative of interactive creativity were found. Relevance: The research is relevant to two Conference Domains: Learning and Cognitive Science:

Knowledge creation and creativity are central to this study, which indicates how creative thinking in young children can be conceptualised and assessed. Teaching and Instructional Design: The Community of Enquiry is an effective pedagogical framework in which many children savour the opportunity to display and develop their creative thinking.

– 130 –

Adults and pre-school children talking science in three learning environments

Monica Strang, Gothenburg University, Sweden

Lisbeth Aberg-Bengtsson, Gothenburg University, Sweden

This paper presents some tentative results from a study of ten 5-year-old pre-school

Children's talking about scientific phenomena with adults in different learning environments (a science center, their pre-school, and a guided conversation with a researcher). Taking a socio-cultural perspective, the aims of the study were to illuminate and describe strategies used by adults when talking with children about science in the different settings and to point out communicative patterns between the interactors in the activities. Moreover, the enhancing (if any) of particular components of the scientific content and the use of scaffolding were looked for. At the time of the study the children had taken part in a theme-work about water at their pre-school. They visited a science center at the end of the theme-work period. The collected data include audio and video recordings from (a) the visit to the science center, (b) circle time at the pre-school, and (c) the guided conversation with the researcher. The analysis shows that the adults in the three different settings used different strategies when talking about science with the children and that these strategies were important as to how the communicative patterns were shaped. In addition,

Children's familiarity with the scientific phenomena, as well as their ability to use it along with the new information, were closely related to their level of communicated knowing. The analysis also indicate the importance of the adults' scaffolding the

Children's construing of their ìstory. In the present paper, the results will be related to theories within the socio-cultural framework and to some previous research.

Moreover, some educational implications will be further discussed in the paper.

Kindergarten children producing humor: A combination of artistic and conceptual incongruities with language use in the process

Eleni Loizou, University of Cyprus, Cyprus

This study aimed in investigating the ability kindergarten children to produce humor and the role of language in this process. Semi-sructured interviews and the actual drawings of ninety kindergarten children were the main sources of the data.

The children were asked to draw a funny picture and then describe what made their picture funny. This study was developed on the basis that humor involves an incongruous event and the data was examined against McGhee`s stages of humor development. Findings of this study suggest that children are capable of producing one or more incongruities when drawing a funny picture. This is in coherence with

– 131 –

McGhee`s stage 3 of humor development; conceptual change. More specifically the incongruities used by the children in their drawings were categorized into 1. Artistic incongruity (based on color or features), 2. Conceptual Incongruity (animism and combinations) and 3. Humorous symbols (clown, gestures, laughter, joke). In reference to the role of language in the process of producing a funny picture, the findings show that a. children use language before, during or after drawing their picture; b. language helps them describe their thoughts and drawings in terms of humor; and that c. they use either a list of information, or a story to describe their funny picture.

This paper asserts the importance of considering multiple perspectives(cognitive, linguistic and artistic) in examining children`s knowledge and in providing for their learning.

Supporting oral language development in nursery schools: challenges and choices

Julie Dockrell, Institute of Education, University of London, United Kingdom

Morag Stuart, Institute of Education, University of London, United Kingdom

Diane King, Institute of Education, University of London, United Kingdom

There is a growing concern about the limited oral language skills of some children on entry to school. The current project addressed this problem by devising an oral language intervention based on current models of language acquisition. The intervention, Talking Time, focussed on three core language skills: vocabulary, inferencing and oral narrative. Both oral comprehension and expression were targeted.

Progress for the children in the Talking Time intervention (N= 53) was contrasted with a comparison group who received the same amount of exposure in a Story reading intervention (N=41) and a control group who received typical good nursery practice (N=48). The ‘intervention’ took place in small groups, twice a week in the nursery setting and was delivered by nursery staff. Both the Talking Time and

Story Time intervention continued over a 6-month time frame. Adherence to the intervention programme was monitored. Language skills were measured prior to the start of the interventions and on completion of the intervention period. All groups improved over the six-month period. Differential improvement patterns were evident for some targeted language skills. ‘Talking Time’ enhanced comprehension, naming vocabulary and sentence repetition. Gains for a general verbal measure were also evident for the ‘story reading’ intervention in comparison to the control.

Expressive narrative language was not differentially improved and the Children's performance in this area remained depressed. The results are discussed in relation to current theories of language development and the ways in which such models can support language programmes that are aimed to address Children's needs and

– 132 –

progress. Issues of viability, fidelity and training are explored in relation to preschool intervention.

B 6 23 August 2005 17:00 - 18:20 Room E002

Paper Presentation

Reasoning

REASONING AND ARGUMENTATION

Chair: Gunilla Petersson, Stockholm university, Sweden

Immediacy and intuitive reasoning in probability judgment

Tali Brecher, Tel-Aviv University, Israel

Reuven Babai, Tel-Aviv University, Israel

Ruth Stavy, Tel-Aviv University, Israel

Stavy and Tirosh (2000) have observed that students react in similar ways to a wide variety of conceptually non-related tasks that differ with regard either to their content area and/or to the type of reasoning required, but share some common, external features. Based on these observations, they proposed the Intuitive Rules Theory, which explains and predicts students' responses to mathematics and science tasks.

Many responses, often described as ìalternative conceptions, may be interpreted as evolving from a number of intuitive rules, which are activated by specific external task features. The motivation behind this study was to empirically address the immediacy characteristics of intuitive responses in the context of science and mathematics. For this purpose, we employed the framework of the Intuitive Rules

Theory and compared the reaction times of two types of responses: those that are regarded as intuitive and those that are viewed as counter-intuitive. We found that the reaction times of the former were, indeed, shorter than the latter. Accordingly we are currently studying if specific teaching methods, in accordance with the Intuitive

Rules Theory, will promote students' ability to overcome their intuitive mistakes.

– 133 –

Thinking, arguing, and counter-arguing: The effects of issue relevance and minority influence on the choice of argumentation strategies

Carlo Tomasetto, University of Bologna - Faculty of Psychology, Italy

Francesca Romana Alparone, University of Chieti - Dpt of Bio-medical Sciences,

Italy

Angelica Mucchi-Faina, University of Perugia - Dpt of Institutions & Soc, Italy

The relationship between argumentation and reasoning has received much attention in educational psychology, as argumentation is considered the tool that supports any cognitive activity. Nevertheless, being confronted with conflicting points of view does not guarantee a cognitive progress, since different argumentation strategies can reflect more or less accurate cognitive elaboration. In that, we make a parallel between argumentation and hypothesis testing tasks, where confirmatory biases, anchoring effects, etc., are examples of a general focussing effect (i.e., a tendency to consider only the elements already present in one's cognitive field and to ignore any divergent information), while falsification of alternatives witnesses a decentering tendency that ensures the use/acquisition of more advanced cognitive skills. We propose that the same focussing and decentering tendencies can be retrieved also in the field of argumentation. In fact, when arguing about an issue, speakers can hold their position either a) enforcing some statements they already agree with, or b) trying to confute alternative standpoints. Developing this parallel, we also predict that socio-cognitive conditions that are proved to foster decentering in hypothesis testing, will favour the appearance of counter-arguing strategies. In particular, we hypothesize that high personal relevance, and confrontation with a minority source of influence, should enhance the use of counter-arguments, whereas low relevance of the issue, and exposure to majority influence, should favour the enforcement of already shared arguments. In a 2 (relevance: high vs. low) x 2 (source of influence: majority vs. minority) experimental design, 114 university students were asked to develop three arguments in order to express their point of view about the introduction of a final comprehensive exam, choosing between confirmatory and disconfirmatory arguments. Results confirm the predicted source*relevance interaction. Implication for the design of argumentation activities in education will be discussed.

– 134 –

Individual differences in young children's proportional reasoning problem-solving and explanations

Fiona Reynolds, University of Melbourne, Australia

Robert Reeve, University of Melbourne, Australia

Proportional reasoning involves understanding relations among quantities and is central to understanding rational number, and success in high school mathematics.

Many researchers have used aggregate analysis (usually age-based) to characterise proportional reasoning development. However, aggregate analysis may not be the most appropriate means of describing proportional reasoning development because of substantial within-age variability in understanding. It may be more appropriate to explore the significance of such variability by using methods that allow identification of different patterns of understanding. In this study we examined proportional reasoning understanding (using Noelting's Orange Juice task) to address two questions: Can we identify distinct profiles of proportional reasoning understanding on the basis of (1) variations in problem-solving performance, and (2) explanations of another person's problem-solving? If we can identify such profiles, what, if any, relationship exists between them? Ninety-five 6- to 9-year olds participated in a pretest-intervention-posttest study in which children were required to exhaustively share a quantity among n recipients. The pretest and posttest comprised 18 trials where quantities of juice, or juice and water, were to be distributed among two, four, or three recipients. Children were asked to choose which of four options (one correct, three common errors) showed one person's ìfair share of the distribution. In the intervention, children were told that another child had solved similar problems and were asked whether, and why, that child was right/wrong. Hierarchical cluster analyses identified distinct problem-solving competency profiles, and distinct explanation profiles. The sets of profiles were significantly related. Both sets of profiles were well-ordered, representing different forms of proportional reasoning understanding. The educational implications of this study are that it is possible to diagnose levels of proportional reasoning understanding in terms of both problemsolving competence and explanation ability. Further, the identified profiles indicate different developmental progressions in the acquisition of proportional reasoning understanding.

– 135 –

Investigating speech therapists' clinical reasoning: Analysing think-aloud protocols and integrating multiple-source data

Barbara Howarth, University of Newcastle, United Kingdom

Kirsty Hoben, University of Sheffield, United Kingdom

Julie Morris, University of Newcastle, United Kingdom

Rosemary Varley, University of Sheffield, United Kingdom

John Lee, University of Edinburgh, United Kingdom

Richard Cox, University of Sussex, United Kingdom

We report results from pilot studies of speech and language therapists who used a web-based case-based teaching resource (www.patsy.ac.uk) to diagnose previously unseen patient cases. The aim was to identify sources of impasse in their clinical reasoning. Pairs of therapists collaborated - they were encouraged to `think aloud' and were videotaped. Other data were derived from the PATSy system logs. The time course and sequence in which students administered standardised language tests to the (virtual) patient were determined from the log data. The think aloud discourses and the PATSy log data were triangulated. A coding scheme for the think aloud protocols was also developed. This required, inter alia, making decisions about the grain size of analysis. Together, the two sources of rich data provided useful insights into students' clinical reasoning. The results will inform subsequent stages of the project, specifically: 1. by elucidating the extent to which domainspecific knowledge versus more general reasoning heuristics are involved, and 2. by identifying topics around which structured task-directed discussion (TDD) exercises can be developed. The TDDs will be used in the next project phases as a principled method for eliciting educational dialogues for re-use as vicarious learning resources.

– 136 –

B 7 23 August 2005 17:00 - 18:20

Paper Presentation

EMOTION AND EDUCATION

Chair: Philip Mayring, University og Klagenfurt, Austria

Room E005

Children's emotions

Maria Eracleous, University of Cambridge, United Kingdom

Emotions have been a subject of study from very early on. There is evidence even before Plato that emotions were an issue of reference. The present study aims to examine Children's emotions in the educational context. More specifically, this is case study of a pre-school class in Cyprus which explores Children's emotions. Research Questions: The research questions can be articulated as follows:

ï What are the ways that children express their emotions? ï What are the different situations that evoke Children's emotions. ï What are the ways that children deal with their emotions? A range of research methods was used in the study, for methodological as well as practical reasons. More specifically, the research methods were interviews, drawings, research diaries, observations and video extracts. Data

Analysis: The analysis of the data was qualitative with some quantifiable data. The analysis of the data was multidimensional due to the various sources of information. The first phase contained the data from each individual method whereas the second phase consisted of the data that emerged from all the different methods as a whole. Findings: One important conclusion drawn from the study concerns the complexities and difficulties of undertaking research concerning Children's emotions. The study has also shown the impact of cultural norms upon the perceptions of emotions as well as perceptions of research procedure. As regards the findings concerning Children's emotions the study showed that important themes regarding children and emotions emerged from the data analysis. The study showed that children consider facial expressions as an important source of information regarding emotions. In addition, the children seemed to have mingled various emotions and the situations that evoke those emotions. Last but not least, the study provided a small amount of evidence regarding possible ways of dealing with one's emotions and helping mechanisms.

– 137 –

Children's understanding of emotion: Initial validation of Italian test of emotion comprehension (TEC)

Ottavia Albanese, University of Milano Bicocca, Italy

Carla Antoniotti, University of Milano Bicocca, Italy

Eleonora Farina, University of Milano Bicocca, Italy

Caterina Fiorilli, University of Milano Bicocca, Italy

Ilaria Grazzani Gavazzi, University of Milano Bicocca, Italy

The Emotion Comprehension Test (Pons, Harris, 2000) is an instrument designed to evaluate Children's understanding of emotion between the ages of 3 and 11 years.

More specifically, it allows to examine Children's understanding of the nature of emotions, their causes and the possibility of control. We wish to present preliminary data of Italian validation in relation to a group of subjects within the overall sample. Validity was assessed using a sample of 100 children (6 to 10 years old) recruited in elementary schools located in two big cities from northern and central part of Italy. There were 20 subjects (10 males and 10 females) in each age group

(age range: +/- 3 months). All were normal children, with no learning, language and developmental problems. A cognitive test (MT Test, Cornoldi et al., 1991) was also administered. Statistical analysis revealed a trend in Children's understanding of emotion that supports empirically developmental trajectories described by

Pons, Harris and de Rosnay (2004). A significant correlation was found between the MT test and the TEC. Results will be discussed in terms of their theoretical and educational implications, underlining the possibility to apply specific training in

Children's metacognition of emotion.

Enjoying teaching: Enthusiasm and teaching behaviors in secondary school mathematics teachers

Mareike Kunter, Max Planck Institute of Human Development and Educ, Germany

Yi-Miau Tsai, Max Planck Institute of Human Development and Educ, Germany

Martin Brunner, Max Planck Institute of Human Development and Educ, Germany

Stefan Krauss, Max Planck Institute for Human Development, Germany

This presentation explores teacher enthusiasm and how it relates to teaching behaviors. We distinguish between teachers' enthusiasm for 1) the subject matter of mathematics and 2) teaching mathematics to their class, and examine how both types of enthusiasm relate to teachers' behavior and class context variables. A total

– 138 –

of 288 teachers and their 9th grade classes participated in this study. Teachers' motivations were measured with teacher questionnaires (enthusiasm for mathematics, enthusiasm for teaching a class, general job satisfaction, and self-efficacy beliefs) and student rating of their teacher's enthusiasm. Teacher behaviors were assessed using teacher questionnaires (use of social co-construction, peer tutoring, cognitively stimulating tasks, classroom management) and student questionnaires (ratings of teachers' elaboration in teaching, cognitive stimulation, discipline). Class context variables were represented by the class means of students' achievement, social background, and motivation scores. Validity analyses show that subject matter enthusiasm and teaching enthusiasm can be considered distinguishable concepts.

Furthermore, teachers who were more enthusiastic about teaching showed more innovative and elaborative teaching behavior and experienced fewer disciplinary problems in their class - both self- and student reported. By contrast, enthusiasm for math as a subject correlated only with teachers' self-reported behavior, and not with student observations. As regards context variables, teachers from the higher track showed more enthusiasm for mathematics than teachers from lower track, and teachers in classes reporting higher interest in math were more enthusiastic about teaching. The value of enthusiasm as an additional concept reflecting teacher motivation is discussed in terms of its theoretical and educational implications.

B 8 23 August 2005

Paper Presentation

17:00 - 18:20 Room E003

PHENOMENOGRAPHY AND VARIATION THEORY

Chair: Ference Marton, Gotenborg University, Sweden

To limit the time: How we can use variation theory to improve pupils learning to tell elapsed time

Anna Wernberg, Behavioural science, Sweden

Mona Holmqvist, Behavioural science, Sweden

The aim of this paper is to describe how teachers, by using the variation theory in a lesson, enable students to build new learning patterns. The variation theory focuses on the distinction between an ìenacted object of learning and ìlived object of learning, where the enacted object of learning consists of how the teacher actually teach the object of learning and the lived object of learning taking account of how the pu-

– 139 –

pils understand this object of learning. In this study, the participating teachers had felt that learning to tell the time, and especially elapsed time, was experienced to be a difficult concept for pupils 9-10 years old. Hence the object of learning became elapsed time. Praxis-oriented basic research is a growing field in educational science and teaching research. One method used in praxis-oriented research is learning study which has a focus on an object of learning, not teaching methods. In our learning study, three similar teaching situations were designed and the analysis of the classroom data was on what features of the object of learning varied, varied simultaneously, or remained invariant. The analysis shows which aspects of telling the time were focused, focused simultaneously or left unfocused, made impact on the pupils' learning-outcome, i.e. the lived object of learning. The purpose of this research was to find out what pupils learn in the classroom. The point of departure was that the entanglement of what teachers do in the classroom concerning the object of learning is crucial for the learning outcome. The result from the study shows that when teachers plan a lesson, taking the variation theory into consideration, it has a positive effect on the pupils learning outcome.

Teachers matters!

Mona Holmqvist, Kristianstad College University, Sweden

This study aims to describe the ways teachers' instructions affect learning outcome by the students when learning English as a second language in class four in a Swedish nine-year compulsory school. The learning object is in this study the infinitive verb to be in present form (am, are, is). The theoretical framework is the variationtheory, and the method used is learning study. The results show how the learning outcome did change according to the instructions the teachers gave during the three different lessons. By using the variationtheory, the teachers managed to catch the critical aspects of the learning object, shown by an increased learning outcome in lessons two and three. However, even if the learning outcome did increase, some errors were still made. In this study it is possible to see the connection between these errors and the kind of instruction the students were offered in the lessons.

Even in this part of analyse the variation theory is used. While the connections are strong, it is possible to encounter the teachers a strong importance of the learning possibilities in a classroom. To use a theory about learning, in this study the variationtheory, seems to develop insights about the students' learning which the teachers have not experienced before. To develop such insights are crucial in a powerful teaching situation.

– 140 –

Meanings of and relationship between meanings of commonly used expressions, when discussing everyday physical events

Christer Alvegard, Learning Lund, Sweden

Elsie Anderberg, Department of Education, Sweden

Lennart Svensson, Department of Education, Sweden

Thorsten Johansson, Department of Philosophy, Sweden

Students' conceptions within the domain of classical mechanics have been investigated many times, but usually the relationship between expressions and meanings have not been problematized. During the last decade an interest has emerged concerning relationships between expressions, meanings and conceptions made possible by an intentional-expressive view of language. This paper presents empirical findings from a study made in this latter tradition. A number of students at Chalmers

University of Technology, Sweden, were asked to describe a situation concerning the physical movement of a body. In the special dialogue-structure used, the student also reflected upon his/her expressions. This made the distinction between expression and meaning possible, and thus made it possible to reveal the student's meanings of expressions used. We present results concerning the relationships between expressions and meanings, and between these relationships and conceptions. Classical mechanics and everyday life share many expressions, such as acceleration, force and energy. In this paper we examine in depth, from the student's point of view, the use and meanings of these expressions. Our analytic distinction between expressions and meanings makes it possible to analyse students' conceptions in terms of meanings separated from expressions. Earlier research has shown that students often use the above-mentioned expressions in common sense ways, even in a disciplinary context. In the dialogues we have found instances of this, as well as of disciplinary meanings of non-disciplinary expressions or expressions shared by the disciplinary and everyday contexts. Example of this latter relationship is when the expression force is given the disciplinary meaning energy. From an educational point of view this is quite different from giving the expression force the everyday meaning of force. In the paper we discuss and categorise relationships between meanings and expressions used by students, as well as the educational implications.

– 141 –

Highly structured open inquiry labs

Jonte Bernhard, Linkoping University, Sweden

Anna-Karin Carstensen, Jonkoping University, Sweden

Oskar Lindwall, Gothenburg University, Sweden

According to several studies MBL is very effective in fostering a good functional understanding of physics. By using video recordings of students' interaction, we have made in-depth analyses of students' courses of action in two introductory mechanics courses and in an electrical circuit theory course. We have explored the different ways students orient to, interpret, and participate in MBL. Traditional taxonomy of laboratory instruction styles often suggests that there is a dichotomy between labs characterised as structured and open inquiry. We question this dichotomy and based on our empirical data we will show that students' courses of action in some dimensions are framed by encounters with the instructions, the technology, the teacher, and other peers wile they in other dimensions are free to explore. Therefore we propose that MBL-tasks could best be described as being both highly structured in some aspects and open inquiry in other.

B 9 23 August 2005

Paper Presentation

17:00 - 18:20 Room E004

SPECIAL EDUCATION AND STUDENTS AT RISK

Chair: George Spanoudes, University of Cyprus, Cyprus

Directionality of metaphor in language/learning disabled and typically developing children

Christina Karefillidou, University of Cyprus, Cyprus

Demetrios Natsopoulos, University of Cyprus, Cyprus

This study was conducted in order to examine Language/Learning Disabled (LLD)

Children's comprehension of conventional metaphors when compared to chronological-age-matched (CA-matched) children. The study sample consisted of 50 children who were categorized as LLD by the cognitive referencing criterion, and

50 matched controls. That is, each LLD child was required to have a significant discrepancy between his/her Performance IQ (PIQ) and Verbal IQ (VIQ) with

PIQ>VIQ. Children in the control group (typically developing children, TDC) ex-

– 142 –

hibited an equal PIQ (standard score) that is at least the average 100 in comparison with their peers whereas they scored above average on the VIQ scales. In the experiment presented here, children (aged 8:8-11:9) were provided metaphorically related terms with the instruction that they be assigned a-term (i.e., source) or b-term (i.e., target) status in the construction of metaphors. Children's representations of conventional metaphors in their long-term lexical or semantic memories were examined in three conditions: (a) in feature-target pairs, (b) in feature-source context1, and (c) in feature-target context2. The results indicated that overall the TDC group scored significantly higher on comprehension. Children with LLD consistently performed poorly on the metaphor comprehension task. The groups did not appear to respond differently to changes in condition. The group of LLD demonstrated the same comprehension pattern as PIQ-matched controls for the three conditions.

Context variations had a statistically significant effect on Children's performance.

Taken together, the results of this study appear to indicate the nonreversibility of metaphors in either group.

Educational psychology practices in cyprus: the process and value of early identification of children at-risk for learning difficulties

Timothy Padopoulos, University of Cyprus, Cyprus

Merope Iacovou- Kapsali, Ministry of Education and Culture, Cyprus

Michalis Ioannou, Ministry of Education and Culture, Cyprus

The present study piloted the administration of a set of tests that have been previously used only for research purposes, in an attempt to testify their suitability of meeting the Educational Psychology Service's (EPS; Ministry of Education and

Culture) needs for early diagnosis of learning difficulties, attention problems, behaviour difficulties and mild mental retardation. Among the administered tests were the following: the WISC-III-Revised (Wechsler, 1992), the Cognitive Assessment

System (CAS; Naglieri & Das, 1997), a set of phonological awareness, rapid naming speed, orthographic knowledge, and reading accuracy measures, as well as two checklists for teachers, the Attention Checklist (ACL, Das, 1986; Greek adaptation by Papadopoulos, Panayiotou, & Georgis, in preparation) and a subscale for hyperactivity (based on the diagnostic criteria of DSM-IV; APA, 1994; previously used by Papadopoulos, Panayiotou, Spanoudis, & Natsopoulos, 2004). Participating children (9 males and 5 females), drawn from the EPS records, attended kindergarten (n = 6) and grade 1 (n = 8), deriving from 11 schools and 2 school districts, representing cases being at-risk for or exhibiting (1) learning disabilities, (2) attention problems, (3) learning disabilities and attention problems, (3) behaviour difficul-

– 143 –

ties, (4) learning disabilities and behaviour difficulties and (5) mild mental retardation. Results showed that the cognitive and behavioural profiles of the participating children were distinctly different. Also, overall, the ratings given by the classroom teachers appeared to coincide with those obtained by the administration of the psychometric tests. Discussion centres on the differential diagnostic functionality of the used tests emphasizing the relation between the abilities measured by these tasks and the aspects of school performance represented in the given population.

Students with Special Needs in Norway's National Assessment of Writing Proficiency

Ragnar Thygesen, University of Stavanger, Norway

There are reasons for concern as to what such testing will mean to students with special needs. The presentation describes the assessment, its policies and practices for accommodations, and results from the 2005 administration of the test. Issues in implementing large-scale, high-stakes assessments are also discussed.

Predicting which students might be academically at risk in higher education

Lin Norton, Liverpool Hope University College, United Kingdom

Bill Norton, Liverpool Hope University College, United Kingdom

Aim: This paper reports on findings from work in a university college in the UK, using Meyer's (2000) Reflections on Learning Inventory (RoLI) as part of a programme designed to enhance students' meta-learning awareness. The research aims to identify patterns from RoLI scores that might predict which students would be successful and which might be at risk of failure in their studies. Method: As part of a compulsory module on Personal Development Planning (PDP), all first year students complete the RoLI, and then are given information on what their scores mean in the context of the expectations of studying for their main academic subjects. In

2002 and 2003, students were asked if they would be willing to submit their RoLI scores for research purposes. 230 students from the 2002 cohort and 380 from the

2003 cohort submitted completed questionnaires, which were matched to the students' records for measures of academic performance, gender and age variables and details of subjects studied. Outcomes: These data will be explored using factor analysis, multiple regression analysis and structural equation modelling to identify predictive patterns for students who do well and students who may need additional support. Methods of using the RoLI as a diagnostic tool will then be developed taking into account age, gender and subject which will enable students at risk of

– 144 –

failure to be targeted through institutional student support mechanisms. It will be equally important that if patterns are identified predicting success that this is conveyed to students thus giving them advice, which is evidence-based. Significance:

The findings from this research will be discussed in the context of how understanding indicators of success and struggle can be used to build more effective learning environments.

B 10 23 August 2005 17:00 - 18:20

Paper Presentation

Teacher Education

PREPARING FOR THE FIRST YEARS OF TEACHING

Chair: Xenia Hadjioannou, University of Cyprus, Cyprus

Room E010

What is a good ‘teachership’ about? Student teachers' early professional identity

Kirsi Pyhalto, Helsinki University, Finland

Tiina Soini, Tampere University, Finland

Maijaliisa Rauste-von Wright, Helsinki University, Finland

Satu Eerola, Tampere University, Finland

A teacher's professional identity development is - or should be - a vital concern of teacher education programs. The early professional identity of student teachers orients them toward their future profession and act as an active basis for meaningmaking and new learning. It is also the framework through which student teachers view their training and find personal meaningfulness in the profession. Accordingly, knowledge on the part of teacher educators and mentors about professional identity and its formation is important, because it provides them with a better understanding about professional development, and tools for guide and support the process.

During the last decade, teachers' professional identity and identity-development has emerged as a separate research focus. Despite of this recent extensive research interest, the concept of the professional identity of teachers is still vague, and is defined variously in different studies, or sometimes is not defined at all. This implies that to be a functional tool in understanding and guiding the professional identity development more attention should be paid in defining and thus understanding the concept, as well as the interactive role of teacher education as a context of profes-

– 145 –

sional identity formation. This paper analyses and discusses central aspects of the professional identity formation of student teachers. These aspects are analysed both theoretically and empirically within the frame of a pragmatic constructivist learning theory. An examination of the professional identity concept and it's role in student teachers' professional development, in the context of teacher training as a learning environment is followed by an analysis and discussion of two empirical case studies about first year student teachers' analyses of teachership. The study provides an outline of one approach to understanding and guiding the professional identity of student teachers.

Micro-politics in the relationship between a beginning teacher and his pupils

Steven Janssens, University of Leuven, Belgium

Rita Romont, University of Leuven, Belgium

Saar Van Hool, University of Leuven, Belgium

"Micro-politics" refer to processes at the school level through which individuals or groups use power to realise desirable working conditions (professional interests).

They perform micro-political actions (strategies and tactics) and show a degree of

"micro-political literacy" (they "read" working conditions as "coloured" by a struggle between different interests). The development of micro-political literacy is part of a (beginning) teacher's professional growth. Traditionally, the concept of micropolitics refers to the school level. In this study it is extended to the classroom level, more specifically to the relationship between a (beginning) teacher and his pupils.

The present study is meant to answer five research questions:

(1) what do beginning teachers know about micro-politics at the classroom level?

(2) which professional interests do they identify with respect to themselves?

(3) which strategies do they use to realise these interests?

(4) which interests do they identify with their pupils?

(5) which strategies do these pupils (according to them) use to realise these interests?

Eleven beginning teachers were interviewed three times. The "September" interview was mainly introductory. The "November" interview focused on micro-political knowledge and the "Christmas" interview on micro-political interests and strategies. Beginning teachers show detailed knowledge about "power in the classroom". They aim for three "didactical" interests: stimulation of learning processes, motivating pupils and keeping discipline. More generally, they stimulate the development of values. The interviewees link the strategies they use with these interests.

"Revision", "being patient" and "structuring" are important for the stimulation of

– 146 –

learning processes. For the development of values, beginning teachers try to be "a good role model". The interviewees identify several interests and strategies from the side of the pupils. Pupils "argue" and "ask questions" to "postpone or prevent something they dislike". "Asking questions" however helps as well to "slow down the class" and to "make it more pleasant".

Anxiety of future teachers - self-report and projective data

Yvona Mazehoova, University of South Bohemia, Czech Republic

Iva Stuchlikova, University of South Bohemia, Czech Republic

The aim of the study is to investigate the anxiety of future teachers. Self-reported anxiety is compared with and projective data assessment. As the rules for negative emotions expression are rather strict for helping profession, there is an assumption, that the self-report of anxiety of future teachers could be distorted toward low anxiety. Self-report (Spielberger's State Trait Anxiety Inventory) and projective measures of anxiety (coloured self-portrait and colour modification of Wartegg's drawing test) were administered to 174 teacher students (136 women, 38 men). The social desirability was controlled by Marlowe Crown Social Desirability Scale. The repressors subgroup was identified using Wienberger's combination of low self-reported trait anxiety (STAI) and high social desirability (MCSD). Results shows that self-report scores of anxiety were lower then population norms based assumption, whereas composed projective scores indicated higher than normal anxiety. Incremental differentiation power of proposed projective assessment of anxiety was assessed via comparison of repressors and truly low anxious. Discriminant analysis based on projective criteria yielded significant differentiation between repressors and true low anxious students. Future teachers seem to present themselves as low anxious whereas in projective assessment they show remarkable amount of anxiety.

The combination of the two approaches - self-reporting and projective assessment provided more elaborated view on student teachers anxiety.

Student teachers' action-control beliefs across teacher education

Lars-Erik Malmberg, University of Oxford, United Kingdom

Todd D. Little, University of Kansas, United States

Brigitte Wanner, University of Montreal, Canada

We investigated student teachers' action-control beliefs about making children learn, subdivided into agency, control and means-ends beliefs. Four intra-agent (effort, ability, personal characteristics, structure), four inter-agent (esteem-supportiveness,

– 147 –

realistic feedback, performance enhancing, and linguistic awareness), two extraagent (supporting collaborative learning and creating a positive relationship with parents), and two external causality belief constructs (luck and unknown) where included in the Teacher's Control, Agency and Means-ends (TCAM) questionnaire.

Construct reliability and structural validity of the constructs were satisfactory. Student teachers were followed up in a cross-sequential study at a teacher education department, 2000-2003. When changes in action-control beliefs across teacher studies were investigated, using individual growth models (in MLWin) from entrance exam to the fifth year among 191 student teachers, it was found that agency and control beliefs declined slightly from the entrance exam and then began to increase to a high after the second year of studies. Means-ends beliefs showed an upward trend across teacher education. Further, we inspected effects of gender, age, entrance exam points, accumulation of credits across the studies and stability of well-being. Older student teachers where more agentic, entrance exam observations and stability of well-being predicted higher levels of agency beliefs, and entrance exam interviews both agency and means-ends beliefs.

B 11 23 August 2005 17:00 - 18:20 Room A009

Paper Presentation

INSTRUCTIONAL DESIGN AND DESIGN RESEARCH

Chair: Lucy Avraamidou, University of Cyprus, Cyprus

Design based research: Design principles and learning processes

Sten Ludvigsen, University of Oslo, Norway

In this paper I take a socio-genetic approach in order to understand human reasoning and to show how categories are a fundamental part of learning in a specific type of institutional practice. The scientific categories are built into a web-based discussion forum (FLE2) as part of a pedagogical and technological design. The scientific categories are based on the concepts of the progressive inquiry model for knowledge construction. On the basis of the theoretical framework and the empirical analysis,

I argue that the categories in the progressive inquiry model could under certain conditions stimulate the students towards a more systematic orientation when solving problems. An important question to consider in analyses of design experiments such as the project Design of Collaborative Telelearning Artifacts (DoCTA NSS); is how

– 148 –

we should understand the activities that unfold. When we challenge the schools and the students practices with ideas about fostering higher order knowledge and skills, we create tensions. Between what most of the students usually do, and the interactional work with what new types of artifacts - prompting categories - can afford.

As analysts we should try to identify both the historical traditions in the schools and the transformative practices that evolve through the design experiment. Design experiments provide rich opportunities to focus on specific educational problems and create models that point towards improvement of practice.

Towards a robust pedagogical design for 'blended learning': A model for individual and group activities in a virtual learning environment (IGVLE)

P.J. van Eijl, University Utrecht, Netherlands

Peter de Voogd, University Utrecht, Netherlands

Albert Pilot, University Utrecht, Netherlands

Wilfried Admiraal, University Utrecht, Netherlands

Both teachers and educational designers attempt to construct higher education courses that function well both under different circumstances and for different groups of students. Whenever such an attempt is realised, we consider it as a robust pedagogical design. However, it turns out to be quite a difficult task to account for all the success factors provided by educational research about the effectiveness of virtual learning modes and aids. In the present research, a model has been evaluated in which a virtual learning environment (VLE) is combined with face-to-face learning into blended learning, in which individual and collaborative learning are integrated in a steady rhythm of weekly activities and assessments. The effects of the characteristics of different student groups (team learners, individual learners, part-time working students, and commuting students) on learning, satisfaction, and on the learning results have been investigated. The outcome of this research is that, for these groups of students, the IGVLE-model turns out to be a robust pedagogical design. This robust design is meant to be a source of inspiration for teachers to apply a VLE effectively in their own courses and to realise a sustainable quality improvement. For educational researchers it contributes to an area of research to models of pedagogical design that are so successful and robust that their application can improve the educational practice in higher education in a significant way.

– 149 –

Theory into design into theory: Two pedagogical conversations

Peter Goodyear, University of Sydney, Australia

Maarten de Laat, University of Southampton, United Kingdom

Victor Lally, University of Sheffield, United Kingdom

This paper reports on research in the area of networked (online) learning at the university level. More specifically, it considers relationships between theory and practice in teachers' educational design activity. We summarise the outcomes of our recent research using (a) pedagogical design patterns (b) critical event-based interviews with teachers, to focus on the theorisation of educational design practice and on the capacity of practice to be informed by theoretical constructs. We situate educational design activity as a complex praxis (that is, theory-laden practice) and examine the conversations between theory and design praxis. In so doing, we pay special attention to the influence of forms and representations of theoretical knowledge - concluding that the affordances of these forms and representations play an important role in shaping this field of educational design praxis. In short, educational design praxis is distributed across minds, tools, texts and other artefacts. In our specific context of design for networked learning, the tools, texts and artefacts come to hand in distinctive ways. Additionally, some of the interim outcomes of design activity take on a persisting physical/digital form and are available to scrutiny and reflection (by their creator, and others) at later times. This situated and distributed design praxis sits in a different relationship to theory than is the case in educational design praxis more generally - at least as it is conceived and represented in the literature drawn on for staff development in higher education.

Self-organized and reflective learning with technology: An exploratory design study

Priya Sharma, Pennsylvania State University, United States

Sebastian Fiedler, University of Augsburg, Germany

Ying Xie, Pennsylvania State University, United States

This paper reports on the results of a qualitative exploration of learners' practices, uses, and lived experiences of personal Webpublishing technologies to support selforganization and reflective thinking. Developing students to become responsible, independent learners requires that they become self-organized and reflective and engagement in self-organization requires one to be able to observe, reflect on, and review processes of learning. To meet the requirements for this type of learning, we designed specific interventions using individual Weblogs and RSS Webfeeds

– 150 –

to support learners' ability to record and return to thoughts and feelings. In this exploratory design study, existing Personal Webpublishing technologies were adapted to support self-organized learning and reflective thinking. Data were gathered from two designed implementations. In the first implementation, five participants engaged in Weblogging over the course of 12 weeks and gathered data included interviews with individual participants as well as existing Weblog document data.

Based on the design evaluation of and data collected in the first implementation, we modified the design of the second implementation to include more modeling and encourage group discourse. Eight participants were interviewed and their Weblog data was gathered. Initial data analysis of these 13 interviews and Weblogs reveals different levels of self-organization in the two implementations. However, both were marked by lack of consistent interactive conversation between participants, with time, privacy, and perceptions of writing emerging as significant influences on self-organized and reflective practice. An evaluation of design of the learning environment identified the need for more structured initial introduction to the technology, followed by need to encourage dialogue among learners. Research, practice, and design implications for supporting the design of learning environments for selforganized and reflective learning are presented.

B 12 23 August 2005

Paper Presentation

17:00 - 18:20 Room A010

INSTRUCTIONAL DESIGN

Chair: Elaine Munthe, University of Stavanger, Norway

The clia-model: a framework for designing powerful learning environments for thinking and problem solving

Erik De Corte, University of Leuven, Belgium

Chris Masui, Limburg University Center, Belgium

A major challenge for education and educational research is to build on our present understanding of learning for designing environments for education that are conducive to fostering in students self-regulatory and cooperative learning skills, transferable knowledge, and a disposition toward competent thinking and problem solving.

Taking into account inquiry-based knowledge on learning and recent instructional research, this paper presents the CLIA-model (Competence, Learning, Interven-

– 151 –

tion, Assessment) as a framework for the design of learning environments aimed to be powerful in eliciting in students learning processes that facilitate the acquisition of productive knowledge and competent learning and thinking skills. Competence refers to the components of competence in a domain; Learning pertains to the characteristics of effective learning processes; Intervention relates to principles guiding the design of learning environments; and Assessment refers to instruments for monitoring learning and teaching. Next, an intervention study is described that embodies major components of this framework, and focuses on the acquisition of metaknowledge and self-regulatory skills in university freshmen. Starting from a constructivist perspective on learning, several aspects of competence, namely metacognitive and affective skills and related metaknowledge, were integrated into the real instructional context of an experimental group (E) of 47 first year students in business economics. The study yielded promising support for the CLIA-model by showing that CLIA-based learning environments are indeed powerful in facilitating in students the acquisition of higher-order learning results, especially the acquisition and transfer of self-regulation skills for learning and problem solving. The study also proves to be supportive for promises of design-based research put forward by the Design-Based Research Collective in the 2003 volume of the Educational Researcher, namely the exploration of the potential of novel learning environments, and the development of contextualized theories of learning and instruction.

New learning and types of learning environments in dutch upper secondary education

Jos de Kock, University of Amsterdam, Netherlands

Peter Sleegers, University of Amsterdam, Netherlands

Rinus Voeten, Radboud University Nijmegen, Netherlands

In this study, the results are presented of a large-scale survey in which Dutch secondary geography, English, and physics teachers participated. The survey aimed to find out what types of learning environments are dominant in contemporary Dutch upper secondary education and to what extent those types meet the conditions for

New Learning. For this aim, an instrument (the QALE) was developed with subscales on teacher-learner roles (three scales), learner-learner roles (one scale), and learning goals (two scales). On the basis of cluster analyses two, about equally sized, groups of teachers are identified, reflecting two types of learning environments. Type 1 largely appears to fit New Learning features. This first cluster includes teachers who tend to strive for intrinsically motivated competence, for selfregulation by students, for cooperation between students, and for developing social

– 152 –

skills more than teachers in the second cluster. The second cluster is better characterized by traditional features of learning environments. Also, teachers in the first cluster implement the developmental and apprenticeship model of teacher-learner roles more than teachers in the second cluster do. There are no large differences between clusters with regard to the realization of the behavioral model. Importantly, cluster membership appears unrelated to the kind of subject a teacher teaches. A statistically significant relationship, however, is found with the grade level at which a teacher teaches; higher grade levels show a higher proportion of seemingly new learning environments. Even though the first cluster shows features of New Learning, it must be said that, all in all, there appears to be only a limited degree of interest in acting according to New Learning principles.

A professional change that made a difference: An analysis of an innovative instructional design

Mary Pearson, University of Sherbrooke, Canada

Francois Larose, University of Sherbrooke, Canada

Aims: This study examined the innovative practice of a teacher in a low-income school who restructured his teaching to provide his 23 pupils with the right tools to learn: a classroom environment supported by ICT within an interdisciplinary, project-based perspective. The research question was posited to examine the effects of such an approach to teaching and didactical and pedagogical mediation of a socioconstructivst orientation on teacher- pupil interactions and on the building of cross-curricular competencies. Methodology: Based on a theoretical perspective consistent with the cultural anthropology of Vygotsky (1978), the study focused on the social interaction of pupils who embarked on a three- year adventure to enhance their learning through computer skills and developing cross-curricular competencies. The research design allowed us to see developmental changes and witness the development of behaviour in individuals participating in the study (Bordens and

Abbot, 2002). In order to link the data collected to the research question, a pragmatist paradigm was used to incorporate qualitative and quantitative approaches: lexicometrical analysis of teacher discourse and video observation of behaviour.

Since the study relied on multiple sources of evidence, specific approaches were used to collect and analyze data (Yin, 1994). Three distinct samples were used: 1) the pupil sample (N=23); 2) 47 elementary teachers with characteristics similar to the teacher targeted (N=48); 3) a documentary sample relevant to the instructional device (lesson planning, pupils' work, etc.). Outcomes/Theoretical and educational significance: At the end of the study, the 23 pupils fared better on district tests in

– 153 –

reading and writing than the 600 pupils combined and compared well in Mathematics - an exemplary feat for a teacher whose professional decision tested the socioconstructivist theory and shed light on how children learn.

Instructional design and class discussions for computer tools in mathematics education

Michiel Doorman, Freudenthal Institute, Netherlands

The use of computer tools influences the process of students' mathematical sense making and the role of the teacher. In this paper we present results of a design research project on the use of computers in mathematics education. We focused on the contribution of the use of computer tools to the students' learning and on ways for supporting the teaching. Specific computer lessons were designed together with a scenario for the teacher. The activities and the scenario were tested during teaching experiments in two tenth-grade classes. We collected data by video and by audio taping whole class discussions and group work. The data was interpreted in terms of what preceded the lessons and the tools provided. The qualitative analyses showed that the students used a variety of strategies of trial and error during the computer lessons. As a result, the teacher had difficulties in discussing the computer activities and connecting them with the following activities. In a second teaching experiment we added open-ended activities before the computer lessons. These activities were designed to support students to contribute to the understanding of a problem that has to be solved, and to the invention of representations which might be helpful for solving the problem. In addition, after the computer lessons we added small group activities and a scenario for a class discussion to exchange experiences. The changes in the instructional design had as a result that students used more qualitative reasoning during the computer lesson. Their activities were more productive as a result of the compatibility between the representations in the tools and the students' current reasoning. In addition, the teacher was able to guide the students in both their understanding of a global problem, and in the relation between the tools and the solution of this problem.

– 154 –

B 13 23 August 2005 17:00 - 18:20 Room A008

Paper Presentation

Higher Education

TEACHING IN HIGHER EDUCATION

Chair: Sari Lindblom-Ylanne, University of Helsinki, Finland

Academics' recollections of the reflective processes they engage in when learning about teaching: An analysis across two ìdisciplinary groups

Carolin Kreber, University of Edinburgh, United Kingdom

One consideration guiding this study was whether due to the nature of their discipline academics might engage in reflection on teaching differently. Specifically, the study compared academic teaching staff in the natural and life sciences to academic teaching staff in the humanities and social sciences with regards to the extent to which they engage in the reflective processes suggested by transformative learning theory (Mezirow, 1991) when thinking about their teaching. The three reflective processes distinguished by Mezirow's transformative learning theory are content, process and premise reflection, which in essence are three different levels of reflection. Individuals engaged in premise reflection, by questioning the presuppositions underlying the problem they encounter (critical reflection), may re-frame teaching problems they encounter in new ways and as such arrive at a transformed, and possibly more discriminating and integrative, conceptualization of teaching and learning. Mezirow's three levels of reflection were applied to three domains of teaching knowledge: 1. instructional methods, 2. student learning, and

3. educational purposes. Thirty-six academic staff in the natural and life sciences and eleven academic staff in the humanities and social sciences participated in a semi-structured interview to identify the extent to which they engaged in content, process and premise reflection in the three domains of teaching knowledge. Data were analyzed primarily deductively using the three levels of reflection as a priori codes. The greatest difference between the two groups of staff was found in relation to premise reflection, with humanities and social science staff engaging in this process seemingly more often than science staff. Most reflection occurred in the domain of instructional methods followed by reflection on student learning and development. Reflection on educational goals and purposes was observed less often but was identified more frequently with humanities and social science staff than with

– 155 –

natural and life science staff.

How do academic disciplines influence the teaching knowledge used by university professors?

Denis Berthiaume, McGill University, Canada

Academic disciplines are generally believed to influence teaching and learning at the university level. Yet, little is known about the specific impact they have on the teaching decisions of university professors. Shulman's notion of pedagogical content knowledge (PCK) is often used to explain how teachers use particular strategies to teach particular subjects in particular contexts. However, attempts at empirically applying PCK to university teaching have revealed the inadequacy of the concept.

This may be due in part to the difference existing between Shulman's notion of content and the nature of academic disciplines in higher education. Therefore, a new constructódisciplinary pedagogical knowledge (DPK)óis envisaged to better represent the nature of teaching within university disciplines. The overall purpose of the study is to devise a conceptual framework that would describe the various sources contributing to the formation of DPK and their interrelations. As such, the aim of the presentation is to introduce the DPK conceptual framework, to describe its various components and their interrelations, as well as to provide illustrations from particular cases in university teaching. Using a multicase approach, data were collected from four university professors teaching in four different disciplines in order to build the DPK conceptual framework. Preliminary results seem to point to three grand sources of influence in the formation of DPK, namely the professor's teaching knowledge base, his/her personal epistemology, as well as the discipline's sociocultural characteristics and epistemological structure. In addition to refining our understanding of the impact of academic disciplines on teaching at the university level, this study contributes a framework for identifying ways to better support university professors as they grow as teachers within their discipline of instruction, a need identified in the faculty development literature.

Differences in an experience of difference itself: A study of teaching approaches in senior accounting classes

Lynne Leveson, La Trobe University, Australia

As professional accounting programs are often criticised for failing to produce adequately prepared graduates, good quality teaching at the senior levels, is considered essential to a faculty's reputation and future prospects. Critical appraisals have

– 156 –

prompted accounting education researchers to investigate more closely their course structure and delivery, although only recently have attempts at understanding this from the teachers' perspective received attention. This paper describes a study into the teaching approach of a group of Australian academics teaching senior classes in accounting undergraduate degrees. Anecdotally, there is a belief that student characteristics and challenging curricula mean that better quality learning occurs at the senior level and that teachers will adopt more sophisticated teaching approaches as well. The study described here explores this assumption on an empirical basis.

It adopts a phenomenographic approach to investigate the ways in which a group of academics experience teaching their senior compared to their junior accounting classes. The variation in this experience is described in two ways, on an intra-individual and an inter-individual basis and the results are presented within a two dimensional outcome space. Contrary to anecdotal evidence, no lecturers' accounts were interpreted as adopting a more complex teaching approach at the senior level. However, within the group this ìconsistency was experienced in a number of qualitatively different ways. The results highlight the two orientations to teaching, broadly known as teacher and student-centred. Another factor which appears to sharpen the distinction between these orientations was also identified. This is the perceived relevance of students' experience in the learning process. That considered relevant by the teacher-centred lecturers concerns experience gained through exposure to subject content whilst the student-centred lecturers also recognised and used their students' personal life experiences as well.

Curriculum development in Higher Education: Participatory analysis of core content

Lena Levander, University of Helsinki, Finland

Minna Mikkola, University of Helsinki, Finland

Designing and managing the teaching content has traditionally been to a large extent a private issue of the individual university teacher. There are, of course, educational goals and written syllabuses for the degree programmes but within the programmes individual teachers do not necessarily have to communicate the content of their teaching in sufficient detail. However, there are pressures to break this tradition. There is an increased interest in enhancing the quality of student learning, and questions about public accountability of higher education. At the same time there is growing international competition as well as demands for promoting mobility in higher education. This paper examines the outcome of dimensional core content analysis of courses taught in the degree programmes at the Faculty of Agriculture

– 157 –

and Forestry, University of Helsinki, Finland. Our aim is to review especially how the core content was presented in the descriptions. This analysis process was an integral part of an educational development project called Development of Structure and Content of Teaching. The aim of this project was to compact and clarify the curriculum in the degree programs. An electronic course questionnaire was designed for the description of the scientific and professional knowledge and skills as analytical dimensions. The resulting descriptions were stored at a teaching data base for the teachers to browse and study. However, we discovered that the quality of the descriptions varied. The generation of comprehensive and thorough descriptions was impacted by the application of the participatory process when developing content for higher education courses. The narrowness of some descriptions calls for further elaboration of the process. The electronic description instrument used by the teachers seems to be a feasible solution for the sharing and managing the core content of degree programmes.

B 14 23 August 2005 17:00 - 18:20 Room A107

Paper Presentation

Student Learning in Higher Education

PROBLEM BASED LEARNING

Chair: Vermunt Jan, University Utrecht, Netherlands

Change in student skills following the implementation of a PBL curriculum

Benoit Galand, University of Louvain, Belgium

Etienne Bourgeois, University of Louvain, Belgium

Mariane Frenay, University of Louvain, Belgium

Instructional practices inspired by Problem-Based Learning (PBL) are more and more often implemented in Higher Education throughout the world (Evensen &

Hmelo, 2000). In spite of its growing popularity, evidence regarding the efficacy of PBL to improve student learning is mitigated (Newman, 2003). The aim of this study was to assess change in student skills following the introduction of a new PBL curriculum in a School of Engineering. The implementation of this new curriculum provided the opportunity to assess the impact of long-term (two years) enrolment in a PBL environment on student skill development. A criterion referenced test was developed to assess several skills and not only factual knowledge acquisition. This

– 158 –

test was completed by four cohorts of students, two of them who had completed the lecture-based curriculum and two who had completed the PBL curriculum. Taken together, 393 students participated to this study. Comparison of student performance before and after the curriculum change indicates several differences in favour of students from the PBL curriculum and no differences in favour of students from the lecture-based curriculum. Results of the present study suggest that the introduction of a new PBL curriculum had no negative effect on student knowledge and skill development. On the contrary, they suggest that students attending the PBL curriculum have developed new skills compared to students from the previous curriculum.

Despite its limitations, this study shows that a curriculum change guided by PBL principles is possible and could foster the development of some skills among students, apparently without deleterious effect on other forms of learning.

The effects of university teachers' approaches to teaching on students' approaches to learning in a problem based learning environment

Jeannette Hommes, University of Maastricht, Netherlands

Mien Segers, University of Leiden, Netherlands

Wim Gijselaers, University of Maastricht, Netherlands

Educational innovations like Problem Based Learning should create a learning environment which encourages students to higher order learning activities. Nevertheless, recent meta- analysis on the effect of Problem based learning do not present conclusive results. It is argued that problem based learning has a strong positive effect on the application of knowledge but not a strong effect on the level of knowledge of the students. The question arises if all the conditions for educational innovations are met. One of the influencing factors on student learning is the teachers' approaches to teaching and learning. In this research a case study in a problem based learning environment is presented. A quantitative exploration of the teachers approaches to teaching and student approaches to learning showed a significant higher preference for Conceptual Change Student Focussed (CCSF) teaching approaches instead of Information Transmission Teacher Focussed (ITTF) teaching approaches.

The results from the students' questionnaires showed a significant higher preference for deep learning approaches instead of surface learning approaches. These results indicate that teachers in an innovative educational environment have a teaching approach which is consistent with the aim of the innovation. However, only low correlations were found between Conceptual change student focussed teaching approaches and deep learning approaches of students. On the basis of the results, it can be argued that more factors have an effect on students' approaches to learning.

– 159 –

Support and development of teachers' conceptions of teaching and learning and support for construction of courses and assessment, which are coherent with their conceptions, are a necessary prerequisite in education al innovations.

PBL in architecture revisited

Erik de Graaff, University of Delft, Netherlands

In 1990 the faculty of Architecture in Delft introduced Problem-Based Learning

(PBL) as the core method of a new educational programme. Important objectives of the curriculum innovation were to integrate theoretical oriented learning with architectural design practice and to strengthen the cohesion within the programme in order to facilitate management of the curriculum. Despite some difficulties with the management of change in an existing programme, the implementation was quite successful. However, the new PBL curriculum was short lived. A little over ten years after the introduction the last remnants referring to were removed from the curriculum. Based on faculty documents and interviews with key players in the educational management of the past ten years, this study will describe the process of disintegration, trying to analyse the factors resulting in the eventual break down.

An authentic task in an Informatics course in higher education

Hanne ten Berge, IVLOS, Utrecht University, Netherlands

Stephan Ramaekers, IVLOS, Utrecht University, Netherlands

Sjaak Brinkkemper, ICS, Utrecht University, Netherlands

Albert Pilot, IVLOS, Utrecht University, Netherlands

A recent review of literature we carried out gave insight in the design features of authentic tasks that provoke higher-order learning in higher education. Usually, these cases refer to complex situations, containing open-ended, ill-defined or even hidden problems and often require a multidisciplinary approach. Confronted with such tasks, students start off with an analysis of the case to establish which (theoretical) questions should be answered in order to address the underlying problems and issues adequately. In the process of analysis, gathering information, constructing and testing possible solutions, they supposedly develop the competencies that are needed to deal with the kind of problems and issues that arise in professional academic practices. Cases with a good design can offer powerful learning environments. Authenticity, structure, complexity and challenge appear to be important features in the design. To gain insight in the concept of the four design features in characterising the design of courses that contain authentic tasks, we explore these

– 160 –

features and their coherence more closely. This will lead to a better description of the component of authenticity in courses in higher education. Furthermore, we draw a provisional conclusion to the positive and negative implications of the use of authentic elements in the course design on learning. Ultimately, our aim is to be able to give course designers principles and guidelines on the design of authentic tasks in higher education. For more specific and detailed research we focus on a Master course called Informatics Business, which is designed around a fictive software enterprise and teaches students what developing their own software product and starting an own software enterprise entails. We describe the design of the course and search for answers to some questions on the design features.

B 15 23 August 2005 17:00 - 18:20 Room A018

Paper Presentation

PROFESSIONAL DEVELOPMENT OF SCHOOL TEACHERS

Chair: Christopher Clark, University of Delaware, United States

Fostering teachers’ professional development through a project-based research learning environment and performance assessment

Miriam Welicker, Oranim College of Education, Tivon 36006, Israel

The researcher conducted three workshops on research methodology for teachers

(n=43), with the aim of investigating and exploring a school project, and fostering reflection on it. They studied educational research methodology (both quantitative and qualitative) and conducted their own chosen project-based "field" research, some of which was action research. The constructivist learning environment was multi-staged, with student-student-instructor interactions. The teachers' research portfolio served as a diagnostic and formative performance assessment strategy.

The research goals were to investigate the teachers' perceptions of their professional development and their attitudes towards the learning environment and the portfolio procedure. The research procedure included content analysis of the teachers' interviews and reflections in the portfolios, as well as quantitative and qualitative analysis of their answers in the questionnaire. Research outcomes revealed that the teachers ascribed great importance [M=3.88, out of 4) to the project-based learning environment and formative assessment strategy. They pointed out its contribution to their self-confidence and pedagogical maturity and their acquisition

– 161 –

of applicable research skills. They discussed their growth regarding a deep and multi-dimensional comprehension of the investigated project, academic improvement throughout the rewriting procedure, and awareness of their professional development, and of the need for life-long learning. Despite difficulties - mainly the

'hard labor' process which accompanied the performance assessment, most of them felt highly motivated autonomous learners (which indicates their internal locus of control and professional development), within this challenging constructive learning process. Some reported that the research procedure contributed to their social and professional status at school, but this was sometimes accompanied by frustration and difficulties in leading the change process (related to their project). As part of their research conclusions and insights, they raised some specific alternatives for coping with these difficulties. This study widens the perspective on constructivist theory and its implications on cognitive and professional development.

Design and evaluation of critical thinking scaffolds to support action research for teacher professional development

Steven J Coombs, Bath Spa University College, United Kingdom

Sarah Fletcher, Bath Spa University College, United Kingdom

The action research mode of enquiry-based learning is one the most important paradigms that validates' classroom-based research and on-the-job professional enquiry of teachers' and trainers in schools and colleges. Indeed, action enquiry has recently been supported by the OECD (2002) as a useful means through which applied research schemes can improve the teaching profession and cited the UK's Best

Practice Research Scholarship. However, the action research qualitative process is generally seen as being less systematic when compared to more traditional positivist experimental methodologies. This paper proposes a systematic and transparent experimental approach to support teachers' carrying out classroom-based enquiries through action research (Lee & Coombs, 2004) and adopts the pedagogy of critical thinking scaffolds (CTS) (Coombs, 2000). CTS can be applied as a professional development toolkit to help scaffold and enrich systematic enquiry through developing enabling heuristics that deliver personal project management. The systems thinking conversational science paradigm of Self-organised-Learning (S-o-L)

(Thomas & Harri-Augstein, 1985) has been applied to an action research paradigm from which the toolkit has been designed. Case study evidences drawn from teacher professional development prototypes will be illustrated from which a generic pedagogical design protocol has emerged.

– 162 –

What can be learnt from beginning teachers’ narratives

Orna Schatz Oppenheimer, Hebrew University, Israel

The aim of this research paper is to learn about the teacher's professional world during the first year of work and to examine the components typical to the critical stage of the entry to the professional world and, through them, to find ways to prepare teachers to deal with the professional reality and help them become better teachers. The paper presents the beginning teachers' world as reflected through their stories, from their point of view and from their experiences and experiments. From a nation-wide Life Stories Competition held amongst beginning teachers in Israel,

80 stories were chosen. These stories were used for narrative research. The hypothesis is that stories are a reflection of the story-teller's consciousness and, therefore, through them, we can learn of the professional problems with which the beginning teacher has to contend. Ten stories were discussed in this research. Five judges from the academic world made the selections. The criteria were based on quality and relevance to the first year as a beginning teacher. The subjects under discussion were analyzed for content and the central theme of each story was revealed. Afterwards, one theme that recurred in all of the stories was chosen.

Secondary school teachers' concept-maps and beliefs regarding active & self-regulative learning

Inge Bakkenes, IVLOS, Utrecht University, Netherlands

Jan Vermunt, IVLOS, Utrecht University, Netherlands

Theo Wubbels, PDI, Utrecht University, Netherlands

Jeroen Imants, Nijmegen University, Netherlands

Based on current knowledge on learning and teaching, the importance of active and self-regulative learning is emphasized. Secondary school teachers are expected to focus more on facilitating, supporting and monitoring learning and less on merely transmitting subject matter knowledge. This change of focus appears to be difficult.

Many teachers are still looking for suitable didactics to promote active and selfregulative learning. They are in the middle of learning processes! Research groups at the universities of Utrecht, Leiden and Nijmegen now collaborate to gain insight in the learning processes of secondary school teachers within the context of active and self-regulative learning. Together they perform a study in which teachers are followed in different learning environments. Part of this large study is describing

(changes in) teachers visions and beliefs regarding active & self-regulative learning. The research questions that are addressed in this paper are 1) whether and

– 163 –

in what way teachers differ in their visions on active and self-regulative learning, and 2) whether and in what way these differences are related to teachers beliefs on learning and teaching. In order to gain insight in teachers visions on active and self-regulative learning, their tacit knowledge was studied using concept-maps.

Concept-mapping traditionally is a time-consuming method. To be able to study the concept-maps of a large group of teachers (N=100) we developed a computer program that enables teachers to sort, select and relate relevant concepts on their own. To elicit and interpret the concept-maps, network analysis was used. Teachers' beliefs were measured with the help of a questionnaire. The questionnaire consisted of eight sets of items representing dimensions that are relevant for active and selfregulative learning. The analyses show that teachers differ markedly in the structure and content of their concept-maps and that these differences are related to teachers' beliefs on teaching and learning.

B 16 23 August 2005 17:00 - 18:20 Room E009

Paper Presentation

Curriculum Studies

CURRICULUM DEVELOPMENT AND IMPLEMENTATION

Chair: Gert Rijlaarsdam, University of Amsterdam, Netherlands

Action research in curriculum development: Integrating movement into current activities of kindergartens

Asher Shkedi, The Hebrew University of Jerusalem, Israel

Hagit Shkedi, Kibbutzim Educational College, Israel

In recent years researchers and educationalists have increasingly recognized that movement and its cultivation are a significant factor in child development but do not receive the attention that they warrant. In response to this deficiency we felt a deep sense of obligation to create a curriculum that would enable kindergarten teachers to bring movement to their students in an effective manner.Using action research methodology, we worked with kindergarten teachers over a period of time, carefully observing their behavior and their reactions in both the workshops and the kindergarten activities. The participants' reactions, comments, and insights influenced the content of the new curriculum. We identified fundamental problems involved in teaching movement in kindergarten as well as different approaches to

– 164 –

incorporating movement in kindergarten activities. We discerned that most movement activity that takes place is random and episodic with no clear relation to the broader framework of kindergarten activities. Although the teachers had expressed positive attitudes to movement in the kindergarten, they did not regard movement as an integrated part of the ongoing learning activities that take place. After working with the curriculum which was developed during the study, the teachers expressed that it facilitated their ability to integrate movement activities in their daily program by providing them with skills and confidence.

On curriculum implementation and student learning

Maria Araceli Ruiz-Primo, Stanford University, United States

Kennedy (1998, 1999) strongly proposes documenting paths of influence from policy manipulations to student outcomes as a strategy for better understanding the process of education and, more specifically, the links between instructional practices and student achievement. This paper proposes and empirically tests a framework for studying the implementation of a middle-school inquiry-based curriculum. We focus on the link between the intended and the achieved curriculum (the measured student learning). Our basic premise is to help explain the student outcomes measured as well as to discern differential effectiveness based on the variations observed in the enacted curriculum. The following questions guided the study:

To what extent does the implementation of the curriculum by the teachers in the study correspond to what the curriculum developers intended? Do the instructional practices observed reflect the objectives of content and pedagogy expressed in the curriculum? Does the fidelity of the implementation have an effect on student performance? We tracked down the types of curriculum using a multi-method strategy and different sources of information: curriculum materials, teachers, observers, and students. We collected data across 12 teachers and 298 students in six states across the US. Teachers were videotaped on every session in which they taught the curriculum. We collected classroom artifacts and students' science notebooks. Students were pre- and post-tested with an integral battery of multiple-choice items, predictobserve-explain prompts, short-answer questions, and performance assessments.

Each classroom was visited once over a two or three-day period focusing on small groups work. The paper will triangulate information across sources of information to provide a profile that describes the variability of the enactment of the curriculum and how this variability relates to students learning.

– 165 –

The same teacher and the same intended curriculum: Is it the same implemented curriculum?

Tammy Eisenmann, Weizmann institute of science, Israel

Ruhama Even, Weizmann institute of science, Israel

The aim of this research is to study the role of the class in curriculum implementation. We examine a case study composed of a teacher who teaches the same intended mathematics curriculum in two seventh-grade classes from schools with different socio-cultural backgrounds. Data sources include: observations of a whole learning unit in each class; teacher and students interviews; and artifacts produced during the school year. Data are analyzed both qualitatively and quantitatively. We mainly use the ìGrounded Theory method and methods used in the two TIMSS video studies. The findings of this study show that in general the implemented curriculum in the two schools was similar and compatible with the intended curriculum. In both classes the teacher taught the same mathematical content, assigned mainly tasks aimed at introducing new concepts and ideas and at advancing mathematical understanding and conceptual knowledge. She also organized her instruction in similar ways. In spite of the above similarities, three major differences in instructional organization were identified. The first is related to the way the teacher launched students' small group work; the second to the way she conducted whole class discussions, and the third to the coverage of tasks in class. These differences are related to the schools' more general culture (for example, the intended curriculum guidelines fit one of the schools' common practices in subjects other than mathematics, but not the other) and students' characteristics (for example, in one of the schools students tended not to complete small group tasks). The findings of this research contribute to the theoretical body of knowledge in highlighting the role of the class and the school in curriculum implementation. They also provide important information for curriculum development and for professional development for teachers that take into account the ìspace where the curriculum is implemented.

– 166 –

B 17 23 August 2005 17:00 - 18:20 Room E113

Paper Presentation

Learning and Instructional Technology

PEDAGOGICAL ISSUES

Chair: Rich Mayer, University of California, Santa Barbara, United States

The structuration of group interaction and its effects on the co-construction of knowledge

Wouter Van Diggelen, Utrecht University, Netherlands

Maarten Overdijk, Utrecht University, Netherlands

Jerry Andriessen, Utrecht University, Netherlands

The notion of structure is a central concept of social constructivist perspectives on learning. Structures arise and evolve from the symbolic, tool mediated discourse within a student group and give meaning and direction to that discourse. The objective of our study is to investigate how structures emerge during face-to-face and tool mediated communication within student groups and how these structures shape the co-construction of knowlegde. In this paper we built upon the concept of structure by referring to structuration theory (Giddens, 1986) as it has been applied in group dynamics. To examine how structures emerge and influence the co-construction of knowledge, we analysed the students' (inter)actions on a mirco-level. The research has been carried out at a secondary school with a 5th grade class of 21 students.

Our focus is on student face-to-face and computer mediated interaction while engaged in small-group discussion about solutions for a societal problem. A process analysis of students' participation has shown that, in contrast to oral, face-to-face communication, the computer mediated interaction is less constrained to one dominant course of action. A content analysis showed that all student computer mediated contributions were related to the co-construction of knowledge. This may be due to the fact that the students could also communicate face-to-face. An analysis of faceto-face interactions revealed that this mode was mainly used for the planning of the task, providing support or telling jokes. We can also conclude that the computer mediated communication was mainly used to remove uncertainty caused by ignorance or imprecision of a shared interpretation of the situation. The communicative acts were directed at acquiring new information that enables the students to form a precise interpretation. Negotiations that are triggered by conflicting viewpoints

– 167 –

were rare, and if they occur they were never properly completed.

E-pedagogies for networked learning

Maarten De Laat, eLearning Research Centre at the University of Sou, United

Kingdom

P. Robert-Jan Simons, Utrecht University: IVLOS, Netherlands

We would like to put the pedagogical approach to online teaching and learning to the centre of our attention and describe how pedagogies in e-learning since the emergence of this field have developed. In this contribution we will compare and discuss pedagogical models used in education and indicate how over time our understanding of e-learning has evolved. De Laat & Simons previously described and reviewed group learning theories and developed some first principles of how collective learning could be supported and implemented. This work is further developed by a review on empirical research on networked learning practices. In this review, studies carried out in asynchronous networked learning communities have been synthesized to develop a body of knowledge of current findings in the field CSCL.

This way we aim to develop an evidence based pedagogical model of networked learning that can critically engage with theories of learning as well as produce a set of principles to inform the current e-pedagogies. In this paper we will present an overview of pedagogical models for networked learning, describe how they have evolved over time and develop, based on research, directions for the future.

Situated pedagogic expertise in technology-integrated mathematics and science teaching

Sara Hennessy, University of Cambridge, United Kingdom

Kenneth Ruthven, University of Cambridge, United Kingdom

Rosemary Deaney, University of Cambridge, United Kingdom

This paper reports on a project which sought to elicit the evolving pedagogic expertise for creating effective learning environments with information and communication technologies (ICT) in secondary school mathematics and science. The research aimed to distinguish transposable components and situational variants in pedagogical thinking across a range of classroom settings and to illuminate how teachers both reframe their goals and actions in response to the affordances offered by new technologies and adapt to the constraints imposed (Wertsch, 1998). Exemplary cases were carefully selected through a multi-stage process. Focus group interviews with recommended subject departments solicited teachers' examples of

– 168 –

successful practices integrating the use of ICT. Five practices were subsequently investigated through 19 case studies: dynamic geometry and graph plotting in mathematics; multimedia simulation, interactive whiteboard, and data capture and analysis in science. Cross-case analyses drew on in-depth data from two lesson observations and semi-structured post-lesson teacher and pupil interviews in each case. The findings illustrate how contextual factors shape practitioners' use of ICT.

For example, while teachers used unanticipated software behaviours as learning opportunities, their rhetorical endorsements of idealised forms of use supporting pupil-regulated experimentation were largely unrealised in practice. Such investigative use was constrained by existing patterns of practice and pedagogy associated with systemic school and subject cultures, curricular and assessment frameworks, time limitations and other pragmatic considerations. Teachers nevertheless configured use of ICT to fit their own settings, strategically managing the mode and extent of use according to perceived software complexity and pupil capability. Careful sequencing and design of lesson activities typically harnessed visual affordances of the technology to support knowledge building and consolidation. Exemplars illustrate responsive, pedagogical adaptations of practice including development of collaborative, dialogic interaction and bridging between informal and formal models, using the technology as a dynamic, manipulable object of joint reference.

The role of the eTutor- Evaluating tutor input in a virtual learning community for psychotherapists and psychologists across Europe

Chris Blackmore, University of Sheffield, United Kingdom

Emmy van Deurzen, University of Sheffield, United Kingdom

Digby Tantam, University of Sheffield, United Kingdom

SEPTIMUS is a new one-year postgraduate training course combining internetbased theory teaching and face-to-face supervision and personal experience, primarily targeted at psychologists and psychotherapists across Europe. The course has been taken by over 150 students from 14 European countries in the last 3 years.

A review of the course after its first year led us to implement a more collaborative learning approach, and to provide eTraining to our tutors in this. We present the results of course evaluations undertaken before and after the implementation of collaborative learning, focusing particular on the role of the course tutor and the effects of different cultural styles of learning in the 8 countries where participating training institutes were based. Our results show that there was a change in tutor style between the first and second year, corresponding to the introduction of the collaborative learning approach. This had significant effects on student satisfaction but not

– 169 –

on self-reported level of understanding. We cannot be sure that these changes are due to the introduction of collaborative learning, as there were also differences in the two student intakes, and there may also have been an effect of tutor familiarization with what was to many of them a novel way of teaching. The results do support evidence from other studies that the nature and frequency of tutor engagement with students is a crucial factor in the success of eLearning programmes. International tutor training has therefore proved essential, as the role of tutor has been interpreted in different ways in different countries.

B 18 23 August 2005 17:00 - 18:20 Room A019

Paper Presentation

Educational Technology

ICT IN EDUCATION

Chair: David Wray, University of Warwick, United Kingdom

Ict in the new emerging teacher education, - an ecological perspective on learning and assessment

Knut Steinar Engelsen, Stord/Haugesund University College, Norway

This paper is based on an analysis of a three year holistic, ICT-oriented action-research project in teacher-education at Stord/Haugesund University College in Norway. I have, from an ecological perspective, studied the roles of ICT as part of the innovation process, in an interplay with other artefacts and innovative didactic agents. My analysis indicates that a core premise for the use of ICT to be legitimated in the activity of learning, seems to be that the actors experience that the tools give them substantial help in their teaching and learning so that they can achieve more overall learning outcomes. My study also shows that competence-development within the field of ICT in learning seems highly depended on ICT having an academic focus and basis and be accorded legitimacy through ordinary academic activities. As an example of academic contextualisation my study takes a closer look at the role of ICT in developing new portfolio-based assessment models. This part of the study is mainly based on an empirical analysis of four different subjectoriented portfolio-models. My conclusions here support other studies that seem to indicate that portfolio-based assessment could be a strong innovative artefact. In addition it shows that digital tools and infrastructure could be central agents for

– 170 –

developing portfolio-models, which to a great extent are collaborative in nature and are characterised by a high degree of student involvement, and which emphasise meta-cognitive aspects like self-assessment and reflection. This seems to be an example of how ICT is legitimated through the assessment models, and that the symbiosis of portfolio based assessment models and ICT could have some backwash effect on the structure of the learning environment in general. My findings will be discussed in light of a socio cultural perspective on learning, with special attention to Etienne Wengers concepts participation, reification and negotiation.

Towards a framework for designing ICT-support for reflective learning activities in competency-based, multiprofessional education

Ilya Zitter, Utrecht University, Netherlands

Robert-jan Simons, Utrecht University, Netherlands

Olle ten Cate, University Medical Center Utrecht, School of Medic, Netherlands

Increasingly professionals are confronted with situations which require them to cooperate with professionals from other backgrounds to reach optimal results. Multiprofessional situations place serious demands on cooperative abilities. Educational institutes are struggling with the question of how to prepare students for these demands. The research aims to result in a framework, which is expected to be usable for educational designers or teachers with experience in educational design. It should be suitable to assess and/or design effective ICT-support for competencybased, multiprofessional education. The aim is not to design The Ideal ICT-Support, but to find guidelines for ICT effectively supporting learning activities in such education. The proposed framework consists of the following elements:

(1) The educational setting, which aims to develop the competencies needed to work in multiprofessional situations.

(2) Learning activities: we believe reflection on one's perspectives and the perspectives of others will help professionals from different backgrounds to connect.

Explicit reflection may help to make implicit representations explicit. Explicit representations will help to establish shared context.

(3) ICT-support: Our assumption is that well-designed ICT-support will persuade students to reflect explicitly on their own perspectives and on the perspectives of students from other professional backgrounds, resulting in making implicit representations explicit.

On the basis of preliminary research and literature research, a version of the proposed framework was made. Consequently, a Delphi-study was carried out. The contributions of 3 panels (in total 17 participants) were analysed and summarised

– 171 –

into 19 guidelines. Subsequently, the guidelines were assessed by the panels. In addition to the above, guidelines to (re)design usable ICT-support were collected from literature. Future research will include expert validations, which will be used to evaluate the framework, afterwards, case studies will be carried out.

A framework for investigating school development through ict

Liisa Ilomaki, University of Helsinki, Finland

Minna Lakkala, University of Helsinki, Finland

The research on ICT in education has concentrated on pedagogical aspects, but ICT has influenced radically a school community as well. ICT seems to foster features characteristics for advanced school communities. A knowledge-creating school is an absorbing idea, which combines the positive effects of ICT in education, and the results of research about learning organizations to an entity, which is a more effective and capable school for the information age. The purpose of the present study was to develop a generic framework for investigating the relationship between ICT and school's working culture and practices, and to investigate schools by using the framework. The study was conducted in nine schools in Finland. In the study, a framework was created, which helped to observe schools from several points of view. The framework consisted of six research phenomena: The goals of the school,

Expert-like working culture in the school, Leadership, Teacher community's working culture, Pedagogical practices, and The ICT resources. The data consisted of interviews of the different actors, questionnaires, classroom observations, and formal and informal documents. The schools differed from each other radically in all dimensions investigated. It was typical for schools on good ICT level that several teachers used ICT, it in many ways. Teachers had more common visions and goals, and commonly agreed working manners. These schools had more networking, and they supported common development work. It is obvious, that a meaningful use of

ICT in a school demands many years' practice and development. The model for the study proved to be a powerful tool to analyse schools. The method was, however, labour-intensive, and there is need to further specify the content of the investigated dimensions, in order to concentrate on the most important factors of the relationship between school development and ICT.

Practices of computer use in elementary education: Perceived and missed opportunities

Tirupalavanam G. Ganesh, University of Houston, United States

David C. Berliner, Arizona State University, United States

– 172 –

This research endeavor reports on a large qualitative dissertation study of eight upper elementary grade (4/5/6) teachers' computer uses in education. Long-term data collection occurred during the 2001-02 and 2002-03 school years in eight classrooms located in seven Arizona (United States) sub-urban schools in a large public school district with 25 elementary (K-6) schools and over 25,000 students. The goals were to identify what local school administrators considered successful educational technology practices and examine the observed practices in relationship to expectations for technology's potential to transform education. The study was set in classrooms that functioned under typical conditionsóone or two computers in the classroom with periodically scheduled access to the computer lab, which is the norm in the nation's public schools. The study revealed that computer technology uses as designed, supported by school district administration, and as enacted by classroom teachers were primarily to improve teacher efficiency and student productivity. Computer technology was seldom used for instruction and curricular technology integration was symbolic. The school as an institution largely served as a barrier to teacher innovation with regard to technology integration. A perceptible focus was on the appearance of modernity. The efforts to infuse and support computer use manifested in a shiny well maintained computer lab and a part-time computer resource technologist for each elementary school. Yet, students' access to computers in school was only 1% of the entire school-year's instruction time. Classroom teachers' computer use was defined by use of software programs in reading and mathematics and its uncritical acceptance as a tool to manage students' reading practices and learning in mathematics. This reliance on computer technology to manage, assess, and track students allowed for the production and documentation of students' progress with ìcomputerized reports, a hallmark of efficiency, modernity, and preparation for life in an industrialized society.

– 173 –

B 19 23 August 2005 17:00 - 18:20

Paper Presentation

Web-based learning

PARTICIPATION IN ONLINE LEARNING COMMUNITIES

Chair: Richard Joiner, University of Bath, United Kingdom

Patterns of participation in online learning communities

Jimmy Jaldemark, Mid Sweden University, Sweden

Room E111

This paper focuses on patterns of participation in on-line teaching. It is about initiation, turn taking and the maintaining of a dialogue; it uses data from a on-line course in higher education; and it contrasts online practices - as retained on the course server - with the IRE model of classroom discourse discussed, thirty years ago, by Sinclair and Coulthard. The heart of the paper reports and discusses the patterns used by three teachers and fifteen students. Particular attention is given to the difference between academic and social discourse. The subsequent discussion focuses on how the affordances offered, on paper, by the course design yet were swamped by other constraints in the actual learning environment. The paper concludes with a discussion of the gap between the practices revealed in this study and the idealizations of the inner life of learning communities that are presented in the international literature. In effect, this paper is a validation, if not a confirmation, of

Wallace's claim that teachers' uses of the Internet are far from the deep and engaging activities implemented by research projects' (2004, p. 448).

Comparing knowledge construction in two cohorts of asynchronous discussion groups with and without role assignment

Tammy Schellens, Ghent University, Belgium

Hilde Van Keer, Ghent University, Belgium

Martin Valcke, Ghent University, Belgium

This paper describes the impact of learning in asynchronous discussion groups on students' levels of knowledge construction. The choice for a design-based approach made it possible to compare the results of two cohorts of students (respectively N =

223 and N = 286) participating in the discussion groups during 2 consecutive years.

Multilevel analyses were applied to uncover the influence of student, group, and task variables on the one hand and the specific impact of the assignment of roles to

– 174 –

group members on the other hand. The results for both cohorts of students indicate that a large part of the overall variability in levels of knowledge construction can be attributed to task characteristics. As to the impact of student characteristics, more intensive and active individual participation in the discussion groups as well as adopting a positive attitude towards the learning environment is positively related to students' achieved level of knowledge construction. As to the impact of task characteristics, significant differences between the consecutive discussion themes were found. The results showed a significant decrease in levels of knowledge construction. Further analysis however illustrated that this significant decrease in levels of knowledge construction disappeared when correcting for task complexity. As to the impact of role assignment, no significant overall differences in students' mean levels of knowledge construction between the role and no role condition was observed. However, additional analyses revealed (1) that the distribution patterns of the levels of knowledge construction differed: students in the role condition more often reached the highest level; and (2) that assigning students the role of summariser resulted in significantly higher levels of knowledge construction. Comparisons between the 2 consecutive cohorts revealed that the introduction of roles lead to higher levels of knowledge construction. An effect size of 0.5 was detected.

Learning together or alone? - Social skills and Web-based environments in higher education

Hannele Niemi, University of Helsinki, Finland

The session focuses on the question of how to advance collaboration in higher education through the Web. The aims of the session are (1) to introduce the theoretical framework a Web-based tool ìThe IQ Team which is an interactive on-line assessment and support system to learn social skills needed in co-operative work, (2) to describe data collection using the IQ Team in virtual environments, and (3) to present the empirical results of on-line students' social skills in different groups, and (4) to present what kinds of different social orientations students have in their learning. The data (N= 275) was collected from on-line students in different disciplines in universities and polytechnics: Theology, Social and Behavioural Sciences, Business, Health Sciences, Teacher education, and Technology and Science.

Gender could be identified in certain groups (67 males, 163 females), as well as a breakdown in HE institutions (university students 82, polytechnic students 48).

The on-line survey data was analysed using exploratory and confirmatory factor analysis, estimating the goodness-of-fit, e.g. a chi-square (? 2), and examining the homogeneity with Cronbach's alpha. Means, standard deviations and t-test were es-

– 175 –

timated for different student groups, as well as correlations between social dimensions. The results indicate that the most of on-line students have high social skills.

The data also gave evidence that there are six different social types among students.

There were no significant differences between university and polytechnic students.

The gender roles seem to be traditional also in on-line courses. The results revealed that students need different strategies: individualistic, competitive and collaborative orientations in joint knowledge creating processes and they must learn to use them in a strategic way to achieve joint objectives.

Graduate students as peer tutors in asynchronous discussion groups: Exploration of individual tutoring styles

Marijke De Smet, Ghent University, Belgium

Hilde Van Keer, Ghent University, Belgium

Bram De Wever, Ghent University, Belgium

Martin Valcke, Ghent University, Belgium

This paper focuses on the implementation of peer tutoring in asynchronous discussion groups in order to stimulate freshmen to negotiate and exchange thoughts while resolving six successive tasks. In the literature on CSCL prominent evidence is found that scripting collaborative activities advances the ongoing interaction.

The present study affiliates with the assumption that collaboration not necessarily and automatically leads to learning. A positive impact on learning strongly depends on the quality of the discourse, more specifically on the content of the arguments and their invitation to construct knowledge. To optimise the quality of the interaction in online discussion groups, trained tutors were introduced to facilitate knowledge building and problem-based learning. As a part of their internship, fourth-year graduate students Educational Sciences' acted as cross-age tutors. The tutor-student ratio within a single discussion forum was two to ten, namely one tutor dyad towards approximately ten freshmen. Tutor contributions were explored by means of a content analysis scheme, stressing three dimensions, namely tutors' (1) social and organisational support in the learning community, (2) support with regard to the learning content, and (3) support related to argumentative knowledge construction by the students. Additionally, tutors' gradual transition from model to coach, as a result of freshmen's growing learning experiences, was incorporated in the inquiry.

Studying tutor activity in the context of e-learning offers possibilities to discuss and optimise learning processes under accompaniment of tutors. This study specifically contributes to a better understanding of supportive interventions of a peer in

CSCL-environments. In the result section of the paper, tendencies in the evolution

– 176 –

of tutors' modelling and coaching activities throughout the discussion themes are reported and a typology of tutoring styles in asynchronous discussion groups is portrayed.

B 20 23 August 2005

Paper Presentation

17:00 - 18:20 Room E103

PROFESSIONAL DEVELOPMENT

Chair: Hans Gruber, University of Regensburg, Germany

Written restitution: A tool to sustain expansive learning

Valentina Ghione, University "La Sapienza" of Rome, Italy

The study of the construction processes of professional competences in teachers, and the experimental research of practices and tools to sustain and to promote such processes finds a useful theoretical-methodological frame in the socio-cultural approach and in particular in the Activity Theory and in Developmental Work

Research (Engestrom, 1993, 2000). Research was carried out on the self-development practices in teachers of a ìSecond Chance School for drop-out students

(Project Chance, Naples, Italy) and was based on the analysis and promotion of the construction of professional competence among operators. The collaboration was structured as a co-configuration work (Engestrom, 2004) and needed the definition of dialogical and reflective tools to make all the participants aware of the knowledges distributed in the system, to share these knowledges and to construct new competences by experimentation through practice (Shon, 1983). Written restitution of the observed elements permitted the experimentation of a form of controlled communication between the researcher and the operators, able to sustain the construction of intersubjectivity (Wells, 1993), which is an essential presupposition for the activation of change processes both in individuals and within the group.

Moreover, written restitution permitted the construction of an area of reflection on professional activity that sustained the expansive learning processes within the activity system. The possibility of anchoring reflective thought to a shared object makes this artefact particularly useful in the building up of communities of practices (Wenger, 1997). Bringing out emerging elements of collective practice is useful in contexts aimed at the empowerment and development of self-evaluation abilities because it can help the enhancement of self-esteem, of the sense of effectiveness,

– 177 –

and of the personal and professional identity of individuals and groups.

Development of counsellors' conceptual skills

Josef Strasser, University of Regensburg, Germany

Hans Gruber, University of Regensburg, Germany

Daniela Geissler, University of Regensburg, Germany

To establish a helpful working alliance, psychological counsellors have to develop an adequate view of their clients' problems. Counsellors have to decide what treatment program is appropriate, to what extent it is applicable, or in what way it has to be adapted. These decisions are guided by a counsellor's case conceptualisation.

Little is known, however, about the development of counsellors' conceptual skills.

Research on expertise and professional performance in other domains revealed that there are substantial differences between novices and experts as to the speed and way they process information. To become an expert, a minimum amount of at least ten years of deliberate practice is needed. In the present study, counsellors of different levels of expertise were analysed during working on a computer-based case presentation. Processes of problem understanding and hypothesis formation were analysed in order to investigate whether evidence from related domains can be found in counselling as well. Specifically the following aspects were focused: (1) duration and extent of problem understanding and solving; (2) extent and elaborateness of case conceptualisation; (3) thematic reference of information processing;

(4) aspects of hypothesis formation. The results suggest expertise-related differences in the development of counsellors' conceptual skills. Experts in counselling develop a working model of a case that fits the specific situation of the client. This model has constantly to be adapted to new information gathered. Therefore expert counsellors try to stay open to new information by producing more general hypotheses. Experience helps them to recognise more relevant information, so they can express more aspects when working on the case and when explaining it. These results imply consequences for the design of learning environments that foster counsellors' conceptual skill development.

– 178 –

Measuring different aspects of transfer from training in OSH

Wolfgang Gallenberger, BG Institute Work and Health, Germany

Kati Masuhr, Technical University Dresden, Germany

Professional training should result in a transfer to the workplace (Simons, 2004).

But until now there was a lack of instruments to measure the transfer from participants who attended seminars with different learning goals. Therefore an approach was created to measure the transfer independent from training content and to differentiate between several aspects of transfer. The survey refers to three theoretical approaches, the four levels of program evaluation (Kirkpatrick, 1998), the framework model of Baldwin and Ford (1988) and the four criteria for evaluating transfer suggested by Stiefel (1976). 171 training participants who had visited different seminars at the German BG-Institute for Occupational Work and Health responded to this questionnaire. The results indicated that half of the participants estimated to utilise 50 percent or more of that learnt in training. The practical relevance of the course content and the work environment, for example supervisory support, have been identified as variables that have a high impact on the amount of knowledge that may be transferred at the workplace. The assumed correlation between the four levels of program evaluation could also be replicated in this study. A factor analysis of the 25 items that measured the several aspects of transfer from training, indicates that the transfer from training could be described by four dimensions. Thus, transfer can be described firstly by the application benefit, secondly by the skill enhancement, thirdly by the detection of new problems in the occupational activity and the motivation for advanced vocational training, and finally through the communication (e.g. to colleagues) of what a participant has learned. The study shows that transfer is a multidimensional construct which can be measured with the revised questionnaire.

– 179 –

B 21 23 August 2005 17:00 - 18:20

Paper Presentation

ENVIRONMENTAL EDUCATION RESEARCH

Chair: Kostas Korfiatis, University of Cyprus, Cyprus

Room E104

Using theoretical concepts in practical situations: A study of pupils' work on freshwater pollution

Karolina Osterlind, Department of education, Stockholm university, Sweden

The paper reports a study of five pupils learning about freshwater pollution. The pupils, aged 13-14, in the upper level of compulsory school, co-operate in a group.

The instruction is organised so that includes explanations of theoretical concepts and processes, like drainage basin and water pollution, and fieldwork at a lake that are used to explore these concepts. Data consists of tape-recorded conversations among the pupils in the group and the pupils' written material. The result shows that the pupils experience difficulty in applying the theoretical knowledge in the practical contexts. These difficulties are of two kinds. One is a problem of linking an abstract concept, like drainage basin, to a concrete reality. The other is a problem of linking a micro level to a macro level of description in the understanding of water pollution.

The pupils interpret the practical activity within a practical context, which make up a context for interpretation other than a theoretical context. The pupils' difficulty in learning the theoretical concepts is seen as a problem of contextualization, a view that is then related to ideas of concept formation within former research.

Environmental knowledge and attitudes of secondary school students in Turkey

Elvan Alp, Middle East Technical University, Turkey

Hamide Ertepinar, Middle East Technical University, Turkey

Ceren Tekkaya, Middle East Technical University, Turkey

Ayhan Yilmaz, Hacettepe University, Turkey

The aim of the present study is two fold: (1) to determine 10th grade high school students' environmental knowledge and attitudes in Ankara, Turkey (2) to examine whether there are differences between 10th grade students, exposed to an environmental science course and those not exposed to this course, on environmental attitudes and knowledge. The data was obtained by the administration of Turkish

– 180 –

version of Children's Environmental Attitudes and Knowledge Scale as measuring instrument to 559 high school students exposed to a one-year environmental science course and 484 high school students not exposed to this course. The students had favorable attitudes toward the environment while their knowledge on environmental issues was fragmentary and incorrect. The environmental science course had a significant effect on students' attitudes toward the environment in favor of 10th grade students exposed to this course. On the other hand, no statistically significant effect of environmental science course was found on environmental knowledge of the students. The positive impact of the environmental science course was desirable for students' attitudes toward the environment. However, it must be considered that it is needed to prepare environmentally literate students who would protect the environment through making informed decisions.

Exploring students' learning in environmental education

Cecilia Lundholm, Department of Education, Sweden

Mark Rickinson, National Foundation for Educational Research, United Kingdom

This paper explores students' learning and learning experiences in environmental education (EE). Recent reviews in the field of environmental education research have made clear that insufficient attention has been paid to question of learning. In particular, there have been very few empirical investigations into the process (as opposed to the outcomes) of students' learning experiences. There has also been a failure by environmental education researchers to engage with learning theory.

In the light of this situation, this paper will report findings from two studies that focused specifically on learners' experiences of and responses to environmental curricula. The purpose of the paper will be to highlight the complexity of the learning experience within EE and to draw attention to the need for improved research-based understandings of learning processes. The studies that inform this paper were undertaken independently but drew on similar theoretical perspectives and methodological approaches. The first focused on students' responses to environmental geography lessons within three English secondary schools. The other looked at Swedish university students' learning about environmental issues as part of undergraduate programmes. Drawing on data, the paper will:

- Present detailed empirical illustrations of learning challenges experienced by school and university students during environmental education courses

- Reflect upon the possibilities and limitations of what we call an intentional approach for researching students' learning in environmental education

Against the backdrop of wider research on students' learning in other subjects and

– 181 –

contexts, the paper will consider the influence of the values-rich and contested nature of environmental subject matter.

We see this paper as a contribution to the development of environmental education as a research-informed practice, and the strengthening of research and theory relating to students' learning in these contexts.

The environmental dimension in pre-service teacher training: Three case studies from Europe

Chrysanthi Kadji Beltran, University of Zurich, Switzerland

Petra Lindemann-Matthies, University of Zurich, Switzerland

Because of the multiplier effect, awareness of environmental issues among teacher trainers is of particular importance. Every teacher trainer will educate a large number of students, who themselves again will educate a much larger number of pupils. It is thus essential to ensure the quality of training offered to the teacher students, because it will contribute to the formation of future citizens. The present study is part of a large research project on Biodiversity as a value and concept in education: Initial training and professional readiness of primary teacher students'. The project is financed by the Research Promotion Foundation in Cyprus and carried out from 2004-2006 in three teacher training institutions in Cyprus, Switzerland and England. We are investigating with the help of document review and analysis, interviews with the teacher trainers, and a triangulation of the information obtained, to which degree and depth environmental and, in particular, biodiversity issues are integrated into pre-service teacher training in the three countries. Moreover, we are looking at the overlap between the content of the training programmes and the environmental education philosophy and principles of the teacher trainers. Finally, we will highlight examples of good practice in the area of environmental education with special reference to biodiversity issues. First results show that all teacher training institutions offer at least one specialized environmental education related module. However, these modules strongly differ in their context. Cyprus and England have a rather science-oriented approach whose emphasis is on information about environmental issues, whereas Switzerland has a more holistic, social-critical approach to environmental education. In all three countries biodiversity issues are integrated at least into the science-oriented modules.

– 182 –

C 1 24 August 2005 08:30 - 10:30 Room E002

Symposium

Teacher Professional Development

TEACHER LEARNING IN COMMUNITIES

Chair: Harm Tillema, Leiden univ, Netherlands

Shoshana Keiny, ben Gurion Univ of the Negev, Israel

Organiser: Shoshana Keiny, Ben Gurion Univ of teh Negev, Israel

Harm Tillema, Leiden Univ, Netherlands

Discussant: Malka Gorodetsky, Ben Gurion Univ of the Negev, Israel

Collaborative inquiry of teachers into their work and professional lives has been advocated as re-valuing the teachers' position in educational practice and eveloping their professional identity. The co-construction of knowledge can be looked upon as a prime route to building and sharing a knowledge base produced by teachers themselves, as well as a means towards self directed professional development. In particular, it features of partnership and joint collegiality are regarded as a stronghold, not only to help teachers to recognize, interpret and respond to the challenges of their practice, but also to transform their work in the context of educational reform. This teacher research, therefore, offers new ways of seeing, being and acting in the professional world of teachers; emancipating them from the mere application of externally generated knowledge . Our aim in this symposium is to study and identify areas for sharing key knowledge on issues that deal with participation in a community of inquiry which can be said to support professional learning. With this focus in mind, we will address issues of learning, of knowledge and knowing, and of practicing the profession as surfaced by the various collaborative inquiry studies conducted by the contributors of this symposium. The contributions cover the following themes as questions to be raised and debated in the symposium:

1. What characterizes the collaborative inquiry framework, and how to enhance its development?

2. How can learning and teachers' professional development be shown and validated in acollaborative community of inquiry?

3. What are the dynamics, the intra- and interactive processes within the collaborative inquiry community that contribute to the construction of knowledge?

We invited several contributions from authors active in this domain of research to provide an input for our discussion that circle around four key words: Personal

– 183 –

Autonomy; Authenticity; Identity and Uncertainty

Authenticity in collaborative learning of professionals

Harm Tillema, Leiden Univ, Netherlands

To promote knowledge productive learning in collaboratively working teams a self organized group of teachers worked together engaged in self-study and collaborative inquiry to redesign their own work environment Knowledge productive learning of professionals can be framed by taking the perspective of authenticity as a starting point, thereby integrating the elements of exchange and dialogue, self directedness and freedom, as well as collaboration in teams to create and renew a personal knowledge base. The goal of our study is to explore the parameters for possible interventions which may enhance knowledge productive learning among professionals. Evaluations of the collaborative learning processes were collected on three criteria: problem understanding, conceptual or perspective shift and commitment to implement ideas in practice. The findings revealed insights about the potential power of learning as a discovery oriented construction of knowledge. A main finding is the improvement in abilities to shift perspectives and to represent a problem under study. The collaborative inquiry method used by these teachers is evaluated as a tool for situational understanding and progressive discourse.

Collaborative learning: A mutual process of knowledge construction and of Personal autonomy development

Shoshana Keiny, Ben Gurion University of Negev, Israel

"Organizations are made of conversations" (Perkins 2003, p. 17). By that we mean that their unique nature, as compared to other forms of discussion, is that the topic of the conversation cannot be predetermined, but arises in the process of conversing.

Rather than leading the conversation, the participants are being led by it, opening themselves to others and at the same time open the possibility of affecting our understanding of the world (Gadamer 1990). Communities of Learners', are mediums that enhance open conversation, where participants can voice their unique individuality. As such, they can serve as optimal sites for participants, to define themselves and create their own identities in relationship with others (Arendt, 1958). As opposed to the common notion of collaborative frameworks as ìbrain-washing sites, that enhance uniformity and conformity, I argue that they form optimal mediums for developing personal autonomy and freedom. My aim in this paper is to show how collaborative inquiry frameworks, which develop to communities of learners',

– 184 –

enhance two coupled processes: the emergence of new joint professional knowledge, as well as the participants' personal autonomy and freedom

Teacher knowledge as identity rationality

Jukka Husu, University of Helsinki, Finland

This study aims to show how pedagogical practices can be rendered into pedagogical knowledge with the aid of concept and means of teachers' identity rationality.

When teachers undertake teaching they analyze i) their situation - what is possible; ii) their students - what their students need and what they can do; and iii) themselves as teachers - what kinds of teachers they are themselves. When teachers act and interact in a given context, they recognize themselves (and others recognize them) as acting and interacting as certain kind of persons' or even as different kinds' at once. These multiple identities are connected not only to teachers' internal states but also to their performances in schools and classrooms. Within this stance, teachers' pedagogical knowledge resides in relations as they encounter with others and situations. Negotiation processes characterize these relations and reflect the situational nature of teachers' knowledge construction though inquiry. Also, these negotiations take place in the context of larger political, historical and structural contexts of a pedagogical situation. The goal of this study is to highlight the importance of taking care the multiple contexts within which teachers and students are practically engaged. Also, it underlines the evidence how pedagogical knowledge develops through practical activities and communicative interchange. Consequently, the research task is to understand and explicate more deeply the complexity of pedagogical relations, and to use these results in the work of teacher education and teacher development

A survey and follow-up study of Norwegian teachers' professional uncertainty / certainty and collaboration

Munthe Elaine, Stavanger University College, Norway

Teacher collaboration has been advocated as an important means to enhance teacher learning for several decades. Uncertainty has been described as being endemic to teaching, and collaboration has been regarded as a means to enhance teachers' professional knowledge and competence, creating a space to address the many uncertainties of teaching. A vast amount of research has also demonstrated both the potential of collaboration as well as the pitfalls. And yet, despite the fact that teachers in Norway have a number of hours per week stipulated for collaboration, we still

– 185 –

know fairly little about uncertainty among teachers and how teachers in Norway collaborate, what they collaborate about, and how they perceive this collaboration.

Two research projects were conducted during the years 1998-2004 to shed more light on teachers' perceptions of professional uncertainty/ certainty and their participation in collaboration. A quantitative study from 1998 will address issues of covariance and variance among groups (elementary and secondary school teachers and groups differentiated by age), and a follow-up study of student teachers (2001-

2004) as they venture into teaching includes qualitative analyses of how schools may enhance uncertainty as well as how novice teachers describe the collaboration they take part in at school. The two research projects represent both quantitative and qualitative approaches to the questions, and the questions will be discussed in light of possibilities for further professional learning through collaborative inquiry

C 2 24 August 2005 08:30 - 10:30 Room A110

Symposium

Instructional Strategies

RECENT WORKED EXAMPLES RESEARCH: INNOVATIVE WAYS TO

INCREASE GERMANE COGNITIVE LOAD AND FOSTER LEARNING

AND UNDERSTANDING

Chair: Fred Paas, Open University of the Netherlands, Netherlands

Organiser: Fred Paas, Open University of the Netherlands, Netherlands

Tamara Van gog, Open University of the Netherlands, Netherlands

Discussant: Roxana Moreno, University of New Mexico, United States

Worked examples are an effective means to decrease extraneous cognitive load and free up working memory resources that are necessary for meaningful learning of complex cognitive tasks. However, freeing cognitive capacity by reducing extraneous load is not a sufficient condition for instructional formats to be effective. Hence, cognitive load research has started to shift its focus towards finding instructional techniques that stimulate the allocation of resources to activities that are relevant to learning and understanding (i.e., increase germane cognitive load). This symposium provides a state of the art overview of research on innovative ways of designing and delivering worked examples in order to increase germane load. With regard to design, the techniques presented are: the inclusion of errors in the worked-out solution

(Grosse & Renkl), the inclusion of an expert's rationale behind the selected solution

– 186 –

steps (i.e., ìthe how and why; Van Gog, Paas, & Van Merrinboer), requiring learners to self-explain the rationale behind the steps and providing feedback on that explanation (Scheiter, Gerjets, & Catrambone), and varying how actively learners process condition-action information for procedures (Catrambone). Another study explores how a learners' prior knowledge interacts with the several ways of transitioning from worked examples to conventional problems (Reisslein, Atkinson, &

Reisslein). Furthermore, two studies (Grosse & Renkl, and Van Gog et al.) do not only consider learning outcomes, but also try to advance our understanding of the actual cognitive processes and effects that different designs induce through the use of think-aloud protocols.

Exploring the relative impact of learners’ prior knowledge and type of examplebased instruction in an environment teaching circuit analysis techniques

Jana Reisslein, Arizona State University, United States

Robert Atkinson, Arizona State University, United States

Martin Reisslein, Arizona State University, United States

This study examined the effectiveness of a computer-based learning environment employing several different example-based instructional sequences to teach series and parallel electrical circuit analysis. In light of recent research suggesting that the effectiveness of these sequences may depend upon on learners' level of knowledge

(i.e., the ìexpertise reversal effect), a sample of engineering college freshmen was first grouped according to their level of prior knowledge -low or high- of electrical circuit analysis. These students were then randomly assigned to one of three instructional sequences: (a) example-problem pairs, the traditional format where worked examples preceded practice problems; (b) problem-example pairs, where practice problems preceded examples; or (c) backward fading, where a steady transition from modeling (complete example) to scaffolded problem solving (incomplete example) to independent problem solving was implemented. We found that lowprior knowledge participants provided with the example-problem pair instructional sequence outperformed their high-prior knowledge counterparts on near transfer whereas the high-prior knowledge participants presented with the problem-example pair sequence outperformed their low-prior knowledge peers. Similarly, the highprior knowledge participants assigned to the backward fading instruction sequence outperformed their low-prior knowledge counterparts on near transfer items. In sum, low-prior knowledge learners benefited from the traditional example-problem pair instructional sequence whereas high-prior knowledge learners benefited from the more non-traditional instructional sequences including problem-example pairs

– 187 –

and backward fading. Overall, this study suggests that it is important to consider learners' level of prior knowledge when designing learning environments that rely on example-based instruction. This study also indicates that the results from studies involving the design of computer-based learning environments in other knowledge domains may not directly carry over to the electrical engineering domain and that there is a need for further research on the specific characteristics of the electrical engineering domain that give rise to this phenomenon.

Can learning from molar and modular worked examples be further enhanced by prompting self-explanations and providing feedback?

Katharina Scheiter, University of Tuebingen, Germany

Peter Gerjets, Knowledge Media Research Center, Germany

Richard Catrambone, Georgia Institute of Technology, United States

Traditional molar examples are characterized by teaching problem categories and solution procedures in a rather holistic way, and therefore require that learners keep in working memory a large amount of information with a high relational complexity. In modular examples the solution procedure is broken down into as small as possible meaningful solution elements that can be conveyed separately, which resulted in less time required for learning, reduced cognitive load during learning, and improved problem-solving performance on near as well as for far transfer problems.

In this experiment we investigated whether including self-explanation prompts and feedback can enhance learning from both types of worked examples even further.

To address this question we implemented both modular and molar examples as either fully worked-out or incomplete examples. In the fully worked-out conditions, elaborated explanations for each solution step were contained in each example. In the incomplete conditions, problem categories were explained by first presenting a fully worked-out example, followed by a second example in which only the solution steps were given and learners had to specify the reasons for why a particular solution had to be applied to the problem themselves. Learners were prompted to give these explanations through questions (e.g., why is the numerator 1 in this case?).

Because learners' self-explanations may be incomplete or false, learners' answers were displayed together with corrective feedback. Subsequent to the learning phase learners were asked to solve 9 test problems, which were either isomorphic to the examples or novel. We hypothesized that the helpfulness of prompting self-explanations and giving feedback would be moderated by learners' prior knowledge as well as by the example format (molar vs. modular). These assumptions were tested in an experiment with 120 students. The data collection is finished and the results

– 188 –

will be presented at the conference.

Finding and fixing errors in worked examples: can this foster learning outcomes?

Cornelia Grosse, University of Freiburg, Germany

Alexander Renkl, University of Freiburg, Germany

Learning from worked examples is a very effective instructional method in wellstructured domains. This is, however, only true when relevant design features of an example-based learning environment are considered. In two experiments, a new feature aiming to optimize example-based learning was investigated: learning with incorrect examples. It was tested whether a combination of correct and incorrect worked examples enhances learning outcomes in comparison to correct examples only. The participants learned to understand and solve probability problems either learning with correct and incorrect examples or learning with correct examples only. The learning outcomes were assessed by a post-test. In Experiment 1, it was found that a combination of correct and incorrect worked examples fostered far transfer performance if the learners had favorable domain-specific learning prerequisites. Experiment 2 replicated this effect. Thus, incorrect examples enhanced learning outcomes, but only for "good" learners. In addition, in Experiment 2 spontaneous self-explanations were assessed by thinking-aloud protocols. It was shown that confronting learners with incorrect solutions changed the quality of their selfexplanations: On the one hand, new types of effective self-explanations could be observed, but on the other hand the amount of the very important principle-based self-explanations was substantially reduced independently of the amount of prior knowledge of the students. Thus, it can be concluded that it might be promising to foster principle-based self-explanations when learning with faulty worked examples.

Effects of process-oriented worked examples on transfer performance and understanding

Tamara van Gog, Open University of the Netherlands, Netherlands

Fred Paas, Open University of the Netherlands, Netherlands

Jeroen van Merrienboer, Open University of the Netherlands, Netherlands

This study aims to empirically test the claim by Van Gog, Paas, and Van Merri?nboer

(2004) that process-oriented worked examples would be able to foster understanding and transfer performance further than product-oriented worked examples. Process-oriented worked examples show a learner not only the solution steps, but also

– 189 –

the strategic (heuristics, systematic approaches to problem-solving) and domainprincipled information used in selecting the steps. A 2 x 2 factorial design is used, in which the following factors are varied: solution worked out (yes/no) and process information (yes/no). The resulting four conditions are: conventional problem solving, conventional problem solving with process information given, studying product-oriented worked examples, and studying process-oriented worked examples. After working on the electrical circuits troubleshooting training in one of those conditions, all participants complete a transfer test consisting of conventional problems. Participants are 56 first year electrotechnics students of 2 Dutch schools of senior secondary vocational education. They are randomly assigned to one of the four conditions. In each condition, 4 randomly selected participants are asked to think-aloud while working on some of the training tasks and all of the test tasks.

To determine whether the process-information has the desired effects on understanding and transfer performance, students' mental effort during the training and the test, and students test performance is measured. To determine how the process information might influence students' understanding, the think aloud protocols are analyzed.

C 3 24 August 2005 08:30 - 10:30 Room A112

Symposium

Knowledge Acquisition

THE ROLE OF THE BOOK IN STORYBOOK READING AS A CONTEXT

Chair: Adriana Bus, Leiden University, Netherlands

Organiser: Adriana Bus, Leiden University, Netherlands

Molly Collins, Boston University, United States

Mary Ann Evans, University of Guelph, Canada

Kathleen Roskos, John Carroll University, United States

Dorit Aram, Tel Aviv University, Israel

Maria de Jong, Leiden University, Netherlands

Discussant: Susan Neuman, University of Michigan, United States

Storybook reading is a highly acclaimed practice in providing children with early literacy experiences, although its efficacy for developing early literacy skills that support later reading achievement remains in question. Research over the past decade has examined different sources of variability within storybook reading. How-

– 190 –

ever, less attention has been paid to the book per se as a mediator for influencing interactive processes between adult and child. Recent research suggests that book qualities may be more decisive elements in the language and literacy learning outcomes of storybook reading than previously thought. This symposium includes 5 research studies that examine different book qualities, such as video fragments in televised books, on Children's language and literacy development. The scientific quality of the studies will be critiqued and implications of the findings for early literacy practice will be highlighted.

Attention to print and pictures during shared book reading

Mary Ann Evans, University of Guelph, Canada

Reading to children is often cited as a basis for children learning to read, in that children are exposed to printed text at close hand in a quiet, intimate context. It is thought to develop an understanding of print conventions, the appearance of words, and the shapes of the characters in written script. For this to be the case, children would have to look at the print while being read to. A few studies have examined parents' and Children's comments during shared reading as a window to the objects of their attention. We will report on three studies where we attempted to more directly assess this.

Storybook reading and vocabulary acquisition in preschoolers

Molly Collins, Boston University, United States

To examine the effects of rich explanation on ESL's L2 vocabulary acquisition from storybook reading, 70 Portuguese-speaking ESL preschoolers were pre-tested in L1 receptive vocabulary, L2 receptive vocabulary, and L2 expressive vocabulary.

Matched according to L2 receptive vocabulary scores, subjects were assigned to control or experimental groups. Subjects in the experimental group heard stories read with explanations of target vocabulary. Subjects in the control group heard stories read without explanations. ANOVA showed significant, robust effects of treatment (i.e., explanation). Regression analysis showed that treatment accounted for 32% of the variance in target vocabulary scores. The presentation focuses on the selection of sophisticated vocabulary, strategies for teaching vocabulary during reading, and examples of vocabulary instruction.

– 191 –

Rare word vocabulary growth in at-risk preschool english language learners

Kathleen Roskos, John Carroll University, United States

This study describes the rare word vocabulary growth of 398 children enrolled in

Southwest border community Head Starts. All children are native Spanish speakers who are participating in an Early Reading First Project designed to improve

Children's early reading skills. The early literacy curriculum (Doors to Discovery,

Wright Group/McGraw-Hill) provides daily instruction in oral language, alphabet knowledge, phonological awareness and print concepts. The curriculum offers extensive opportunities for learning new words as an integral part of storybook reading, with a special emphasis on learning content vocabulary that builds domain knowledge, referred to in the program as wonderful words (275 words). Children are afforded ample practice to hear and use new words in large group, small group and play activities. Research has shown that the Doors program is effective in promoting the oral language and print knowledge of both English-and Spanish-speaking Head Start students.

Do we need the written text in illustrated children's books?

Dorit Aram, Tel Aviv University, Israel

Participants were 30 middle income SES dyads of mothers and their children aged

4 - 6. We videotaped the mother twice in their home while ìtelling two books to her child (with text/ without text). The order of reading the two books was random.

We used two books that were new to the children. The books were illustrated by the same illustrator and we presented randomly, once with and once without text.

After the mother finished telling the story, the child was asked to retell the story, and he/she was asked questions regarding the story content and novel words from the story. Children's language level was assessed via PPVT and the Definition test.

Beyond book's genre, mothers have their own style of telling a book to their child related to the child's retelling and understanding of the story. Books without text yielded richer, more creative mother-child interactions, suggesting the language potential of this genre.

Televised books: a scaffold to preschoolers’ story understanding and language development?

Adriana Bus, Leiden University, Netherlands

With an experimental design we tested the potential role of new media in develop-

– 192 –

ing narrative comprehension and language skills in preschool children with a low level of language proficiency. In the first condition children heard the same story six times supported by static pictures; the second group heard the same story six times, but three times through televised presentations. The third group was read to six times but they heard two different stories both supported by static pictures. The fourth group did also hear two stories, each three times, every story once through televised presentation. A fifth group was only pre- and post tested. With the help of a physiological measure (skin conductance responsivity) we assessed Children's level of arousal while hearing the story. The results show an advantage for televised stories: children are more aroused. It also appeared that in a young group at-risk the availability of another symbolic system created by presenting a story with rich images and sounds promotes story understanding and language skills. Furthermore, positive effects of multimedia accumulate over sessions.

C 4 24 August 2005 08:30 - 10:30 Room A111

Symposium

Emotion

ACADEMIC EMOTIONS IN STUDENTS' SELF-REGULATED LEARN-

ING AND ACHIEVEMENT

Chair: Peter Op 't Eynde, University of Leuven, Belgium

Organiser: Peter Op 't Eynde, University of Leuven, Belgium

Reinhard Pekrun, University of Munich, Germany

Discussant: Monique Boekaerts, Leiden University, Netherlands

In the past ten years, there has been a call for empirical studies that focus on the emotional life of students. In contrast to ìcold cognition, the role of emotions has been neglected by educational research, with a few notable exceptions - research on test anxiety, and Weiner's program of attribution research (see Pekrun & Frese,

1992; Weiner, 1985). As a consequence, we lack knowledge about the interrelations of students' emotions with their goals, motivation, use of self-regulatory strategies, and academic achievement. Recently, a number of American, European, and Australian research groups have begun to conduct in-depth investigations into these interlinkages. Many of these studies build on the traditions of test anxiety and attributional research, but their scope goes well beyond that. This symposium brings together emotion researchers and programs of research from around the world as well

– 193 –

as a discussant that is a world-renowned European scholar in the field of students' affect and motivation. The five presentations all share a common focus on students' emotions. Based on the results of empirical studies, all five presentations will discuss emotions and their relationship to aspects of students' goal strivings, selfregulated learning, and academic achievement. The structure of the presentations implies that the symposium (1) discusses a number of facets of students' emotional life, including anxiety, but also negative and positive academic emotions other than anxiety, as well as students' regulation of their emotions; (2) presents findings from different populations of students; (3) addresses emotions both from a trait perspective and from a process-oriented, situated perspective; (4) examines functions as well as antecedents of emotions, and their reciprocal linkages; (5) presents empirical findings gained by different, complementary methodologies. The format of the symposium will be a traditional one with five presentations followed by remarks from the discussant, Monique Boekaerts.

Investigating the role of academic emotions in negotiating multiple goal-strivings

Jeannine E. Turner, Florida State University, United States

Cathryn E. Simmons, Florida State University, United States

As college students progress towards the completion of their degrees, many life goals; i.e., social goals and/or future career goals occupy their current concerns. Although the pursuit of their academic goals may be quite salient, students' motivation for academic achievements fit within their broader goal-striving context. Choosing and striving for goals entail complex processes that involve cognitive, motivational, and emotional systems. Sustained learning is a complex phenomenon involving a myriad of complex, adaptive processes, such as those involved in perceptual-cognitive appraisals, affective responses, fulfilling motivational goals, striving for future goals, and self-regulation as well as complex skills such as reading, problem-solving, and learning strategies. Understanding students' processes for negotiating and regulating their emotions, cognitions, and behaviors was the focus of this study. Approximately 30 undergraduate students, from various majors, volunteered to participate in this research. Throughout the semester, we both collected quantitative data

(e.g., motivational goals, effort regulation, study strategies, emotions, and grades) and qualitative data (interviews). The data suggest that, following test feedback, students experience a variety of blended emotions (e.g., feeling disappointed and worried, disappointed but content, angry, ashamed, pleased and proud). Students' subsequent motivations, self-regulation strategies, and academic achievements are influenced by their initial emotional reactions; their attributions concerning test feedback; their appraisals of their multiple workloads and resources; and the clarity

– 194 –

and importance of their future academic and occupational goals. Throughout the semester, students appraised the importance of their multiple goals, their emotional experiences, and their academic feedback and made strategic decisions and negotiations regarding their goals, effort, and strategies for each task within each class. The results of the initial analysis are congruent with current theories of goal striving, but highlight the strategic processes in which students engage to manage the many tasks that are associated with their many life goals.

Girls and mathematics : A hopeless issue? An analysis of girls’ and boys’ emotional experiences in mathematics

Anne Zirngibl, University of Munich, Germany

Reinhard Pekrun, University of Munich, Germany

Rudolf vom Hofe, University of Regensburg, Germany

Werner Blum, University of Kassel, Germany

Raymond Perry, University of Manitoba, Canada

This study focuses on gender differences in mathematics emotions. Based on

Pekrun's (2000) control-value theory of academic emotions, competence beliefs and values are assumed to be important cognitive antecedents of academic emotions mediating gender differences in students' emotional feelings. Furthermore, we hypothesized that gender-linked emotions contribute to gender differences in mathematics performance. N = 1,036 male and 1,017 female 5th grade students participated in the study, representing a broad range of student abilities and socioeconomic backgrounds. Students' emotions, competence beliefs, and value beliefs were assessed through self-report measures. A standardized mathematics test and students' academic grades were used to assess mathematics performance. Girls reported significantly less enjoyment, pride, and boredom than boys, but more anxiety, hopelessness, and shame. No significant gender differences were found for anger.

The assumption that gender differences in emotions were mediated by control and value beliefs was confirmed. Regression analysis showed that the female maladaptive affective pattern was due to girls' low competence beliefs and low subjective value of mathematics, combined with high subjective values of achievement in math which were comparable to boys' achievement values. Concerning performance, grades of girls and boys did not differ significantly, but considerable gender differences were found in the standardized mathematics test. Structural equation modelling confirmed the postulated relationships between gender, competence and value beliefs, emotions, and performance in mathematics. Furthermore, a multigroup comparison procedure revealed that these functional relations were similar

– 195 –

in boys and girls. Gender-linked differences in emotions and achievement were due to differences in mean scores of competence and value beliefs, rather than to differences in relations between variables. In conclusion, the need for interventions to improve girls' cognitive-affective attitudes towards mathematics will be discussed.

Affect during small group instruction: Implications for students’ engagement and learning

Elisabeth A. Linnenbrink, University of Toledo, United States

Kristin L. Kelly, University of Toledo, United States

Toni M. Kempler, University of Michigan, Ann Arbor, United States

Both psychologists and educators have become increasingly interested in considering affect in academic contexts (Schutz & Lanehart, 2002). Two studies (Study 1: n

= 138; Study 2: n = 192) were conducted to investigate upper-elementary students' affect during small group instruction. In both studies, a multi-dimensional model of affect consisting of both valence (pleasant, unpleasant) and activation (high, low) was used to consider how affect relates to behavioral and cognitive engagement during group work and perceptions of the quality of the group's interaction. Results indicate that it is useful to use a multi-dimensional model, as the findings varied both in terms of valence and activation. Across both studies, high positive affect

(excitement) was associated with higher levels of individual self-regulation. In contrast, both high (tense) and low (tired) negatively valenced affect was associated with a tendency to loaf, or allow other group members to do the activity. Students' perceptions of the quality of the group's interactions also varied based on affect. In study 1, low activation was associated with group interaction with reports of feeling calm associated with reporting higher quality group interactions and reports of feeling tired associated with lower quality group interactions. In study 2, feeling more happy than sad or more excited than tired was linked to higher quality social interaction suggesting that a positive valence is also important. Interestingly, students' learning during the group activities was unrelated to the affect they experienced, suggesting that affect may be more important in shaping engagement than in predicting learning.

– 196 –

The effect of achievement goals on academic emotions and task performance at a mathematical task

Jacques Gregoire, Catholic University of Louvain, Belgium

Sophie Govaerts, Catholic University of Louvain, Belgium

Benoit Galand, Catholic University of Louvain, Belgium

We studied the influence of achievement goals on academic emotions and performance at a mathematical reasoning task, and the way achievement goals are related to task performance through academic emotions. A 246 students sample (grade 11) answered a questionnaire assessing their value of mathematics and their general self-efficacy in mathematics. Their specific task self-efficacy was also measured.

The students were then assigned to three groups. Each group received different instructions designed to manipulate their goal orientation (learning goal, performance-competition goal, performance-approval goal) during a logical reasoning task. Six academic emotions (enjoyment, hope, pride, anxiety, boredom-anger, and shame) were measured with the Academic Emotions Scale (Govaerts & Gregoire, in press). Fifteen logical reasoning items were presented in three blocks of five items of equivalent difficulty. As expected, MANOVA showed that goal orientation induction had an effect on three academic emotions reported during the logical reasoning task. Students in the learning goal condition presented more hope, and less anxiety and shame than those in the performance-competition and performance-approval goal conditions. Repeated measures ANOVA showed an interaction effect between goal orientation and performance blocks. The students in the learning goal condition remained more stable across task performance than those in performancecompetition and performance-approval conditions whose performance decreased.

As expected, academic emotions were found to mediate the interaction effect between task performance and goal orientation. In the learning goal condition, students' performance remained more stable across the blocks because they felt more hope than those in the two others goal conditions. These results supported a model linking motivational and emotional components in learning context. They showed the benefit on achievement of learning goal versus performance goal. They also emphasized the role of positive academic emotions in learning and achievement.

– 197 –

Students’ self regulation of emotions in mathematics learning: How do they cope?

Peter Op 't Eynde, University of Leuven, Belgium

Erik De Corte, University of Leuven, Belgium

Inge Mercken, University of Leuven, Belgium

Over the years the concept of self-regulated learning has broadened to include motivational, volitional, and emotional components next to (meta)cognitive ones. The exact nature and role of these different components is, however, not always clearly understood. More specifically, very few data are available on the regulation of emotions in school contexts. We lack a proper understanding of students' knowledge and use of strategies to regulate their emotions in school contexts. Therefore, in this paper we will discuss the results of a survey study aimed at identifying the different categories of emotional regulation strategies students use in different contexts related to mathematics learning in school. To identify the kind of regulation strategies students employ when managing their emotions we developed a questionnaire and administered it to 393 students. Students participating in the study were in their second or fifth year of secondary education following either a classical, humanities, or vocational track. They were asked to indicate the emotional regulation strategies they draw on to manage their emotions in three different stressful situations related to mathematics learning. An exploratory factor analysis was performed to identify the different categories of emotional regulation strategies used by students. Next, variance analyses clarified the relations between the strategies used and the different situations, students' familiarity with the stressful nature of these situations, students' track level, age, and gender. Results show that students know and make use of 6 different categories of emotional regulation strategies in stressful school situations related to mathematics learning, including active and problem focused strategies as well as more emotion focused strategies. There are, however, clear differences in the kind of strategies used by students depending on the situation confronted with, their familiarity with the stressful nature of this situation, the track level they are in, their age and gender.

– 198 –

C 5 24 August 2005 08:30 - 10:30 Room A008

Symposium

Teaching and Instructional Design

THE ROLE OF PEDAGOGICAL AGENTS: WHEN DO THEY WORK?

Chair: Jan Elen, K.U.Leuven, Belgium

Organiser: Geraldine Clarebout, K.U.Leuven, Belgium

Jan Elen, K.U.Leuven, Belgium

Discussant: Peter Goodyear, University of Sidney, Australia

Pedagogical agents are animated characters designed to operate in an educational setting for supporting and facilitating learning (Shaw, Johnson & Ganeshan, 1999).

They may adapt their support to learning paths followed by students and provide students with non-verbal feedback through means of facial expressions and gestures

(Gr?goire, Zettlemoyer, & Lester, 1999; Johnson, Rickel & Lester, 2000). Pedagogical agents can adopt different roles and deliver support on different dimensions

(see Clareobut, Elen, Johnson & Shaw, 2002). Most research on pedagogical agents has focussed on more format issues of pedagogical agents, while this symposium puts the pedagogical scaffold at the core. Empirical studies are presented where the agent provides support on different levels and adopts different roles. In a first study by Domagk et al. the differentiating effect of motivation is studied in relation to the influence of pedagogical agents on leanrning outcomes. The study of Choi and

Clark focusses on the cognitive and affective effect of pedagogical agents, while the study of Clarebout and Elen investigates whether an agent providing metacognitive support is benefical for students' learning. The last contribution by Atkinson,

Dunsworth, and Reisslein discusses whether learners' English language proficiency influence the utility of pedagogical agents.

The discussion integrates the different research findings to identify functionalities these pedagogical agnet can take and issues that still need ot be studied.

– 199 –

The effect of pedagogical agents on student’s motivation and the learning outcomes

Steffi Domagk, University of Erfurt, Germany

Helmut Niegemann, university of Erfurt, Germany

Pedagogical agents are defined as animated lifelike characters in multimedia learning environments that facilitate the learning process (Johnson et al., 1998). Although this definition already implies a learning supportive role of pedagogical agents, the kind of support that they should provide and their role in the learning environment remains vaguely. In most cases an unspecific and unjustified motivational effect of these agents is assumed (Bendel, 2003). The differentiated examination of this motivational effect is the goal of this study.

The research on pedagogical agents is still at its beginning. The different roles that pedagogical agents can take over in multimedia learning environments and the different variables that they could have an effect on (such as motivation, emotion, learning outcomes) make this research field a very large one. The focus of this study is the effect of pedagogical agents on the learning outcomes mediated through the learner's current motivation. Therefore the current motivation of the learners was measured three times during the learning process to detect effects over time. After the learning period, the learners passed a learning test to investigate the effect of the pedagogical agent on retention, near and far transfer tasks. The pedagogical agent in this study has the role to guide the learner through a multimedia learning environment. It gives short introductions before the start of a new topic. The subject of the learning environment used was the visual perception of human beings. The data analysis is based on the calculation of structural equation models that take both into account, the motivational effects over time and the expected mediating role of the learner's current motivation on the learning outcomes.

Cognitive and affective benefits of animated pedagogical agent in learning english as a second language

Sunhee Choi, University of Southern California, United States

Richard Clark, University of Southern California, United States

An animated pedagogical agent is a lifelike computerized character that inhabits a computer-based learning environment to direct learning attention and provide learners with pedagogical assistance such as information and contextualized feedback. The present paper discusses the results of an experimental study that explores

– 200 –

the cognitive and affective impact of an animated pedagogical agent as well as an alternative delivery system (i.e., an electronic arrow with audio) when they are used to teach English relative clauses. Of particular interest is to investigate relative effects of an animated pedagogical agent and an electronic arrow in directing learner attention which is necessary for learning second language grammar. The study also examines the cognitive efficiency of these two media systems used to deliver am instructional methods - metalinguistic rule presentation. 60 college level

ESL students are randomly assigned to two treatment groups (Agent Group and Arrow Group), and work with respective versions of computer-assisted ESL learning software. The total of seven dependent variables are measured, which include mental effort measures, time measures, noticing measures, learner interest measures, cognitive efficiency measures, and performance measures.

Does a learner’s english language proficiency influence the utility of an animated agent and labels during multimedia learning?

Robert Atkinson, Arizona State University, United States

Qi Dunsworth, Arizona State University, United States

Jana Reisslein, Arizona State University, United States

Animated pedagogical agents are capable of delivering instruction using life-like behaviors such as gaze, gesture, and locomotion. They are designed to broaden the communicative relationship between learners and computers, which may in turn foster active engagement with the learning environment and increase the probability of future interactions with the instructional program. The research findings on whether an animated agent can facilitate learning are mixed, however. In an effort to clarify the conditions under which the deployment of an animated agent in a multimedia learning environment might be productive, we conducted a study with 111 middle-school students learning about the human circulatory system. We explored how learner performance was impacted by: (a) the presence of agent's image in a learning environment involving the human cardiovascular system; (b) the use textual labels as signals to guide the process of instructional information; and (c) the interaction between learners' English language proficiency and the presence of an agent and labels. The agent was animated to move around the screen in coordination with narration, using gaze and gestures to direct learners' attention. The labels emphasized key words from the narration and were explicitly linked to the visual material that the narration was referencing. We found that learners with limited

English proficiency benefited most from the presence of labels coupled with narration but did not profit from the presence of labels in combination with an agent. On

– 201 –

the other hand, learners classified as English proficient performed best when presented with both an agent and labels. Overall, our results revealed that learners' language proficiency interacted with the presence and absence of an agent and labels in a number of important ways. This suggests that the existing instructional design recommendations for authoring multimedia learning environments may need to be adapted in order to accommodate learners' language proficiency.

Influence of pedagogical agent on complex problem solving in open learning environments

Geraldine Clarebout, K.U.Leuven, Belgium

Jan Elen, K.U.Leuven, Belgium

Research show that pedagogical agents can acts as tools to direct students' learning processes. However, research performed with pedagogical aagents is mainly done in so-called closed learning environments. Only limited studies have been done with more open learning environmetns that aim at the student acquisition of complex problem solving skills. In this contribution research is reported that investigates the effect of a pedagogical agent on solving complex problems. Two groups are compared, one with and one without agent. An ANOVA revealed no difference between the agent and no-agent condition for the post-test score. A difference was found however, for the transfer test where the no-agent group outperformed the agent group. Apparantly, a mathematentic effect of agents was found.

C 6 24 August 2005 08:30 - 10:30 Room A009

Symposium

Teacher Education

MOTIVATIONS FOR TEACHING: PERSPECTIVES FROM ISRAEL,

GERMANY, AUSTRALIA, USA

Chair: Juergen Baumert, MPI fuer Bildungsforschung, Germany

Organiser: Jens Moeller, University of Kiel, Germany

Helen Watt, University of Michigan, United States

Discussant: Juergen Baumert, MPI fuer Bildungsforschung, Germany

Our symposium aims to bring together researchers around the world investigating individuals' motivations for teaching and the choice of teaching as a career. Re-

– 202 –

search approaches are informed by lenses from the motivation literature, and findings have important implications for teacher recruitment and teacher motivation.

This is a timely symposium, in the current climate of severe teacher shortages.

Goal orientations for teaching: A novel framework for conceptualizing and studying teacher motivation

Ruth Butler, Hebrew University, Israel

Policy papers and recommendations in many European countries in recent years emphasize the importance of attracting and retaining teachers who are not only well qualified but also well motivated. Surprisingly, however, we know little about teacher motivation, at least in part because of the dearth of compelling theoretical frameworks. After discussing how research can be informed by application of general theories of motivation, this presentation will extrapolate from achievement goal theory to develop a novel conceptual framework that is based in the assumption that the classroom is an achievement arena not only for students, whose role is to learn, but also for teachers who presumably strive to feel successful in their work, but may differ in the ways they define success, and thus in their personal goals for teaching.

Specifically, there are grounds for predicting that it will be possible to identify four distinct classes of achievement goals for teaching (mastery, ability-approach, ability-avoidance, work-avoidance) that correspond to previously identified goals for learning. Supporting data will be presented from two studies conducted with Israeli teachers. In Study 1, 30 teachers were interviewed about their reasons for feeling that they had had a successful day at work. Data from this study also served to create the new "Orientations for Teaching" measure that was administered in Study 2 to 100 teachers. Discussion will focus on the implications of this perspective for studying and understanding a) personal and organizational influences on teachers' goal orientations and b) influences of teachers' goal orientations on other features of teacher thinking and behaviour, on student motivation and learning, and thus on the creation of effective learning environments.

Different types of teacher students' motivation

Britta Pohlmann, University of Kiel, Germany

Jens Moeller, University of Kiel, Germany

The purpose of the present study was to examine profiles of motivation for choosing teacher education. For that reason we developed a questionnaire which is guided by the Expectancy-Value (Wigfield & Eccles, 2000) model and the research by Watt

– 203 –

and Richardson (2004). The expectancy-value model provides the most elaborated approach for explaining academic choices. Besides the motivations for the choice we asked teacher students for the certainty of their decisions and for their doubts about it. Exploratory factor analyses for first-year students elicited two motivational factors (an intrinsic and an extrinsic factor) and one doubt factor. By using

Mixed-Rasch models we identified 3 different types of teacher students. The three types which are characterised by the extent in intrinsic and extrinsic motivation differ in the certainty and the doubts about their decision. Additional factor analyses for senior students revealed that the motivation structure as well as the doubts differ over time. The intrinsic factor splits in four components: -educational motivationì,

-subject-specific interestì, -teaching self-conceptì, and -low demands, whereas the extrinsic factor remains stable. Regarding the doubts senior students differentiate further than first year students in -bad imageì, -stressì, and -low challengeì. In further studies the development of the different types of first-year students will be followed to obtain more information about the impact of motivation profiles on success and satisfaction in the teaching profession.

A new theoretical model for studying motivations for teaching: Theorisation and empirical evidence

Helen Watt, University of Michigan, United States

Objectives of the present study were to apply current influential models from the motivational literature to issues and findings relating to teacher recruitment. Specifically, our goals were to develop a comprehensive scale (the Factors Influencing

Teaching Choice [FIT-Choice] scale) measuring factors influencing the choice of teaching as a career, validate the scale on two large cohorts of preservice teachers in Sydney Australia, and determine the factors most important in choosing teaching for these cohorts. Our study makes important theoretical contributions through establishing a comprehensive framework to guide future investigations, as well as important applied contributions through providing an integrative measure to assess motivations for the choice of a teaching career. Values and ability beliefs emerged as strong influences on the choice of teaching as a career, as the motivational literature would suggest. Understanding preservice teachers' motivations for choosing teaching suggests implications for teacher recruitment, having clear social and political relevance in the current climate of severe teacher shortages.

– 204 –

Career change into teaching: Different motivations for career switchers?

Paul Richardson, Monash University, Australia

Governments around the globe are finding it difficult to deal with the supply, recruitment and retention of the teaching workforce, and in response are increasingly focusing recruitment campaigns on suitably qualified graduates seeking a career change. Using the recently validated FIT-Choice [Factors Influencing Teaching

Choice] Scale (Watt & Richardson, 2003, in review) we compare the motivations and beliefs of graduate and undergraduate participants entering pre-service teacher education from three universities across two States in Australia, a sample size of

1645 participants. Our scale identifies seventeen constructs related to self-perceptions of teaching ability, perceptions about the demands and rewards of teaching, as well as intrinsic, subjective attainment and social utility values about teaching as a profession. The study is distinguished from previous smaller scale, opportunistic research designs, through its sample size and scope. The profile of motivations and beliefs of graduate entrants from these three universities, making the career change into teaching, shows they are aware that teaching is a demanding and skilled career which does not offer compensations of high salary and high social status and yet, they want to become teachers so as to make a social contribution. For these graduates, the desire to make a social contribution was a major motivation for choosing teaching as a career - more so than undergraduates. However, in identifying and clarifying the motivations and beliefs underpinning the career change decisions of graduate entrants coming into teaching, we find that they are not very different from undergraduates. This suggests that recruitment campaigns could best attract graduate entrants through an emphasis on teachers' social contribution. Other major motivations for both groups were intrinsic value, job security and perceived teaching ability.

Seniority in teaching and perceptions of motivations for entering the teaching profession: A comparison between beginning and experienced teachers

Yael Katzir, Levinsky College of Education, Israel

Tal Yaakobi, Levinsky College of Education, Israel

We examine motivations for entering the teaching profession among teachers of varying seniority in Israel. Studies suggest motivations for the choice of a professional career are influenced by interactions between family, environmental and genetic factors, relationships among economic, social and cultural factors, and the sex, age and developmental stage of the individual. Most studies categorize moti-

– 205 –

vational factors into three major categories: intrinsic, extrinsic and altruistic. While the majority of such studies were carried out with populations of student teachers, our study addresses the effects of experience in teaching on motivations for the choice of a teaching career. The population under study included 414 teachers, all women, aged 23 to 66 years. Seniority in teaching was categorized into three groups by years of experience: "beginning teachers", "mature teachers" and "senior teachers". The research tool was a structured questionnaire. Findings show a predominance of intrinsic and altruistic motivations. Factor analysis elicited 6 factors

(2 intrinsic, 2 altruistic, 2 extrinsic) in the choice of teaching. Correlations were found among all four intrinsic and altruistic factors, but not between the intrinsic or altruistic factors, and the extrinsic ones. Significant differences were found between beginning teachers, and the mature and senior teachers concerning two motivational factors: "contribution to the educational system" and "convenience". In both cases the beginning teachers perceived their motivations significantly lower than the other groups. While the actual motivations for the choice of a career in teaching may be a function of personality, age, sex, social, cultural and economic factors, the individuals' perceptions of their motivations may change over time as a function of experience in teaching and professional position. Our findings demonstrate the importance of including teachers having varying degrees of experience in research concerning perceptions of motivations for the choice of a teaching career.

– 206 –

C 7 24 August 2005 08:30 - 10:30 Room A007

Symposium

Web-based learning

KNOWLEDGE BUILDING IN WEB-BASED LEARNING ENVIRON-

MENTS

Chair: Begona Gros, University of Barcelona, Spain

Organiser: Jeroen Van Merrienboer, Open University of Netherlands,

Netherlands

Isabel Alvarez, University of Barcelona, Spain

Therese Laferriere, University of Laval, Canada

Mary Lemon, OISE, Canada

Discussant: Jan Elen, University of Leuven, Belgium

Steffano Cacciamani, University of Valle d’Aosta, Italy

Much of the conventional development of web-based learning environments is based on the creation of electronic material based on existing print-based format.

In this case, the web-based learning environments have tended to display limited evidence of the learning and affordances of the new technologies. The main goal of this Symposium is to analysis different perspectives and researchers about the design of web-based learning environments to facilitate the process of building knowledge. The contributions will be address to different aspects of the process of knowledge building: scaffolds, mentoring, web-based support, affordances, etc.

Five design principles for web-based learning: Their psychological basis and effects on transfer of learning

Jeroen Van Merrienboer, Open University of Netherlands, Netherlands

Van Merrienboer's four-component instructional design model aims at meaningful learning. The first component of the model pertains to learning tasks, which are whole-task experiences that are based on real-life tasks. It may be case studies that must be studied, tasks or projects that must be performed, or problems that must be solved by the learners - all aiming at knowledge construction through inductive learning. The educational medium must allow learners to work on those tasks and in web-based learning it takes the form of a computer-simulated task environment.

This presentation focuses on the role of learning tasks in web-based learning. In

– 207 –

particular, it presents empirical studies that investigate five principles of multimedia learning, which are directly relevant for the organization and presentation of learning tasks. The sequencing principle indicates that it is better to sequence learning tasks from simple to complex, than to present them in their full complexity at once. The fidelity principle indicates that for novice learners it is better to work in a low-fidelity environment (not trying to mimic the real task environment) than in a high fidelity task environment, which contains irrelevant details that may deteriorate learning. The variability principle indicates that learning tasks must be sufficiently different from each other to allow for the construction of general, abstract schemata that make transfer of learning possible. The individualization principle indicates that adaptive web-based systems, which dynamically select learning tasks based on the characteristics of the individual learner, yield higher transfer than non-adaptive training systems. And the scaffolding principle indicates that learners should receive a high level of support and guidance for the first tasks at a particular level of difficulty, but that support and guidance should diminish as they acquire more expertise. This presentation discusses empirical studies that provide support for the five design principles.

New challenges for teachers professional development: creating knowledge building communities

Isabel Alvarez, University of Barcelona, Spain

This paper is about the introduction of Knowledge Forum for bringing Teacher

Professional Development into a more meaningful and participatory idea. Finally, the paper finishes by discussing some final considerations to help create knowledge building communities in our Spanish Teacher Professional Development context.

There are certain elements that need some further thought before we even introduce knowledge forum software into classrooms: 1) Serious awareness of the teachers,

2) Stressing the idea of community, 2) Collaborative work, and, 4) Understanding knowledge Forum as a tool to enhance knowledge building communities.

Pre-service teachers’ use of web-based environments to support field experiences

Therese Laferriere, University of Laval, Canada

This presentation will focus on pre-service teachers' use of Web-based environments to support field experiences, reflective practice, and knowledge building. Using both quantitative and qualitative methods, data was gathered over a three-year period, with pre-service teachers going to the same innovative school for learning

– 208 –

to teach in their last year of their four-year B. Ed. in Secondary Education. The university-school partnership applies the professional development school model

(PDS), and the school has a student-owned laptop program and the Web is accessed daily by school learners. Student teachers' patterns of use will be identified, and the role of accompanying school-based and university-based teacher educators will be described. At a finer level of analysis, student teachers' online discourse was analyzed in relation to the professional beliefs and skills that they manifested at the beginning of the term. Progress was assessed both in terms of reflective practice and knowledge building, and advances as regards student teachers' understanding of teaching in a web-enhanced classroom are identified.

Fostering reading comprehension with information and communications technology: The teacher’s role in online discourse

Mary Lemon, OISE, University of Toronto, Canada

This paper reports an investigation of how students' text comprehension improved through on-line discourse. Two English literature classes in an inner city multicultural secondary school participated. The teacher taught both classes-an experimental and a control class. The experimental group was a class of Grade 9 students with access to the discourse space, Knowledge ForumÆ, for writing narrative and expository texts, critiquing others' texts, and exploring ideas. The control class did not have access to Knowledge Forum. Writing autobiographies, students teaching other students how to use html, creating texts (narrative and expository) that were evaluated and revised according to teacher and student commentary, and online discourse led to a database view focusing on idea improvement that moved the curricula from a focus on tasks to a focus on real ideas and authentic problems. Reading comprehension measures showed no difference on a pretest but the experimental class excelled on question-oriented items compared to the control class on the posttest. Additionally, experimental students' intellectual activity in the database was assessed using the Analytic Toolkit (ATK), a suite of tools underlying Knowledge

Forum. ATK results were positively correlated with students' final grades. Randomly selected students in the experimental class were interviewed about their online experiences. The teacher and his students reported that interactions in the database led to a more harmonious classroom culture both on and off-line. Overall, discourse in the database resulted in improved reading comprehension, a positive correlation between database activity and final grades, and to improved class relationships.

Teacher interviews and his online scaffolding of students' work are of specific interest for this paper.

– 209 –

C 8 24 August 2005 08:30 - 10:30 Room E003

SIG Invited Symposium

Higher Education

EDUCATIONAL INNOVATIONS IN HIGHER EDUCATION: CONDI-

TIONS TO MAKE IT WORK

Chair: Mien Segers, University Leiden, Netherlands

Organiser: Mien Segers, University Leiden, Netherlands

Sari Lindblom-Ylanne, University of Helsinki, Finland

Discussant: Vermunt Jan, University Utrecht, Netherlands

Many teachers and program leaders in Higher Education invest in optimising the learning and teaching environment in order to facilitate student learning. However, educational innovations are complex processes, not always leading to the expected positive results. In this Invited Symposium of the SIG HE, four studies are presented offering empirical evidence for the role of different factors in educational innovations. By exploring the impact of innovations on student learning from different perspectives and on a variety of dimensions, we aim to contribute to the current insights in educational innovations and the conditions to make it work.The Trigwell and Lindblom-Ylanne paper focuses on the role of the teacher in innovations.

A range of dimensions in university teaching are studied: approaches to teaching, self-efficacy, regulation, interest in student learning and interest in the subject matter being taught. The results are discussed in the light of their possible impact on innovations. The Hounsell, Hounsell, Litjens and McCune paper as well as Struyven, Dochy and Janssens paper address the main theme of this symposium from the learner perspective. The former investigates the impact of changes on the courselevel in various disciplines, designed to bring about a greater congruence in the provision of extrinsic feedback and guidance and so to facilitate student learning. The latter compares the effects of being assessed by more traditional and new modes of assessment on student performance and discusses the reasons for the observed effects. As with both previous papers, the Segers, Nijhuis and Gijselaers paper explores innovation from the learner's perspective, considering assessment as well as instructional practices. It present the results of an evaluation of the effectiveness of an educational innovation in terms of the meaningfulness of students' learning

(Kember et al, 1997), taking into account students' perceptions of the learning and assessment environment.

– 210 –

Enhancing guidance and feedback to students: Findings on the impact of four evidence-informed initiatives

Hounsell Dai, University of Edinburgh, United Kingdom

Jenny Hounsell, University of Edinburgh, United Kingdom

Judith Litjens, University of Edinburgh, United Kingdom

Velda McCune, University of Edinburgh, United Kingdom

At the 2003 EARLI Conference in Padova, we presented a symposium paper which discussed the findings of a study of three final-year courses in Biology . While the first part of those findings examined the development by the students of characteristic ways of thinking and practising in biology as a discipline, the second part of the findings highlighted features of the teaching-learning environments represented by the three course units -- and notably, variations in the provision of feedback to the students --- which appeared to facilitate or inhibit the quality of their learning.

The present paper reports follow-up work in the same three university departments, focusing on two of those three final-year course units together with two first-year course units, in each of which varying steps had been taken to introduce changes designed to address students' concerns about the provision of feedback. As in the preceding empirical studies, data had been gathered from students via two questionnaires which had been designed specifically for use in the large-scale research project of which the present study forms part, and by means of semi-structured interviews with groups of students. These follow-up enquiries consequently drew upon data from 968 questionnaires and 18 interviews with 45 students. The analysis reveals a mixed set of outcomes, with marked variations across the four course settings in the extent and scope of the impact of the guidance and feedback initiatives. The paper explores possible reasons for these variations in impact, while also highlighting fundamental differences between first- and final-year undergraduate courses which may affect the ease with which changes can be implemented.

How dimensions of university teaching are related to educational innovations

Keith Trigwell, University of Oxford, United Kingdom

Sari Lindblom-Ylanne, University of Helsinki, Finland

This study aims to explore qualitatively and quantitatively a range of dimensions of university teaching and how those dimensions might be a part of successful educational innovation. Using interviews and a multi-dimensional university teaching questionnaire, we investigated areas of teaching such as self-efficacy, interest in subject matter, approach to teaching, and how they are related. We observed

– 211 –

zero correlations between Conceptual change (CCSF) approaches to teaching and

Self-efficacy and small positive correlations between Information transfer (ITTF) approaches to teaching and Self-efficacy. Importance of student learning correlated positively with both approaches, whereas interest in the subject being taught correlates positively with a CCSF approach and negatively with an ITTF approach.

Self-regulation correlates positively and external regulation negatively with a

CCSF approach to teaching. From these and similar results we have assessed the likely effects of these variables on innovation, and reached five conclusions. First, that CCSF approaches are more likely to be associated with effective innovation than ITTF approaches. Second, that in attempts to introduce innovation, both Approach to Teaching and Self-efficacy need to be considered, along with degrees of self-regulation as feelings of lack of control may result in a maintenance of departmental norms and act as a barrier to educational innovation. Third, the confidence teachers have in their ITTF approach to teaching may work against the adoption by these teachers of educationally innovative ideas. Fourth, how teachers view the importance of student learning is not related to their approach to teaching, and therefore may not be a factor in educational innovation. However, teachers who are more interested in their subject matter are more likely to adopt innovative approaches. Finally, we see benefits in educational development and training sessions that include aspects of awareness raising in the key areas described above.

The effects of the method of assessment on student performance

Katrien Struyven, University Leuven, Belgium

Filip Dochy, University Leuven, Belgium

Steven Janssens, University Leuven, Belgium

The quest for the effects from being assessed by a particular assessment mode on student performance is central to the reported research. For this study, a distinction has been made between a lecture based and a student activating learning environment. Five experimental conditions go together with one of four assessment modes, namely: portfolio, case-based, peer assessment, and multiple choice evaluation.

The investigation took place in a first year course on Child Development in the elementary teacher education program of eight different institutions. Data collection was obtained by a pre-test/ post-test-design with the help of standardised tests

(N=816). To find out whether the different assessment modes had differentiating effects on student performance, a first unexpected test (= pretest) was administered to the students during their last class session, so before having prepared for the examination, and the second test was administered after students had completed their

– 212 –

final examination (= posttest). The results show a statistically significant effect of the multiple choice test on student performance, whose students outperform their colleagues in other conditions. However, if students' preparation level is taken into account, the other assessments in the pre-test equal or exceed their scores. Though the closed book format of the tests might have been to the advantage of the multiple choice students, this evaluation method tends to remain strong competition for other assessments.

The impact of an educational innovation on students’ study strategies and the role of students’ perceptions of the learning environment.

Jan Nijhuis, University Maastricht, Netherlands

Mien Segers, University Leiden, Netherlands

Wim Gijselaers, University Maastricht, Netherlands

Many educational innovations in higher education aim to enhance student learning in terms of their study strategies. However, although much time and energy is invested and the design of the innovations seems to be sound, the results in terms of student learning do not always meet the expectations. The present paper explores the conditions for innovations to work, looking from the students' perspective. A so called Assignment-Based Learning (ABL) course was redesigned into a Problem-Based Learning (PBL) course, taking into account the results of studies on the effects of PBL and implementing instructional principles derived from constructivist propositions. The instructional as well as the assessment practices in the course were optimised. In both conditions (ABL and PBL) the students' study strategies (as expected by the students at the start of the course and as adopted during the course) as well as their perceptions of the learning environment were measured. The Biggs

Study Processes Questionnaire and the Ramsden Course Experiences Questionnaire, extended with the Scouller Perceptions of the Assessment Demands Questionnaire were used as instruments to measure these variables. The results indicate that students in the PBL condition (n=312) do not adopt more deep study strategies than the students in the ABL condition (n=406). Comparing the expected study strategies and those adopted, the results show a significant decrease in deep study strategies for both conditions. In the PBL course surface study strategies are significantly more adopted than students expected to do. The results of the logistic regression analysis indicate the importance of students' perceptions of various learning environment factors for their study strategies. It can be concluded that it seems that students' perceptions of the learning environment, including assessment, act as a filter between the learning environment as designed and student study strategies.

– 213 –

C 9 24 August 2005 08:30 - 10:30 Room A107

Symposium

Collaborative Learning

MEASUREMENT CHALLENGES IN COLLABORATIVE LEARNING RE-

SEARCH

Chair: Jan-Willem Strijbos, Leiden University, Netherlands

Frank Fischer, Knowledge Media Research Center, Germany

Organiser: Jan-Willem Strijbos, Leiden University, Netherlands

Frank Fischer, Knowledge Media Research Center, Germany

Discussant: Martin Valcke, Ghent University, Belgium

Gerry Stahl, Drexel University, United States

Collaborative learning is increasingly used at various educational levels and the use of computer-supported collaborative learning is growing. Given the current focus on so-called student-centred and self-regulated learning it is likely to increase in forthcoming years. In addition, the research has shifted from being solely outcome focused to the study of the process of collaboration: in other words the analyses of the communication - whether this is face-to-face or text-based. Yet, the study of collaborative learning has evolved into a multidisciplinary field (i.e., sociology, computer science, social psychology, communication studies) and provides a wealth of analysis approaches (quantitative content analysis, conversation analysis, multilevel modelling, cross case matrices, et cetera), however it is difficult to determine which approach could be used. Moreover, several studies reveal that the analysis is far from straightforward; including the development of an analysis procedure.

Furthermore, analysis methodologies cannot be applied to any type of data at any level of collaboration (i.e., small group, community of learners or community of practice) and since we want students to influence each others behaviour and perception some established analysis methodologies are not readily applicable anymore

(e.g., the assumption of independence - a requirement for ANOVA and regression analyses - is violated). Each type of collaboration requires specific considerations for the methodological approach to be used. In this symposium we aim to make the methodological considerations (which are often summarily dealt with in journal articles) more explicit. Especially the rationale for choosing and applying a specific approach to a specific data collection (each methodology will be illustrated with some of the research outcomes). Collaborative learning has been studied from

– 214 –

many different perspectives and it is time to combine them and extend the interdisciplinary character to research methodology.

Problem solving through chat: Beyond dyadic interaction.

Jan-willem strijbos, leiden university, netherlands

Gerry Stahl, Drexel University, United States

Computer-supported collaborative learning (CSCL) deploys several different communication technologies to support collaboration: chat is one of them. Thus far it has been primarily used during dyadic interaction. In the Virtual Math Teams'

(VMT) project small groups of three to five students collaborate through chat in a one hour session to solve a mathematical problem. As part of a multi-method analysis approach a coding scheme was devised to conduct content analysis to reveal differences between well and poor groups in mathematical problem solving. A first conceptualisation of the coding scheme consisted of four dimensions. Each chatline could be assigned one code from multiple dimensions: communication, social interaction, problem solving and mathematical reasoning. During the calibration several methodological problems emerged that thus far have not been reported in

CSCL and collaborative learning research. Firstly, the diversity of processes proved problematic. Initially the dimensions were assumed to be independent, but analyses revealed that ties between the dimensions could not be avoided. Secondly, the analysis required that the interaction structure (i.e., who responds to whom) was reconstructed. Obviously reconstruction needs to be conducted reliably. Also, not all chat-lines are eligible to receive a code in every dimension - further complicating analysis. Reconstructing the conversational thread proved to be crucial for the whole coding process. The mathematical dimension proved to be most difficult, because of the many subtle nuances involved. So far we are able to detect mathematical content, yet a specific typology requires additional work and conversation analysis might be more applicable to uncover this process. Yet the coding reveals a collaborative process evolving on multiple concurrent strands. The reliabilities for most dimensions proved to be satisfactory. The implications of the observed methodological issues for analysis of chat communication will be discussed.

– 215 –

What’s in the mix? Combining coding and conversation analysis to investigate chatbased problem solving

Alan Zemel, Drexel University, United States

Fatos Xhafa, Open University of Catalonia, Spain

The coding of interactional data for statistical analysis presents theoretical and practical challenges. Coding schemes that rely on categories that are decided by their relevance to the analytical problem under investigation carry with them the presumption that the analyst's perspective and concerns are privileged over the demonstrable and publicly displayed perspectives and concerns of the participants in the interaction. We suggest that 1) the endogenous and publicly displayed concerns of participants in interaction provide for the actual performance of the observed actions of participants and for the observable organization of that participation, and 2) these publicly observable features of the ongoing interaction properly constitute the appropriate data for analysis. Typically, such data has been examined using the methods of Conversation Analysis (CA). While CA can provide detailed descriptions of the methods actors use to organize their interaction in ways that are meaningful to the actors themselves, there has been very little statistical analysis of interactional data for the distributions of these methods or their consequences for ongoing interaction. In this paper, we explore a respecification of coding procedures for doing statistical analysis of interactional data in a manner consistent with a conversation analytic approach (Heritage & Roth, 1995). We will analytically identify coding categories that are consistent with and are constitutive of the production of the local order of interaction. By treating observed phenomena in terms of what they are designed to achieve at the point in the sequential unfolding of interaction where they occur, we will be able to capture the sequentiality of interaction. With this approach, we believe that the statistical analysis would be more parsimonious and produce a more precise description of group process and interaction.

Measuring knowledge convergence: Achievement similarity and shared knowledge in computer-supported collaborative learning

Armin Weinberger, Knowledge Media Research Center, Germany

Karsten Stegmann, Knowledge Media Research Center, Germany

Frank Fischer, Knowledge Media Research Center, Germany

Learning in small groups may result in convergent knowledge outcomes or foster possible prior differences between learners. Few studies, however, measure convergence or divergence of knowledge as an outcome of small group learning. This

– 216 –

contribution analyzes knowledge convergence/divergence as an outcome of learning in small groups with the concepts of achievement similarity and shared knowledge rooted in two different theoretical frameworks. Achievement similarity means that learners acquired similar amounts of knowledge regardless if they share similar knowledge. Shared knowledge means that learners acquired similar knowledge. In order to test application-oriented knowledge, learners had to individually apply concepts of a psychological theory in analyzing a problem case. Trained coders identified correct relations between theoretical concepts and case information. These data were used to compute both, achievement similarity and shared knowledge. A 2x2 factorial design was used (number of triads = 32, N = 96) to examine the following questions: (1) can theoretical concepts of achievement similarity and shared knowledge be measured independently and (2) to what extend are these measures of convergence sensitive enough to indicate effects of specific instructional interventions?

We examined effects of two instructional interventions on achievement similarity and shared knowledge. Achievement similarity was conceptualized as difference between the amount of known theoretical concepts of an individual learner and the average of his or her group. As a measure for achievement similarity, standard deviations of learners within one group were used to indicate dissimilarity and multiplied by -1 to indicate similarity. Shared knowledge was measured as number of pairs within one group, which applied the same theoretical concept in the individual analysis of the problem case. Results show that the different instructional interventions significantly and independently affect both convergence measures.

Validity and interpretability will be discussed against the background of theoretical approaches to learning in small groups.

The analysis of negotiation of common ground in CSCL

Pieter Jelle Beers, Open University of the Netherlands, Netherlands

Henny P. A. Boshuizen, Open University of the Netherlands, Netherlands

Paul A. Kirschner, Open University of the Netherlands, Netherlands

The recent growth in CSCL research has given rise to a plethora of analysis methods, all with specific analysis goals, specific units of analysis, and made for specific types of data (chat, threaded discussions, etc.). This paper describes the development of such an analysis method, with the ultimate aim of drawing general conclusions about CSCL-analysis. Special attention is paid to choices made and changes in those choices through the course of development of the analysis scheme, and its underlying assumptions. The analysis scheme reported on was developed as part of a research project about ìKnowledge sharing and knowledge building in expert

– 217 –

teams with ICT. This project involves the development of support for negotiation of common ground. A new coding scheme needed to be developed because no existing scheme addressed negotiation. The first step concerned operationalising negotiation with negotiation theory as starting point. This resulted in a set of codes that characterised the negotiation process in task communication. The second step involved development of a reliable measurement strategy, primarily aimed at assisting the assignment of codes to segments in ambiguous situations. This resulted, in the case of face-to-face communication, in a necessary lowering of the number of different codes so as to code reliably. Reliability was not a problem in electronic communication. The results show that the main challenge in analysing negotiation involved achieving reliability without compromising the original operationalisation of the research question. In general, it can be concluded that CSCL-analysis needs to take into account both validity of the coding scheme and the associated measurement strategy. Aiming for reliability can render a coding scheme invalid if one applies a measurement strategy without proper care.

Comparing analysis schemes and applying multilevel modelling: Two methodological issues in the study of the impact of role assignment in asynchronous discussion groups

Bram De Wever, Ghent University, Belgium

Hilde Van Keer, Ghent University, Belgium

Tammy Schellens, Ghent University, Belgium

Martin Valcke, Ghent University, Belgium

CSCL is often presented as a promising learning environment. To gain deeper understanding of its mechanisms, adequate research methodologies are however needed. This presentation focuses on two important methodological issues in this research area: (1) the use of adequate, reliable, and valid ways to analyse the discussion corpora, and (2) the importance of taking into account the hierarchically structured setting. The context for discussing both issues is a study conducted in the field of educational sciences with first year university students (n = 273). Asynchronous discussion groups were introduced to enhance collaborative learning. In order to explore students' sharing information, elaborating, negotiating, and consensus building, 10 discussion groups of approximately 10 students were randomly selected (n = 98). The presentation focuses first on content analysis as a way to uncover the information locked in the transcripts of asynchronous discussion groups. In the study, two content analysis instruments were employed: the model of Gunawardena and colleagues (1997), and the coding scheme of Veldhuis-Diermanse (2002). Both

– 218 –

instruments were selected on the basis of their social-constructivist background.

The motive to apply two analysis instruments was to explore potential relations between both approaches and to validate the schemes empirically. Whole messages were chosen as the unit of analysis and inter-rater reliability was checked. Secondly, to take into account that individual students are clustered into collaborative groups, we opted for a multilevel approach. Because data of students within a group cannot be considered as completely independent and thus the assumption of independency for using traditional analysis techniques is violated, we need to use this multilevel technique, since it takes the interdependency in groups explicitly into account. Moreover, it allows us to uncover the specific influence of, and cross-level interactions between student, group, and task variables.

Learning together: Combining individual and group-level perspectives for studying collaborative learning

Maarit Arvaja, University of Jyvaskyla, Finland

Hanna Salovaara, University of Oulu, Finland

Paivi Hakkinen, University of Jyvaskyla, Finland

Sanna Jarvela, University of Oulu, Finland

Collaborative learning has been critisized for neglecting the broader classroom context in which the collaboration is embedded. A specific challenge is to find ways to combine individual and group level perspectives for evaluating collaborative knowledge construction activity in groups, as well as building models for analysing the interactions that occur between individual and group level learning processes.

The aim of this paper is to identify conceptualizations and methods for studying collaboration in context and for understanding in detail how individuals, groups and contexts interact. In order to effectively study the impact of context on the collaboration at the individual and group level, process-oriented data will be used such as videotapes of students' activity, interviews and questionnaires. Furthermore, studying the knowledge construction at the group level should focus on analysis of the negotiation processes of the students. Linell's (1998) notion of contextual resources will be used as an analytical tool in studying how students negotiate meanings in their activity and what resources mediate that activity. Interviews and questionnaires will give insight into students' short-term impressions after meaningful activities or events and personal meanings students' attach to different activities. Results obtained form several groups during a teacher education course on educational psychology (N = 150), under face-to-face or computer-mediated conditions, will be used to illustrate the applicability of the methods that were selected to reveal

– 219 –

the mediating influence of context on collaborative activity. The specific focus of this paper is to illustrate the methods used to identify the influence of context on students' collaborative activities and their applicability to reveal the interaction between individual and group level perspectives, ultimately generating information of individual students' interpretations of their collaboration to identify aspects that support or hinder successful collaboration.

C 10 24 August 2005 08:30 - 10:30 Room E004

Symposium

Qualitative and Quantitative Approaches to Learning and Instruction

DEVELOPMENT OF ARGUMENTATION COMPETENCIES IN THE

MATHEMATICS CLASSROOM

Chair: Kristina Reiss, University of Augsburg, Germany

Fou-Lai Lin, National Taiwan Normal University, Taiwan

Organiser: Kristina Reiss, University of Augsburg, Germany

Fou-Lai Lin, National Taiwan Normal University, Taiwan

Discussant: Bracha Kramarski, Bar-Illan University, Israel

Reasoning correctly and arguing coherently may be regarded important goals of mathematics education (NCTM, 2000), although it is a well-known fact that many students are facing difficulties when they learn mathematical reasoning, argumentation, and proof. Within the last decade several international studies on mathematics achievement like TIMSS or PISA and a number of specific studies on students' proof competencies have shown that arguing mathematically and proving are difficult mathematical activities for students to perform. However, some of these investigations give reason to assume that there is a particular influence of the specific mathematics classroom. Moreover, these findings suggest that it might be possible to create learning environments for students in order to foster their proving abilities.

The participants of the symposium will discuss this topic in some detail. Relying on the results of empirical studies with secondary school students on argumentation and mathematical proof in the different countries (Germany, Taiwan, United Kingdom) conducted by the participants, the discussion will focus on the development of argumentation skills in the mathematics classroom and on possible interventions that might help students learn a coherent mathematical argumentation. This discussion will be supported by the presentation of results from quantitative and qualita-

– 220 –

tive research studies. We will argue from a perspective where learning is regarded a constructive process that can be initiated and supported by the teacher. The symposium aims at describing aspects of the development of argumentation skills and at identifying adequate learning environments with respect to cultural specifics of the mathematics classroom.

One more step toward acceptable proof in geometry

Ying-Hao Cheng, Chung Kuo Institute of Technology, Taiwan

Fou-Lai Lin, National Taiwan Normal University, Taiwan

The results of a national survey in December 2002 show that there are one quarter of our 9th grade of students, they had just learnt formal proof in geometry lessons, can construct an acceptable proof in a 2-step unfamiliar question. More than one third of them do not response to this question. And about one third of them may recognize some crucial elements and prove this question but lack the process of concluding or miss one step in proving, we call this type of proof as incomplete. These students seem to have a gap to construct an acceptable proof. There are two different modes of incomplete proof. One is lacking the process of concluding. It is only ritually incomplete because all the key elements of proof are included and correct.

The other is missing one step in proving. This mode of performance means that the students cannot recognize all the necessary information from the premise. It may be due to the invisible of literal information because students always find the cue of proof on figures. It may also be due to the lack of finding useful information from the premise. An interview was conducted with 12 grade 9 students whose proofs were incomplete in order to overcome their obstacles and help them to construct an acceptable proof. There were two main strategies used in our interview based on the idea of visualizing the implicit information. One is picking out the key terms and drawing on figures, the other is setting goal by one step backward reasoning. The results show that the strategies were effective

Students' development in geometrical reasoning: Insights from a large-scale longitudinal survey

Celia Hoyles, University of London, United Kingdom

Dietmar Kuechemann, University of London, United Kingdom

We report some results from a longitudinal study of mathematical reasoning among a nationwide large sample (n=1512) of high-attaining students in England, from age 131/2 years to age 151/2 years. Using a mixture of quantitative and qualita-

– 221 –

tive methods, we report clear progress in reasoning in response to standard items in algebra and geometry; however, on less standard items, and ones requiring an explanation, a sizeable minority of students persist in basing their explanations on perception in geometry or on numerical evidence in algebra, with at best modest progress in the construction of mathematical explanations along with some regression. Finally robust gender differences were identified along with subtle influences of curriculum and school organisation.

Reasoning and proof in geometry: Effects of a learning environment based on heuristic worked-out examples

Kristina Reiss, University of Augsburg, Germany

Aiso Heinze, University of Augsburg, Germany

Christian Gross, University of Augsburg, Germany

In this presentation, we argue that heuristic worked-out examples are an adequate learning environment for mathematical argumentation and proof. It enhances the traditional worked-out examples that turned out to be efficient for the learning of algorithmic problem solving. The basic idea of heuristic worked-out examples is providing students with opportunities to explore, recognize, and use explicitly different phases in the process of performing a proof. The results of an intervention study with 243 students from grade 8 suggest that this learning environment might be more efficient than regular mathematics instruction on proof. The ability to give correct mathematical argumentation and to generate a proof for a specific hypothesis is based on several aspects like knowing mathematical concepts, knowing heuristic strategies and being able to use them, having metacognitive control strategies available, and understanding the nature of mathematical proof. Empirical studies indicate that students lack one or more of these facets of proof. In this study we investigated to what extent learning to prove can be fostered by heuristic worked-out examples. We addressed the question, whether this learning environment is of special use for low-achieving or high-achieving students. Comparing the mean posttest scores of the experimental and the control group we found a significant difference: the experimental group performed much better in the posttest than the control Summarizing the analysis of the posttest results the experimental group performed significantly better than the control group. Moreover, students from the experimental group performed significantly better on complex proofs. We found a positive effect on the achieve-ment of students learning with heuristic worked-out examples (in comparison to a control group receiving traditional mathe-matics instruction).

– 222 –

A model of reading comprehension on geometry proof

Kai-Lin Yang, Chung Yuan Christian University, Taiwan

Fou-Lai Lin, National Taiwan Normal University, Taiwan

Research on the learning and teaching of mathematical proofs mainly focused on constructing mathematical proofs. However, we approached to the learning of mathematics proof with the perspective of reading comprehension. One purpose, to explore the characteristics of students' reading comprehension on geometry proof, was proposed in this study. Herein, reading comprehension means what was grasped and inferred while reading a text. From theoretical analysis, a framework of the levels of reading comprehension on geometry proof was formulated. From empirical exploration, the assessment instrument was developed to identify the level of each student's reading comprehension on geometry proof, and students' cognitive and thinking characteristics in each level were explored through interviews. In sum, a model was constructed to describe what the levels of reading comprehension on geometry proof were and to explain why it is difficult to advance students' ìreading comprehension level in this research.

C 11 24 August 2005 08:30 - 10:30 Room A010

Symposium

Teacher Professional Development

THE SCHOLARSHIP OF RESEARCH AND SCHOLARSHIP OF TEACH-

ING IN SCIENCE AND ENGINEERING

Chair: Brandon Reed, University of Cape Town, South Africa

Organiser: Ake Ingerman, Chalmers University of Technology, Sweden

Discussant: Keith Trigwell, University of Oxford, United Kingdom

Lee Shulman, in his capacity as president of the Carnegie Foundation for the Advancement of Teaching, recently wrote the following in the conclusion to an article entitled: Opening Lines: Approaches to the scholarship of teaching:

ìIn modern times, we regularly distinguish between two kinds of method: the methods we use in our research, on the one hand, and our methods of teaching on the other. In the older traditions of the university, however, these two aspects of method converged (or were never separated). The methods of scholarship and the methods of teaching were identical; ones ìmethods were those strategies used to

– 223 –

marshal evidence in a scholarly and persuasive manner for instructing one's students. Both pedagogical and scholarly arguments involved warrant (evidence) and explanation, in a persuasive rhetorical form. It is ironic that the two have not only drifted apart; they are seen as competitive.In this spirit this symposium explores the nature of the research and teaching context experienced in science and engineering.

Ake Ingerman & Shirley Booth describe how physicists relate to the trustworthiness of their own research and that of their peers. This furthers the understanding of the ground for the collaboration, peer review and communication in the research community. Liselott Dominicus, Rebecca Lippman Kung, Cedric Linder & Delia Marshall describe the nature of the relationship between how physics lecturers craft their practice and the kinds of perceptions their students have developed about what constitutes quality teaching in physics. Lotta Antman & Thomas Olsson have investigated peer evaluation of scholarly approaches to teaching and use this to discuss the relationship of scholarship of teaching to perceived teaching expertise, providing rich exemplars of how a scholarly approach to teaching in practice is and could be, supported.

Trusting research: An exploration of physicists' conceptions of the trustworthiness of their own and others' research

Ake Ingerman, Chalmers University of Technology, Sweden

Shirley Booth, Lund University, Sweden

Research is, also in science and engineering, done in collaboration with, in relation to and reviewed and collated by peers. To act and work in that world, the researcher have continually to make judgements of the quality, appeal, usefulness and how good their own as well as their peer's research outcome is. This presentation focuses on how researchers in condensed matter physics make judgments, or conceptualize the making of judgments, of the trustworthiness of their and others' research results. Through an analysis informed by phenomenography, four different aspects of the complex phenomenon trustworthiness are brought out, in which dimensions of variation are identified and explored. The four different aspects are: the object of trust, the anchor for trust, the context for trust, and the relation between the object and the anchor. The relations between the ways in which the different aspects are handled are explored and the establishment of trustworthiness is related to participating in a community of practice. Judging trustworthiness as related to both content and social aspects is considered in relation to the peer review process and to the development of PhD students from newcomers to full-fledged researchers.

Crafting of teaching practice and physics students' perceptions of quality teaching

– 224 –

Liselott Dominicus, Uppsala University, Sweden

Cedric Linder, Uppsala University, Sweden

Rebecca Lippmann Kung, Uppsala University, Sweden

Delia Marshall, University of the Western Cape, South Africa

This study explores the relationship between university physics lecturers crafting of their teaching practice and their students perceptions about what constitutes quality teaching. The work described is limited to introductory and intermediate undergraduate teaching and learning in three university physics departments, two in

Sweden and one in South Africa. Physics lecturers and their students, who were involved in different physics courses, were selected for the study on the basis of being able to distinctly identify qualitatively different characterizations of crafting-of-practice for the participating lecturers. This selection was based upon the results of interviews with lecturers in the three participating physics departments.

Then, a selection of students taking these courses were given questionnaires and a further selection interviewed about their perceptions of quality physics teaching.

Drawing on a phenomenographic perspective categories were developed from the interview transcripts in order to capture the variation in perception. The interviews were then collated by course lecturer and sorted in terms of the categories we had developed. The results were captured in terms of sets of data clustering across the perception categories. These clusters were then examined in terms of the lecturers' crafting-of-practice characterizations. The outcomes provide an illustration of how the nature of the crafting-of-practice can significantly influence how students come to think about quality teaching and in turn potentially influence their own development as learners.

Peer review of scholarly approaches to teaching: The case of the pedagogical academy

Lotta Antman, Lund University, Sweden

Thomas Olsson, Lund Institute of Technology, Sweden

The Pedagogical Academy is a model for rewarding excellent teaching, developed at Lund Institute of Technology. A phenomenographic approach was used to study the procedure of reviewing and assessing scholarly approaches to teaching, in the case of the Pedagogical Academy, integrating the analyses of documents, videorecorded observations and interviews. The assessment process was peer-reviewed in the sense that previously awarded and rewarded teachers acted as reviewers.

The results indicated that the group of reviewers worked with three tacit levels of

– 225 –

knowledge, a notion which was problematised and expanded in a matrix model, opening up two complementary knowledge dimensions. The variation in focus was analysed in terms of thematised validity claims with respect to their effect on the assessment procedure as a whole and on the results. In the peer-reviewed processan ostensibly scholarly approach to assessing levels of Scholarship of Teaching and Learning among applicants-issues of power, social status and tradition were nevertheless seen to set the stage.

C 12 24 August 2005 08:30 - 10:30 Room E102

Symposium

Mathematics Education

CONCEPTUAL UNDERSTANDING AND PROCEDURAL SKILL IN

MATHEMATICS LEARNING

Chair: Camilla Gilmore, University of Oxford, United Kingdom

Michael Schneider, Max Planck Institute for Human Development,

Germany

Organiser: Michael Schneider, Max Planck Institute for Human Development,

Germany

Camilla Gilmore, University of Oxford, United Kingdom

Discussant: Lieven Verschaffel, Catholic University of Leuven, Belgium

Several theories of learning and cognition pose that our behaviour is shaped by at least two different kinds of knowledge: one providing abstract understanding of the principles and relations between pieces of knowledge in a domain, and another one enabling us to quickly and efficiently solve problems. In recent empirical research on mathematics learning the former is frequently named conceptual knowledge or understanding, while the latter is labelled procedural knowledge or skill.

Models about these different kinds of knowledge and their interrelations may be helpful for designing the contexts in which knowledge is to be conveyed. They could be used for a theory-guided construction of learning environments and curricula. There seem to be, however, more questions than answers concerning conceptual and procedural knowledge. Their specific characteristics, their interrelations, and their relations to other psychological constructs are controversial.

Despite these controversies and despite the importance of the field, comparatively few empirical studies have addressed the relationships between conceptual under-

– 226 –

standing and procedural skill. This symposium, therefore, is to give an overview over different empirical works in the field. While some of the participants use theory-derived measures for the kinds of knowledge, others evaluate the knowledge demands of different types of tasks by factor analyses or expert-ratings. Some focus on processes that spread over years, e.g. by comparing different age groups, others on processes that may take only hours and can be observed in training studies.

Some study task specific knowledge, others concentrate on broader and more stable constructs like mathematical knowledge or quantitative reasoning skills.

The discussion of the studies will not only focus on empirical results, but also on researchers' experiences with their different designs and methodologies, and will try to work out promising directions for future research.

Children's Understanding of Addition and Subtraction

Katherine Canobi, University of Melbourne, Australia

The research addressed Children's understanding of addition and subtraction by identifying different patterns of performance on various conceptual indices. Investigating patterns of conceptual understanding provides a strong basis for examining the role of (a) different principles, (b) physical referents, (c) types of activities, (d) age-related changes, and (e) conceptual-procedural interactions in Children's addition and subtraction development. In Study 1, 72 6- to 9-year olds judged a puppet's activities involving 3 conceptual relations: (a) a + b = c, b + a = c; (b) a - b = c, a c

= b; and (c) a + b = c, c - b = a and explained their judgements. In Study 2, 60 5- to

8-year olds completed a Guessing task and a Problem-solving Task comprising 3term inverse problems (i.e., a + b - b =?) and pairs of complementary addition and subtraction problems (i.e., a + b = c, c - a =?) along with control problems. In the

Guessing task, children were encouraged to make ìgood guesses about the answers to problems. In the Problem-solving task, children attempted to solve the inverse and control problems. Distinct conceptual profiles were identified in both studies and were related to Children's problem-solving skills (speed, accuracy and self-reported procedures on unrelated problems). The results suggest that many children find the inversion principle relatively difficult although performance differs among children and across tasks. The results also suggest that the usefulness of physical referents for illustrating concepts varies across individuals, concepts and tasks.

The findings highlight the importance of exploring variation among individuals and across tasks for developing effective learning environments. They underscore the complexity and functional significance of understanding the structure of addition and subtraction in Children's mathematical development and suggest that there are

– 227 –

different pathways to conceptual knowledge.

The results of a new mathematical reasoning task given to a large number of 8-year old children

Peter Bryant, Oxford Brookes University, United Kingdom

Terezinha Nunes, Oxford Brookes University, United Kingdom

A new Mathematical Reasoning task, which assesses reasoning and poses very little demands on Children's computational skills, was administered to over 2,000 8year-old children as part of a large-scale longitudinal, epidemiological study that is being carried out in the West of England. The task is designed to measure Children's ability to reason additively and multiplicatively. The scores in this task were related to the same Children's IQ scores, to their performance in a standardised arithmetic task and in working memory tasks, and also to their school achievement scores

(SATS). Our preliminary analyses show that:

1. the Mathematical Reasoning task had a higher correlation than any of the other measures with school achievement in mathematics (SATS).

2. general intelligence, computation skill, and mathematical reasoning make specific contributions to explaining variance in SATS-Maths, and together explain 46% of the variance;

3. working memory is not a strong predictor of SATS-Maths after general intelligence has been entered in the equation: the correlation between working memory and SATS-Maths is low.

We conclude that:

- General intelligence does not account for Children's mathematics achievement by itself. Learned abilities, such as computation skills and mathematical reasoning, play a significant part in explaining variance in mathematical competence.

- Working memory has a negligible role in explaining variance in mathematics achievement in the population as a whole.

- To improve mathematics achievement, it is necessary to invest in developing Children's mathematical reasoning just as it is necessary to invest in teaching computation skills.

– 228 –

Who can judge reasoning and knowledge demands of mathematics problems?

Martin Brunner, Max Planck Institute for Human Development, Berlin, Germany

Mareike Kunter, Max Planck Institute for Human Development, Berlin, Germany

Stefan Krauss, Max Planck Institute for Human Development, Berlin, Germany

Subscores on student achievement tests in mathematics are often interpreted as reflecting (static) mathematical knowledge or some kind of (procedural) quantitative reasoning ability. In many cases, this interpretation is based on expert analyses of task characteristics. In psychometric research on cognitive abilities, the reasoning and knowledge demands of test items are represented by factor loadings on corresponding latent variables. The aim of this study is to compare expert judgments of cognitive task demands with factor loadings indicative of those demands. Three groups of experts (mathematics educationalists, psychometric researchers on intelligence, experts on German mathematics curricula) rated 117 mathematics tasks with regard to characteristics that are theoretically related to reasoning and knowledge demands. By means of confirmatory item factor analyses and using additional reasoning items, a model with a reasoning and a mathematical knowledge factor was specified and corresponding factor loadings were estimated. Performance data were collected from 29,386 ninth-graders attending all academic tracks in Germany. The main result was that correlations between expert ratings and factor loadings were almost negligible, indicating that expert ratings may relate to constructs other than psychometrically defined abilities or knowledge. With regard to educational practice, it might thus be helpful to bear in mind that tasks recommended by experts to focus on reasoning abilities or mathematical knowledge do not measure those demands in a psychometric sense. The implications of these findings for test construction are discussed.

The relationship between children's understanding of inversion and arithmetic skill

Camilla Gilmore, University of Oxford, United Kingdom

Children must understand the concepts underlying arithmetic as well as learning how to perform the operations. One key principle children must learn is the inverse relationship between addition and subtraction. A direct test of Children's understanding of this concept compares their performance on a set of standard arithmetic problems (a+b-c=) and a set of problems which can be solved directly by making use of the inverse principle (a+b-b=). Children's flexibility in using inversion was examined by varying the format of the inverse problems and the position of the

– 229 –

missing number (e.g. b-b+a= ;?+b-b=a). Children were more accurate on the inverse than standard problems, showing they were able to apply the inverse principle, but this was affected by the order of elements in the problem and the position of the missing number. The relationship between Children's understanding of inversion and their calculation skills was examined using cluster analysis. Three distinct profiles of performance were observed: one group of children with good conceptual understanding and calculation skills; one group with poor conceptual understanding and calculation skills; and a final group with good conceptual understanding but poor calculation skills. This suggests that these aspects of arithmetic may be somewhat independent and so it is important to consider different components of mathematics ability in both research and educational settings.

The measurement of children's conceptual and procedural knowledge about a mathematics problem: Findings from confirmatory factor analyses

Michael Schneider, Max Planck Institute for Human Development, Berlin, Germany

Different causal relationships are proposed between conceptual and procedural knowledge. Some theories suggest uni-directional links, others bi-directional ones.

Because of their implications for learning theory and classroom instruction, e.g. task sequencing, these positions should be compared empirically. However, empirical research on these questions is hampered by the lack of knowledge on how the two kinds of knowledge can be measured with a sufficient degree of validity and independent of each other. The present study was designed to examine this point by means of confirmatory factor analyses (CFA). Conceptual and procedural knowledge about decimal fractions of 231 fifth- and sixth-graders was assessed by eight different measures before and after an intervention. The measures were adopted from studies in other mathematical domains, where four of them (evaluation of procedures, translations into diagrams, size comparisons, written explanations) had been used to access conceptual knowledge, while the other four (problem solving correctness, problem solving duration, asymmetry of access, dual-task interference) served to access procedural knowledge. CFA for the pre-test data revealed that two-factor models, i.e. one factor reflecting conceptual knowledge and another one reflecting procedural knowledge, were not significantly superior to a one-factor model in terms of model fit. This indicates a low divergent validity of the eight measures. The convergent validity, estimated by Cronbach's Alpha coefficients, was acceptable for the measures of conceptual knowledge (a = .71), but not for the measures of procedural knowledge (a = -.08). Similar results were found for the

– 230 –

post-test data. These findings highlight the difficulty of measuring conceptual and, especially, procedural knowledge and suggest that results from previous studies should be interpreted with caution. The generalizability of the present findings and their methodological and theoretical implications are discussed.

C 13 24 August 2005 08:30 - 10:30 Room E112

Symposium

Learning and Instructional Technology

SUPPORTING EFFECTIVE LEARNING IN VIRTUAL ENVIRONMENTS

Chair: Karen Littleton, The Open University, United Kingdom

Organiser: Linda Price, The Open University, United Kingdom

Discussant: Eleni A. Kyza, University of Cyprus, Cyprus

Educational programmes are making increasing use of Information and Communication Technologies (ICT) and there is an assumption that the technology itself is the agent of change. However some educational programmes do not take account of how to constructively align the technology with the pedagogical aims, objectives and assessment strategy of the course and this can render less valuable uses of technology and inferior student experiences. Further the design on virtual environments often second guess the student experience and in what way the use of ICT may assist their learning. In this symposium we shall consider how students can be supported by using the technology in both a transformative way and a facilitative way.

Transformative, in terms of why and how technology could and should be used in order to enhance the student learning experience students; where the technology and pedagogy are constructively aligned. Facilitative, in the sense that students are provided with an opportunity that they would not otherwise have had. Within this framework we will examine how we can effectively support the student learning in their particular virtual environment and circumstances. The research illustrates that understanding the student experience from the student's perspective is fundamental to providing an effective virtual environment whether this be from a transformative or facilitative perspective.

– 231 –

Independent learners' conceptions and perceptions of tutorial support in on-line and person to person contexts

Linda Price, The Open University, United Kingdom

On-line tutoring is a means of providing flexible learning support. However there is a concern that in some circumstances students may perceive on-line tutoring as less valuable than face-to-face and telephone tutoring. In a previous study Richardson

& Price (2003) investigated students' perceptions of quality in a distance education course that had two tutoring modes: face-to-face tutoring and on-line tutoring. The two modes were comparable on all scales of the Course Experience Questionnaire

(measuring perceived quality of the course), except Good Tutoring. The students receiving on-line tutoring gave lower ratings to the quality of tutoring than did the students receiving face-to-face support.

To investigate the reasons for these findings we undertook a further study. Semistructured interviews were used to elicit students' beliefs. These were conducted using the virtual research method of epistolary interviews being developed at the

Open University. Student's conceptions and perceptions of tutoring were compared and contrasted in order to understand why on-line tutoring is perceived as poorer and what role on-line tutoring might play in supporting student learning.

The groups did not differ in their conceptions of learning but they did differ in their perceptions of the tutorial support that they had received. Students in the on-line group had expectations that were not met on-line. However, on-line tutoring did provide some students with an option to study that otherwise would not have been available. Hence in this situation the role of ICT was not transformative but facilitative, providing an educational opportunity that would otherwise not have existed.

Pedagogic integration and learners' use of electronic resources

Adrian Kirkwood, The Open University, United Kingdom

Electronic (on-line) resources are progressively being introduced within HE courses and/or provided by university libraries. The Internet enables students to access a range of information sources - including bibliographic databases, on-line journals and other archives and collections - from wherever they are located. Access to these resources is particularly valuable for dispersed independent learners, many of whom are unable to study in an academic or specialist library. However, research evidence from studies of on-campus and distance education students indicates that there are often disappointing levels of use by students of recommended on-line resources and of the information infrastructure developed to support teaching and learning

– 232 –

in UK higher education. This presentation reports a study that aimed to investigate how and why independent learners use on-line resources while undertaking their normal undergraduate coursework. The investigation was concerned not only with the academic context of the course(s) being studied, but also any personal, domestic and employment-related experiences and circumstances that were pertinent. The study built upon quantitative data collected from large-scale surveys. The methods and outcomes of the research will be summarised and evidence presented to support the need for closer integration of on-line resources within course pedagogy.

Students are not averse to using the WWW to find information that they feel will be of benefit to them, particularly in relation to assessed work. They will make use of on-line resources that can be seen to contribute to the achievement of core course and/or personal outcomes to a greater extent than those that appear to be peripheral or optional. However, students often lack appropriate information literacy skills to make effective use if what is available via the Internet.

Repurposing wisdom - applying ‘grounded guidelines' for effective e-tutoring in post-compulsory education

Erica McAteer, University of Strathclyde, United Kingdom

This presentation draws on research funded by the UK Joint Information Systems

Committee (JISC) in 1998, involving a substantive review of research and practice literature, followed by fieldwork undertaken across a range of subject disciplines, where learners and teachers were engaged in what were then thought of as cutting edge uses of communication and information technologies for learning. The research itself is summarised, and evidence is provided to support the rationale for conversion of its dissemination strategy to provide accessible practice guidelines which retain their grounding in the illuminative data from fieldwork and practice literature. The resource itself will be illustrated. The evaluation of its application to practice, conducted with the support of practice networks across the UK, is in a sense another reality check of what are, now, emergent issues of concern for learners and teachers in e-learning environments using CMC as the medium for communication.

Using on-line resources to support students learning physics in Irish secondary schools

Jen Harvey, Dublin Institute of Technology, Eire

As the number of online resources available to support student learning has in-

– 233 –

creased so to have the mechanisms enabling faster access to these materials. However, while individual teachers might make use of subject portals or gateways, there is still a tendency to create individual collections of resources, even if this remains a time consuming activity to undertake. This presentation explores changes in practice observed as part of an ongoing project where a group of teachers have worked together to create a portal of resources tailored to support the needs of the Irish secondary school Physics curriculum. Potentially, the effective use of such resources can both encourage and enable students to engage more with the topic and, in turn, develop a deeper conceptual understanding of the subject area. For the teachers, the speed of access to an appropriate resource for use with students is a priority. Therefore the development of a robust and effective way of evaluating potential resources and materials has been a key element to the project portal development process. The group reports a sense of ownership over the process and a strong feeling of community has developed. As the resource portal has evolved so has the way in which the resources are being used and reviewed. Teachers use and promote the site because the content has been developed and reviewed by a select group of their peers. They also remain actively involved with the community as a means of exchanging ideas and changing practices. The network links to many of the schools have greatly improved and within this current year all schools are expected to have broadband access. This influences the way in which the resources are evaluated, stored and then used. A case study approach is used to explore some of these emergent and potential changes.

Case study approach to evaluating student support in on-line environments

Anne Jelfs, The Open University, United Kingdom

This paper and presentation is based on the premise of supporting students using technology in a transformative and facilitative way. This paper focuses on the facilitating of students in their academic careers with the Open University in the UK

(OU) through the provision of a series of on-line materials. Understanding how effective and useful on-line student support materials are to distance/flexible students requires a strategic evaluation of the provision. This paper refers to a case study of the educational evaluation of on-line resources where the aim is to improve the quality and effectiveness of teaching and learning. Particularly where the resources are mediated through and supported by information and communication technologies (ICTs) for on-line study support. A range of approaches and tools can be drawn on, and I see the role as that of an expert informing and illuminating decision making. The qualitative aspects of the evaluation reported in this paper found that the

– 234 –

main influence on use of the website was the labelling and navigation. A number of the labels were too similar to allow students to differentiate and anticipate which topics were included in specific areas. There needed to be a strategic philosophy on how the pages were to be presented, so that students did not have to learn to navigate each area separately. On the positive side, students thought the content was very valuable and would use it in their studies as it did facilitate their approaches to studying. The case study evaluated the use of on-line resources to support and scaffold students working at a distance. The aim of the paper is to encourage further discussion of provision that can make on-line support materials more facilitative and engaging.

C 14 24 August 2005 08:30 - 10:30 Room E005

Symposium

Assessment and Evaluation

DEVELOPMENT OF CROSS-CURRICULAR COMPETENCIES IN DIF-

FERENT EDUCATIONAL CONTEXTS

Chair: Beno Csapo, University of Szeged, Hungary

Organiser: Beno Csapo, University of Szeged, Hungary

Discussant: Philip Adey, Kings College, London, United Kingdom

The main goal of this symposium is to bring together a collection of papers that examine the development of cognitive competencies and the role they play in education from a variety of perspectives. Improving students' general, broadly applicable, transferable cognitive competencies has been one of the major goals of education systems for several decades, but since life long learning became a reality, the related issues should receive even greater attention. The development of cognitive skills determines students' school career in a number of different ways, from the success in learning of the particular subjects to the selection processes taking places during the transition between grades. The papers present research related to several cognitive skills, including Piagetian operations, propositional logic, scientific reasoning, critical reasoning, inductive reasoning, deductive reasoning, and argumentation skills. The age of the students examined in the studies presented in this symposium ranges from the primary grades to the higher education. A variety of methods of research will be presented, compared and discussed including ethnographic techniques, interviews, discourse analysis, large-scale cross-sectional

– 235 –

and longitudinal assessment, and computer simulation. The first paper (Wegerif at al.) reports on a study of dialogues in several short online courses. The second one (Vidakovich) investigates the relationships between the development of logical operations and deductive patterns. The third paper (Seppala) compares the development of scientific thinking at two types of higher education institutions. Finally, a computer simulation will be presented (Csapo), where a database was simulated by using data of large-scale assessment project in order to study the selection processes of a school-system. Comparing and contrasting these for different approaches may stimulate an active discourse on the status of competencies in educational research and practice.

Investigating the social dimension of induction into argumentation

Rupert Wegerif, University of Southampton, United Kingdom

Simon McAlister, London Metropolitan University, United Kingdom

Andrew Ravenscroft, London Metropolitan University, United Kingdom

This paper reports on a study of synchronous dialogues on several short online courses teaching the core skill of critical reasoning in a range of educational contexts. The focus of this paper is the way in which we combined several methods into a coherent and principled methodology for researching the induction into argumentation online. We used a flexible dialogue support tool as an experimental device, or lens, through which to investigate the interplay of social and cognitive dimensions of online educational dialogue. The tool provides templates (e.g sentence openers), frames and advice shaping online interactions and so enables researchers to explore the impact of different interaction rules. A combination of qualitative interpretation and the use of concordance software was used to facilitate the discourse analysis. Finally ethnographic techniques of participant observation and open-ended interviews were adopted to explore the experience of participants online. Through the use of these methods it emerged the supports offered by dialogue tools, sentence openers for example, did not work independently of social relationships but through the way in which they entered into and mediated those relationships. This paper reports on the development and testing of a new methodology combining and adapting the best of what has been learnt from the analysis of face-to-face discussions with some recent techniques from applied linguistics to understand the experience of learning in a virtual environment. Through this we explore the distinctive affordances that online environments have for the induction of learners into argumentation.

– 236 –

Differences in the development of logical operations and deductive patterns of 10-

16-year-olds

Tibor Vidakovich, University of Szeged, Hungary

The aims of the study were: to construct tests for the evaluation of seven basic operations of propositional logic, and deductive patterns with these operations as premises; to diagnose the status of these operations and deductive patterns in four age groups between 10 and 16 years; to reveal significant differences and characteristic tendencies of the development between the age groups; and to examine whether the development of operations influences significantly the development of deductive patterns. On the basis of the selected logical operations a task system was constructed. All operations were evaluated in two ways. First as operations in complex statements, then in deductive patterns, where one of the premises was a complex statement containing the operation. All tasks were formulated with familiar contents, evoking everyday situations. The tests were administered to a sample of more than 8500 students; 4th, 6th, 8th and 10th graders, the sample was representative for Hungary by settlement and school types (academic, technical and vocational). In the operational tasks, improvements were significant between the age groups, but standard deviations showed a growing tendency. In the primary school the disjunctions, in the secondary school the conditionals showed faster improvement. In the tasks of deductive patterns, differences between the age groups proved to be significant for most tasks, but there were no unambigous correlations between the results of logical operations and deductive patterns. According to our results, significant development of logical operations takes place mostly before adolescence, and development slows down later. The development of deductive patterns did not show similar characteristic tendency. The hypothesis that the development of basic operations influences performances in deductive patterns can not be proved on the basis of our results. The effects of school, especially schooling streams are significant. These results demonstrate the selection mechanism of the Hungarian school system.

Students' scientific reasoning skills in polytechnics and universities in the field of business education

Hannele Seppala, University of Helsinki, Finland

The aim of the research was to investigate students' scientific reasoning skills in the initial, intermediate and final phases of the university and polytechnic studies in business education. Scientific reasoning was defined in the research as hypotheti-

– 237 –

cal-deductive reasoning, which includes the ability to identify causality between variables (possible in the level of formal operations). The level of formal operations includes also the abilities to reflect or to operate on the concrete operations of classification and relation. (Inhelder & Piaget, 1958) In addition - as another part of scientific reasoning - students' metacognitive awareness of the formal operational schema was investigated in the research. Two measures of formal reasoning were used to find out students' scientific reasoning skills: the Pendulum and the Chemical tasks (Science Reasoning Tasks: Shayer, Wylam, Kuchemann & Adey, 1978). 338 business major students from five polytechnics and four universities participated in the study. In the research the university students showed significantly higher formal reasoning abilities than polytechnic students. The results of the Chemical task indicated that there were also significant differences between the higher education sector and the phase of studies. In the initial and intermediate phases of studies in the polytechnics and in the universities the development of formal reasoning corresponded each other, but in the final phase of studies the results of the two sectors differed from each other: the polytechnic students in the final phase of studies showed lower reasoning abilities than the students in the initial and intermediate phases. The research will continue with investigating the connections between the students' formal reasoning abilities and the scientific and vocational orientations in polytechnic and university studies.

Stability and change in the development of general cognitive skills: An analysis of data of large-scale cross-sectional and longitudinal studies

Beno Csapo, University of Szeged, Hungary

This paper presents a model to overcome the difficulties of long-term longitudinal assessments by combining data form different sources and it aims to give reliable answers to questions raised in educational practice. The data used for the analyses were obtained from three different sources: (1) inter-age correlation data form long-term longitudinal studies cited in the literature; (2) data of large-scale crosssectional assessments of Hungarian school-aged population; and (3) data of shortterm longitudinal studies concerning the same population. Sample sizes of the two latter data collection were N>1500 per cohort. The age difference between the cross sectional samples was two years (students of 3rd, 5th, 7th, 9th, and 11th grades); similarly, there was two years difference between the two measurement points in the longitudinal assessment. (A younger sample was assessed when students were in grade 6 and 8, and an older sample when they were in grade 10 and 12). Means and standard deviations were computed from the large-scale assessments, while

– 238 –

inter-age correlations were obtained from the literature and short-term longitudinal studies. These data were used to generate a simulated database that reflects all attributes of the development known from empirical studies. In this paper, the case of inductive reasoning - one of the most important general skills determining school success - will be shown in an analysis that estimated how accurate school selection may be if it takes place at the end of grades 4, 6, 8, and 10. Results show that if 10% of the most advanced students are selected for an academic track in these grades, only 3.9%, 5.2%, 7.1%, and 8.9% will be found in the same upper achievement range at the end of 12th grade. The paper concludes that because of the flexibility of cognitive development no reliable basis may be established for early selection.

C 15 24 August 2005 08:30 - 10:30 Room A108

Symposium

Reading

EARLY READING DEVELOPMENT: LOCAL MODELS AND CROSS-LIN-

GUISTIC COMPARISONS

Chair: Timothy Papadopoulos, University of Cyprus, Cyprus

Organiser: Timothy Papadopoulos, University of Cyprus, Cyprus

Discussant: Irianna Diakidou, University of Cyprus, Cyprus

For many years, the development of theories about the way children learn to read was relied extensively on studies of English-speaking populations. However, as reading research accumulates evidence from other languages as well, we learn more about how children learn to read in other scripts, noticeably different from English. Greek, for instance, is unusual in having a unique mapping between letters and phonemes making it distinctly more transparent than English for both reading and spelling. This, in turn, makes both processes more accurate, even for the least capable readers and/or spellers. This symposium aims to examine and provide some preliminary answers about the similarities and differences existing between

English and Greek in early reading and spelling. With data deriving from three different cohorts, two English-speaking Canadian (Alberta and Ontario cohorts) and a Greek-speaking in Cyprus, we focus on children reading both words and non-words as well as comprehending text. At the same time, the developmental associations between phonological and cognitive correlates with reading and spelling are examined. The Ontario team discusses evidence from children who experi-

– 239 –

ence considerable difficulties with reading and/or spelling, examining the so-called double-deficit hypothesis', substantiating any relevant arguments about the role of phonological and naming speed skills as strong predictors of reading development in English. Overall, it emerges that such an endeavor provides findings that explicitly speak out for substantial differences existing between the two languages: Greek children learn to read rapidly and accurately even only seven months after the start of formal reading instruction. On the other hand, phonological and orthographic processing measures appear to account for more variance in the English- than in the Greek-speaking sample. Finally, although the double-deficit hypothesis is well searched with English-speaking populations, even with the home background being taken into account, it remains an open issue with the Greek-speaking population.

Predictors of word fluency in english and greek: A cross-linguistic comparison

George K. Georgiou, University of Alberta, Canada

Rauno K. Parrila, University of Alberta, Canada

Timothy Papadopoulos, University of Cyprus, Cyprus

This study examines the relative importance of different components (phonological sensitivity, phonological memory, and rapid naming speed) of phonological processing for learning to read an orthographically regular language (Greek) and an orthographically irregular language (English). Sixty English-speaking Canadian children and 70 Greek-speaking Cypriot children participated in this study. Children were followed from grade 1 until grade 2. Independent and dependent measures were similar in both languages. Initial analyses of the data indicate that the accuracy of word decoding was close to ceiling for the Cypriot children already at the end of grade 1, whereas the same was not true for the grade 2 performance of the Canadian children. Regression analyses with word reading fluency as the dependent measure showed that generally phonological and orthographic processing measures accounted for more word reading fluency variance in the Canadian sample than what was true for the Cypriot sample. Fischer's r to z transformations, however, showed that the only significant differences between the two orthographies were found when orthographic processing tasks were used to predict word reading fluency. Our findings suggest that the cognitive processes that are crucial for the development of reading ability in English may play a more limited role in languages such as Greek in which children easily learn to decode any word irrespective of its length and complexity.

– 240 –

Phonological awareness, naming speed, and working memory as predictors of reading comprehension in greek

Timothy Papadopoulos, University of Cyprus, Cyprus

George K. Georgiou, University of Alberta, Canada

Rauno K. Parrila, University of Alberta, Canada

Eleni Anastasiou, Ludwig-Maximilians-Universitat Monchen, Germany

The aims of the present study were two-fold: (a) to identify a combination of predictive measures from grades 1 and 2 that correlate with reading comprehension in grade 2, and (b) to examine the predictive accuracy of these measures, in a Greek-speaking population. Seventy children participated in this study. Measures representing color, object, letter and digit naming, phonological segmentation and blending, phonological memory, word identification, word attack, orthographic choice, and word chains were used as the independent measures, administered in both grades 1 and 2. Reading comprehension was measured at the end of Grade 2, using an adapted version of the task from Woodcock-Johnson Psychoeducational

Battery-Revised. Four predictive models representing different combinations of the predictive constructs were run. Analysis showed that although the phonological variables were highly correlated with passage comprehension, the ability to decode or read isolated words was even higher correlated with the dependent measure. In grade 2, passage comprehension was highly correlated with all grade 2 independent measures, yielding the highest correlations with phonological segmentation and blending, followed by word chains and orthographic choice. The subsequent multiple regression analyses showed that the grade 1 model combining blending of sounds, word chains, and decoding of words was identified as the best predictor of passage comprehension in grade 2. Similarly, the grade 2 model joining blending of sounds and segmentation of words, words chains, and reading of isolated real words was yielded as the best predictive model. These findings are important for two reasons: (a) they indicate that phonological more than the rapid automatized naming measures have higher predictive power on reading comprehension, a striking finding countered to what predicts word reading in Greek, and (b) they strongly support the notion of word recognition modularity in a salient orthography as far as the prediction of reading comprehension is concerned.

– 241 –

Language development and reading: The roles of naming speed and phonological awareness

John R. Kirby, Queen’s University, Canada

Jennifer E. J. Dawson, Queen’s University, Canada

Jennifer Currie, University of Alberta, Canada

The Double-Deficit hypothesis (Wolf & Bowers, 1999) indicates that deficits in phonological awareness and naming speed are associated with reading disabilities.

These two oral language constructs can also be seen as predictors of reading development (Kirby, Parrila & Pfeiffer, 2003). This paper reports results from a four year longitudinal, prospective study of children selected to fit the four double-deficit categories. Our goals in the present paper are to (a) examine the home backgrounds of the children in the four groups, (b) track the development of the children in the four kindergarten-identified groups, and (c) investigate the relationships between the phonological awareness and naming speed dimensions and reading, taking into account other factors, such as home background and home literacy activities. We screened 547 kindergarten children (aged 5 years) on measures of phonological awareness (Word Blending) and naming speed (Object Naming), and selected 214 children who fit in one of the 4 groups (no deficit, phonological deficit, naming speed deficit, or double deficit). During kindergarten, we assessed these Children's letter knowledge and early reading skills (including letter-sound knowledge and word reading), and we also asked their parents to complete a questionnaire about home literacy activities. During Grades 1, 2, and 3, we are assessing further early reading and spelling skills (regular, irregular and pseudoword reading, orthographic processing, and regular and irregular word spelling). Results indicated that home background was associated with oral language deficits in kindergarten. There was also considerable movement between groups, indicating that kindergarten diagnoses were not entirely stable. Reading results supported the expected pattern: children with no observed deficits performed better than those with single deficits, who in turn out-performed those with the double deficit. Phonological awareness and naming speed continued to predict reading development after taking home background into account.

– 242 –

C 16 24 August 2005 08:30 - 10:30 Room E009

Symposium

Education for Peace

PEACE EDUCATION IN REGIONS OF CONFLICT AND TENSION: DOES

IT MAKE A DIFFERENCE?

Chair: Gavriel Salomon, Haifa University, Haifa 31905, Israel

Organiser: Gavriel Salomon, Haifa University, Haifa 31905, Israel

Peace education, co-existence education, education for mutual understanding and their likes are carried out all over the world, particularly in regions of severe conflict and tension. However, despite this richness of activity, there is very little research into these activities (with the exception of research on inter-personal conflict resolution in the schools). In light of this shortage of scholarly activity, research in such countries as Cyprus, Kenya or Israel/Palestine is of particular interest. Much of this interest stems from the examination and analysis of existing programs (Cyprus: Hadjipavlou; Kenya: Owinyo; Israel: Gordon) as well as from analyses of the challenges that face peace education in regions of severe conflicts (Salomon). The latter leads to serious yet researchable questions about the ways that institutional education can contribute to peace education and about the possible limitations of what peace education can achieve. Owinyo and Wildemeersch describe the problems faced by anti-violence programs in Kenya, with particular emphasis on the incongruity between the local culture and the one embedded in programs designed in the industrial West. Hadjipavlou describes the effects of the opening of the line between the two parts of Cyprus on members of the community and asks how could this be incorporated into school-based activities. Gordon describes a qualitative study of a school-based program designed to promote Jewish-Arab co-existence and the difference between the way teachers and students approached it. He focuses in particular on the blindness that extreme enthusiasm can afflict. Salomon lists hurdles facing peace education, research results that are incongruous with these, suggests a theory to account for the discrepancy and offers specific testable hypotheses.

In all, the proposed symposium describes peace education in three conflict countries and raises serious questions about its nature and implementation as a planned educational intervention.

– 243 –

Does peace education in the context of intractable conflicts affect the really important beliefs about self and adversary? Theory and researchable questions

Gavriel Salomon, Haifa University, Haifa 31905, Israel

A discrepancy is pointed out between formidable and thus discouraging hurdles facing peace education in the context of intractable conflicts and actual, encouraging research findings of such programs. It is suggested that the hurdles pertain to the most deep-seated, and thus unchangeable convictions constituting the backbone of a group's collective narrative. On the other hand, the change-objects affected by peace education programs pertain to more peripheral attitudes and beliefs which are more easily changeable, more weakly associated with behaviors and thus less consequential. This hypothetical possibility is briefly examined from both a theoretical and practical perspective, leading to three clusters of research questions: (a) Is the proposed distinction between central and peripheral attitudes and beliefs applicable to peace education programs? (b) How stable are changes of peripheral attitudes in the absence of changes of the more central ones? And (c) to what extent can only long-term, socialization-like programs affect core beliefs and attitudes?

The Blindness of Ideological Commitment: The Educational Tragedy of Arab/Jewish Co-existence

David Gordon, Ben-Gurion University of the Negev, Israel

I will illustrate educational problems that can arise as a result of laudable ideological commitment. The illustrations derive from the evaluation of the Children Learning Together (CLT) Project developed to promote Arab-Jewish coexistence. CLT is a program which emphasizes commonalities and similarities between the two nationalities. A basic assumption is Allport's Contact Hypothesis i.e. under certain conditions, inter-group contact reduces hostility and prejudice. An implicit assumption is that the development of bonds with someone of another culture is part of one's own personal growth as a human being. I engaged in a qualitative evaluation of 3 pairs of Arab and Jewish schools. A consistent motif revealed is the different reaction and perspectives of teachers as opposed to students. Teachers were active, enthusiastic and happy to meet each other whereas students were passive, less enthusiastic and tended to remain within the boundaries of their ethnic group. Much of the teachers' enthusiasm was the outgrowth of teachers' workshops which aimed to produce true believers' of the ideology and ideas that CLT represents. As a result, however, teachers often tended to misread the student's reactions to activities and failed to interpret correctly students' dissatisfaction, which for an outsider seemed

– 244 –

obvious. Extreme commitment may create a blindness' which leads to moral insensitivity, uncritical transfer from teacher to student activity and to teachers' loss of touch with reality. Such conduct may become harmful especially, when ideological commitment overcomes pedagogical considerations. In addition CLT's implicit assumption re personal growth may be problematic because its existentialist-like orientation may be at odds with the pragmatic nature of school life.

A search for peace in secondary education systems in Kenya

Salome Owinyo, Kenya, Kenya

Danny Wildemeersch, Catholic University or Leuven, Belgium

The goal of this study is to explore the nature and effect of peace education in a

Kenyan education context. Following an empirical research carried out in Kenya to understand the despicable culture of violence that has permeated the schools, we observed that the traditional conceptions and practices of peace education do not seem to be very helpful. Violence in the secondary schools has developed into an alarming and worrying trend. Whereas before school unrest was characterised by simple walkouts its now evident that students premeditate and plan actions aimed at causing maximum harm (MOE, 2001). As a result many students have lost their lives in these skirmishes and the schools have lost millions of property which have been destroyed and burnt-up. In retrospect, very little research has been done to understand this phenomenon and as such efforts to intervene and control these incidences have been insufficient as we continue to witness more schools engaging in these culture. Our concern is to explore this helplessness' depicted by traditional conceptions of peace education that do not seem to function in this situation. As stated in part of UNESCO's constitution, ìSince war begins in the minds of men, it is in the minds of men that the defences for peace must be constructed. We believe that it's imperative to first give up the traditional ambitions of peace education and question the conceptions of peace educators by opening up for another understanding.

Multiple stories: the Crossing as part of people's Peace Education in Cyprus?

Maria Hadjipavlou, University of Cyprus, Cyprus

The paper discusses the dynamics of crossing to and from the Line in Cyprus giving examples of the new experiences and stories as these are constructed when meeting the Other. I will argue that these stories form part of the public reconciliation process and constitute part of informal peace education. I define peace education as the

– 245 –

capacity to reach to the Other, feel empathy for their suffering and engage in shared social activities thus challenging the bipolarity of ìus and them. The challenge remains of how to incorporate these ìnew realities and stories in formal peace education curricula. To do this would mean official engagement in a new dialogue about history making. The Line in Cyprus is about 112 miles long stretching across the island separating North and South. This Line, which according to one's positioning in politics, ideology and history is referred to as the green line, the ceasefire line, the dead zone, the demarcation line, the partitioning line, the Attila line, the no-man's land or the border, has been opened on April 23, 2003, in four areas -two in Nicosia, one in the Famagusta area and one in Larnaca. Thousands of people from all Cypriot communities have been crossing to and from since then and new relationships have been formed based on direct face-to-face experiences with the Other. Many of these stories reveal, at first, the presence of the past as was experienced by each other in formal schooling or through mediated information in the mass media. They do not, however, stay there because through the direct meeting with the Other, i.e. the Turkish Cypriot or the Greek Cypriot who has been living in the other's house and memories for over thirty years discover the human being who has also suffered and has been imagining a different life.

C 17 24August 2005 08:30 - 10:30 Room A109

Symposium

Research Methodology

WRITING RESEARCH IN THE LAB AND IN THE FIELD: RECENT

METHODS

Chair: Asa Wengelin, Lund University, Sweden

Organiser: Daniel Perrin, University of Applied Sciences, Switzerland

Discussant: Deborah McCutchen, University of Washington, United States

David Galbraith, Staffordshire University, United Kingdom

During the last 20 years writing research has evolved from the study of stand alone texts to include cognitive and social perspectives. Some reasons for this could be the increased availability of computers, and the possibilities of interactive writing provided by the internet and other interactive tools. . This has lead to a need also for teaching practices to incorporate the concept of process and social dimensions into the teaching of writing as well as into the writing for learning practices.

– 246 –

Therefore, in order for researchers and educationalist to be of use for each other, it would be useful if methodological development within the writing research area, was concerned not only with research methods which can yield research outcomes, but also research with methods which can be used by educationalists as part of their intervention, which in turn could be researched etc. For this purpose we need multimethod approaches that involve both qualitative and quantitative analysis, allowing the possibility of both interpretive and positivist stances in order to cover as well social as cognitive perspectives on writing. Thus the papers in this symposium all put forward recent multi-approach methods that could be used both in research and in the classroom. Williams and Edwards focus on collaborative writing and explore dimensions of group interaction and suggest how email may be a vehicle for realising and recognising those dimensions. Perrin put discusses the relation between interviews and participatory observations before the writing session, keystroke logging of the actual session and verbal protocols after the writing session in progress analysis. Bakker and de Glopper test the combination of the four methods thinking aloud protocols, writing questionnaires, keystroke logging and text analysis.

Finally Wengelin, Johansson, Lindgren and Stevenson reviews different usages of keystroke loggings and combinations of keystroke logging with other methods.

Collaborative writing: The role for effective research and effective learning.

Noel Williams, Sheffield Hallam University, United Kingdom

Kirstie Edwards, Sheffield Hallam University, United Kingdom

Evaluation of student collaborative writing, whether we are looking at product or process, requires better knowledge of the social constructs operating during a project, and the way they may relate to product outcomes than we currently have.

For research, obtaining that knowledge is a commitment to explore socio-cognitivist models of writing, in order to map the interactions and hidden intentions of collaborating students onto the observed features of their products. Examining student emails is one method for providing both quantitative and qualitative data, which allows both positivist models of writing (e.g. data seen as evidence of group goaloriented problem-solving) and, simultaneously, more interpretativist models (e.g. using students' own categorisation of group interactions in addressing the writing task). This is because email data is both ìfixed, as permanent record with countable features, and discursive, implying or entailing a number of different discourse constructs to be fully understood (such as, for example, inter-group politeness strategies designed to maintain group cohesion). Research which aims to link the social process of writing to accounts of its collaborative product should have pedagogic

– 247 –

value in enabling teachers to identify points of connection between formative assessment, which is often process-based, and summative assessment, being generally product-based. We explore dimensions of group interaction which may have value in terms of the contribution to the writing project, suggest how email may be a vehicle for realising and recognising those dimensions, and how their connections and correlations might be examined. If such observations, even measures, can be identified in email exchanges, then they may also be detectable in other media used by collaborating student writers, offering tools for educators which are simultaneously helpful as classroom instruments and research tools. This in turn raises issues about the nature of the relationship between pedagogy and research in the context of writing.

Progression Analysis: A mult-method approach for investigating writing at the workplace

Daniel Perrin, University of Applied Sciences, Switzerland

What exactly do journalists do when they write - and what do they do with language? In 1997, the Swiss Federal Office for Communication (BAKOM) commissioned a qualitative and explorative investigation of journalistic writing in print, radio, TV, and on-line news offices. Before the study could begin, an ecologically valid method had to be developed to investigate writing processes at workplaces: the writing processes had to be accessible from a number of relevant perspectives but not be affected by the data collection. Progression Analysis (PA) was developed and applied for the first time within the framework of the BAKOM research project.

Data on writing processes were obtained from 40 workplaces in media newsrooms and evaluated as case studies. Additional corpora on journalistic writing have since been collected. Findings include detailed information on journalistic language performance, language awareness, language use, and writing strategies. Current research projects are using PA to investigate writing in other domains as well, such as in academy and industry.

PA is an ethnographic, computer-based multi-method approach with which data can be obtained on three levels.

1. Before writing begins, PA determines, through interviews and participatory observations, what the writing situation is and what experience the writer draws on for it.

2. During writing, movements are measured with keystroke-recordings. It's crucial that the computer logging doesn’t influence the performance of the editing system or the writer.

– 248 –

3. After writing, PA deduces the repertoire of an author's writing strategies. Based on cue-based retrospective verbal protocols, this third level of PA opens a window into the mind of the writer. The question is what can be recognized through this window. The presentation will focus on the relations between the thee levels of

Progression Analysis.

Comparing writing process research instruments in an educational environment

Carien Bakker, University of Groningen, Netherlands

Kees de Glopper, University of Groningen, Netherlands

To gain better insight in the way writers write-to-learn and how writing can affect learning, we need careful observations of the processes that take place during writing-to-learn. In order to observe these processes, researchers have the disposal of several observation techniques and methods like thinking aloud, writing questionnaires, keystroke analysis and text analysis. Until now, researchers have used the different types of observation instruments, but have never compared systematically the merits and drawbacks of each instrument or the surplus value of the use of a combination of instruments. Such a comparison will contribute to well- founded methodological choices. In a study of conceptual learning in science and writingto-learn processes of 15-16 year old students we developed and tested four instruments that capture writing processes: one synchronous - direct (thinking aloud), one asynchronous - direct (a task-dependent writing questionnaire), one synchronous

- indirect (keystroke registration) and one asynchronous - indirect (text analysis).

We have tested the combination of the four instruments with 5 students who carried out a knowledge transforming writing assignment while thinking aloud. During this writing act we recorded their key strokes. Afterwards, the students filled out the writing process questionnaire. All in all, we are able to analyze the writing processes that took place during the writing session with the help of thinking out loud protocols, key stroke analyses, responses to the writing process questionnaire and rhetorical and conceptual analysis of the texts produced. We currently analyse our data. We will outline a portrait of every student. We will make clear which contribution each instrument makes to the description of the writing process and we will determine the concrete surplus value of combinations of instruments.

– 249 –

A review of the different usages of keystroke logging - where are we heading next?

Asa Wengelin, Lund University, Sweden

Victoria Johansson, Lund University, Sweden

Marie Stevenson, University of Amterdam, Netherlands

Eva Lindgren, Umea University, Sweden

Keystroke logging has become an increasingly popular method for investigating the production processes during writing, both as a research tool (in experimental as well as naturalistic settings) and as a reflection tool for learners and teachers.

The advantages of keystroke logging are that it can provide the researcher with data not only on the finally edited text but also on editings and temporal aspects of the production process, and that it is does not intrude on the normal production process of the writer. However, a disadvantage of the method is its lack of transparency.

Keystroke logging provides us with information on frequencies, locations and durations of pauses as well as information on how much and what (on the surface level) is edited when and where in the production process, but it does not provide any information on which cognitive processes are going on during the pause, or the purpose of an editing. A possible solution to this is to combine keystroke logging with other data collection techniques, such as think-aloud protocols, stimulated recall, retrospective interviews, observations, writing questionnaires and text analysis. In this paper we will attempt to review the different usages of keystroke logging and keystroke logging in combination with other data collection techniques from both methodological and theoretical perspectives in order to outline where we are and where we should be heading next. Questions to be discussed are: What are the pros and cons of the different techniques? What are the theoretical assumptions underlying the different usages? What are the theoretical, methodological and pedagogical implications of the differens usages?

– 250 –

C 18 24 August 2005 08:30 - 10:30 Room E104

Symposium

Development of Expertise in Specific Domains

LANGUAGE COMPETENCE AND THE DEVELOPMENT OF SCIENTIF-

IC AND MATHEMATICAL KNOWLEDGE

Chair: Eva Teubal, Hebrew University of Jerusalem and David Yellin Te,

Israel

Julie Dockrell, Institute of Education, United Kingdom

Organiser: Eva Teubal, Hebrew University of Jerusalem and David Yellin Te,

Israel

Julie Dockrell, Institute of Education, London University, United

Kingdom

Discussant: Ma del Puy Perez Echeverria, Universidad Autonoma de Madrid,

Spain

Nora Scheuer, University of Comahue - CONICET, Argentina, Argentina

This symposium considers the relationship between language and the development of scientific knowledge and understanding. The four studies identify the language processes that support scientific and mathematic discourse across preschool and elementary school to adulthood. In their paper, Best and her colleagues use a naturally occurring event - an eclipse- to evaluate young Children's lexical and conceptual understanding over a five-month period. They demonstrate how the exposure to the eclipse was effective in helping children acquire lexical knowledge and wider understanding about entities relating to eclipse. This demonstrates the importance of multiple tasks to identify aspects of Children's learning that are central to understanding the development of scientific reasoning. The study by French and Peterson develops this theme through the evaluation of a special preschool science intervention. Both Children's lexical acquisition and their ability to use scientific discourse were examined. This demonstrates how high quality talk, in the form of extended conversations, and the use of multiple discourse forms supports Children's engagement in and understanding of scientific principles. The paper by Garcia, Rojo &

Andersen considers how written language can support the process of scientific understanding. Their results demonstrate how note- taking and report writing in elementary school science can foster the development of scientific inferences. The importance of embedding language within the scientific activity is further supported

– 251 –

by their analyses of science laboratory activities.

The by Teubal, Albert and Guberman discusses the effects of prototypicality and language on adults' conceptions of geometrical shapes. It shows that items belonging to the same categories may differ in prototypicality, and that although labeling may help in sorting geometrical shapes according to various definitions, it may also hinder the recognition of non prototypical items. These presentations emphasize the role of language as a representational system supporting scientific and mathematical discourse .

Children's Semantic Representations of a Science Term

Best Rachel, University of Memphis, United States

Assessments of lexical acquisition are often limited to pre-school children on forced choice comprehension measures. Our study assessed the understandings 30 school-age children (mean age = 6;7) acquired about the science term, eclipse following a naturalistic exposure to a solar eclipse. The knowledge children acquired about eclipses and a control term, comet was assessed at three points in time (baseline-test, two-week post-test and five-month post-test) using a range of assessment tasks (multiple-choice comprehension, picture-naming, drawing and solar system manipulation task). Children's knowledge was compared to 15 adult controls during the baseline-test and two-week post-test. The analysis focused on the range of knowledge children acquired about eclipses and entities related to an eclipse (sun, moon, earth and planets in general). We found that children acquired extensive knowledge about eclipses, but not comets. At the two-week post-test, the majority of children were able to produce the term eclipse and provided evidence of accurate comprehension and wider conceptual knowledge about solar eclipses, which was retained at the five-month post-test. Further, at the two-week post-test children had acquired knowledge about entities relating to eclipse, which was also retained at the five-month post-test.

Learning Language through Preschool Science

Lucia French, University of Rochester, United States

Shira Peterson, University of Rochester, United States

Many children living in poverty in the United States have limited exposure to school-based oral language practices by the time they enter preschool (e.g., Hart &

Risley, 1995; Heath, 1983). Unfortunately, research suggests that many preschool environments do not provide the kind of language environment that supports Chil-

– 252 –

dren's development of school-based language and literacy skills (Dickinson & Tabors, 2001). This study focuses on the language development of children exposed to ScienceStart!, a preschool curriculum designed to enhance Children's receptive and expressive language skills. In the ScienceStart! curriculum, children participate in daily hands-on science activities within a rich language environment stemming from scaffolded participation in a four-part cycle of inquiry. Over 450 children have participated in this study since its inception, 75% of whom come from low income families. Quantitative and qualitative methods were used to assess

Children's language growth. On the standardized PPVT and on an interview-style assessment of explanatory language, children showed significant improvement in their language skills over the course of a single 5-week unit, and over the course of the school year. Videotapes of classroom activities revealed that children participated in high quality talk with teachers, including having extended conversations, engaging in multiple discourse forms, using features of scientific language, and producing causal explanations for science phenomena. These results support our claim that a language rich environment in preschool can support Children's normal acquisition of school-based language and literacy skills. The science-based curriculum is particularly suited for this goal because the science content engages children, and the science cycle provides a predictable linguistic format which scaffolds

Children's participation in classroom discourse.

The role of labels and prototypes in geometrical conceptualization

Eva Teubal, Hebrew University of Jerusalem and David Yellin Te, Israel

Jeanne Albert, David Yellin Teachers' College, Israel

Ainat Guberman, Hebrew University of Jerusalem and David Yellin Te, Israel

The aim of the present study was to assess the effect of conventional labeling on adults' ability to identify geometrical shapes in terms of formal definitions.

Since prototypes are part of a word's meaning, it was hypothesized that labeling would enhance correct identification of prototypical items. Sixty two subjects were presented with 34 numbered shapes, and were given 9 different definitions, in the form of ìparty invitations. Each invitation contained either a conventional or a nonsense name for the ìguests, and a list of defining characteristics: ìOnly the DUCS are invited. A DUC is a quadrilateral whose angles are all equal. All the subjects had to list the shapes that fitted each invitation. Subjects who received nonsense names were asked to provide the conventional names, if they could. The research questions were: 1) Do Ss perceive membership in a geometrical category as a homogenous or as a graded trait? 2) How do conventional labels affect the subjects'

– 253 –

ability to act according to the definitions? 3) What is the relation between success in the sorting task and the ability to relate the conventional name of the category?

Findings: 5. Identification rate was higher for "polygons" that don't have more specific definitions, and for "quadrilaterals" that do (such as rectangles and trapezoids).

6.Labeling helps subjects' sorting in the case of parallelograms and trapezoids. It had no effect in the case of "polygons", "quadrilaterals", "rectangles", "squares" and

"rhomboids". Labeling contributed to the identification of prototypical rhombuses, and hindered the identification of non-prototypical rhombuses (squares). 7.Subjects could identify at least some of the relevant shapes without the ability to specify the shapes' names, but not vice versa. It was concluded that membership in geometrical categories is graded. Conventional labeling promotes the recognition of some prototypical items, but may hinder the recognition

Written language as a metacognitive tool in science knowledge construction

M. Garcia-Mila, Universitat de Barcelona, Spain

N. Rojo, Universitat de Barcelona, Spain

C. Andersen, Ohio State University, United States

The present study aims at analyzing the mediating role of writing in fostering metacognition during scientific inquiry. Sixth grade elementary students were asked to investigate the effect of three factors (type of fertilizer, type of light and type of seed) on the plants growth. 34 students worked over seven sessions (twice a week), which permitted a microgenetic analysis of the data. During the inquiry process, participants had the chance to design the experiments they wanted, interpret the evidence they gathered, and make the corresponding inferences based on such data.

Each participant was provided with a lab notebook to use whenever needed. They were also asked to write a report in the mid and final sessions. Our results show the microgenetic development of scientifc inquiry strategies (data collection and inference making) embedded with note-taking, and related with the reflection fostered by report writing. We found a statistically significant relationship between the quality of note-taking (taking complete notes) and the use of factorial combination strategies in the experimental design. In terms of report writing, our results show that the demand of writing a report in the mid session induced the students to stop and review their work by looking at results obtained so far which may have fostered the reflection process. Our results also show that those students who wrote reports with a high level of metacognition, that is, reports that went beyond the data recording by including inferences showed a statistically significant improvement in their inference making. These results, taken as a whole, show that prescriptive outlines

– 254 –

should not be the closure activity in the science laboratory activities. Report writing should be embedded during the inquiry along with the need to take notes.

C 19 24 August 2005 08:30 - 10:30 Room E103

Symposium

Action Research

COMMUNITIES OF PRACTICE IN HE

Chair: Gina Wisker, Anglia Polytechnic University, United Kingdom

Organiser: Gina Wisker, Anglia Polytechnic University, United Kingdom

Discussant: Erik Meyer, University of Durham, United Kingdom

Building on theories of Communities of Practice, this symposium aims to collate examples of Communities of Practice operating successfully in Higher Education contexts in three countries, UK, Israel and Australia. Two Communities of Practice relate to the continuing professional development of practising HE lecturers and Learning and Teaching advisers. One CoP focuses on continuing professional development for university staff aiming at publication, and another looks at the successful strategies and Communities of Practice activities of school-based Learning and Teaching advisers. Two others consider CoPs developed in relation to a cohort based international PhD. Here, Wisker, Robinson and Shacham, explore how the development of cohorts, guardian supervisors and online supervisor support enhance postgraduate learning and supervisory practice in terms of the personal, institutional and learning development areas. Trafford and Leshem consider the specific role and significance of conceptual frameworks as critical features in research and ask how candidates gain understanding of conceptual frameworks that incorporate ideas into their theses in the context of a cohort based PhD community of practice.

The two final papers focus on web based and e-learning examples, Zimitat explores the effective use of WebCaseStudy which aims to enable health based professional learners to avoid the reification of knowledge and become involved in a community of practice, exploring the use of clinical approaches to cases and developing higher order skills and soft knowledge in the process. Uwe Richter and Sharon Waller use a constructivist approach to investigate the use of online discussions to help cultivate a learning community, in relation to student expectations on two masters modules designed to equip learners with the skills and pedagogical understanding required to facilitate effective learning with ICT. Each paper is research based and

– 255 –

focuses on the development and effective utilisation of Communities of Practice in

HE contexts.

Cultivating a professional learning & teaching community of practice at APU

Sharon Waller, Anglia Polytechnic University, United Kingdom

This paper considers how the predominantly individualistic culture of academics working in isolation from each other at a split campus university - Anglia Polytechnic University (APU) - could be influenced by the cultivation of connections between staff to facilitate the sharing of good learning and teaching practices across the university. It suggests that like most organisations APU can be regarded as a community of Communities of Practice (Brown and Duguid, 2001, p. 203) connected by networked computers. However whilst the networked infrastructure provides the material basis' (Castells, 1996, p. 369) which affords communication flows between staff across the university, knowledge sharing across communities is dependent upon the development of shared practices (Greeno, 1996; Brown and Duguid,

1989; Brown and Duguid, 2001). Using the university's School-based Learning and

Teaching Advisors as a case study the paper explores the notion that by working with groups of people who are bound together by a shared interest whilst also having multi-membership of different communities the university is able to take advantage of their ability to act as knowledge brokers' and mediate across community boundaries. The paper suggests that, as part of their practice, advisors have become stewards of the learning, teaching, assessment and curriculum (LTAC) domain, a role legitimised by their influence on the development of institutional policies. Interviews with the advisors suggest that successful cultivation of a community of practice is a process of negotiation rather than direction dependent upon understanding how it functions, what is important to its members and how to elicit their commitment. Additional considerations for a distributed community include the need to equip members to use technology more effectively, in order to maximise their time and strengthen the connections between them. Lessons learned will be used to inform the cultivation of a new community of Learning Technologists.

– 256 –

Postgraduate research success: Communities of practice involving cohorts, guardian supervisors and online communities

Gina Wisker, Anglia Polytechnic University, United Kingdom

Gillian Robinson, Anglia Polytechnic University, United Kingdom

Miri Shacham, Orte Braude College of Engineering, Israel

Traditionally, supervisors work with students on an individual basis and postgraduate development programmes run onsite. Isolation can be a key feature for postgraduates, particularly international students or those studying at a distance, and for their supervisors. Entry into the international community of research, supervision, study and publication can be enabled, arguably, by the support of Communities of

Practice from the outset of postgraduate student/supervisor interactions. In this context, considerable numbers of international postgraduates at APU (UK) and their supervisors are effectively supported in three key areas of engagement, need and development: The Personal (e.g. support from friends, colleagues and family, stress and coping), Learning (research seen as a form of learning, development of learning levels, research processes and skills) and Institutional (support provided by the institution- postgraduate development programmes, supervision, infrastructure). The support is enabled by building three innovative Communities of Practice:

(1) Guardian supervisors - work with all students on research development programmes, with accompanying meetings focused on strategies of meta-learning. They support students' work with emails, and other distance media. This builds a CoP among the Guardian supervisors, and all who take part in the PhD programme.

(2) Cohorts - PhD students are empowered to develop mutual, critically focused support for each others' work through the enhanced use of the cohort as a CoP in the compulsory research development workshops and through ongoing discussion lists, self help groups and symposia.

(3) Online supervisory interactions - Supervisors of international postgraduates are supported as a CoP thorough the provision of online supervisory discussion and development.

This paper is based on action research with current postgraduate students and those who have successfully completed PhDs, guardian supervisors and supervisors as collaborators to explore the rationale, problems, practices and the richness of experience of working with systems fostering Communities of Practice.

– 257 –

Using online discussions to help cultivate a learning community and manage student expectations: A social constructivist approach

Sharon Waller, Anglia Polytechnic University, United Kingdom

Uwe Richter, Anglia Polytechnic University, United Kingdom

This paper explores the processes of negotiation and mediation which take place between learners and teaching staff on two masters modules designed to equip FE/

HE tutors and educational technologists with the skills and pedagogical understanding required to facilitate effective learning with information and communication technology (ICT). Both modules are delivered in a blended mode of classroombased and online learning activities including facilitated online discussions. Both modules are enhanced by a website integrated within a virtual learning environment

(VLE). The paper examines the rationale for the development and integration of a pedagogical framework which is underpinned by a guided-construction model of learning (Shuell, T., 1992). This is followed by a consideration of how the framework operates in practice as a dynamic mechanism reflecting student learning and social needs as well as supporting the management their expectations. Reflections on the significance of the role played by the modules' online elements in cultivating a learning community (Wenger, et al., 2002) in order to better support students who are predominantly busy professionals are endorsed by student feedback and analysis of their online contributions. Emerging patterns of communication and behaviour within a networked learning environment together with associated implications for the roles of both tutors and learners are discussed. Learning within a networked environment is new to students who have yet to become encultured (Blackler, 1995, p.

1024) in the practices of online learning communities. It is therefore suggested that learners need guidance in how to behave, customs to be observed, what to expect and expectations of them. The paper concludes that successful learning depends on the robustness of the underlying pedagogical framework and the development of a supportive learning community whilst acknowledging that the development of successful networked learning takes time and effort due to the changes required in cultural and professional practices.

Webcasestudy: Using technology to capture soft knowledge within a community of practice

Craig Zimitat, Griffith Institute for Higher Education, Australia

This paper explores the processes of negotiation and mediation which take place between learners and teaching staff on two masters modules designed to equip FE/

– 258 –

HE tutors and educational technologists with the skills and pedagogical understanding required to facilitate effective learning with information and communication technology (ICT). Both modules are delivered in a blended mode consisting of classroom-based and online learning activities including facilitated online discussions. Both modules are enhanced by a website integrated within a virtual learning environment (VLE). The paper examines the rationale for the development and integration of a pedagogical framework which is underpinned by a guided-construction model of learning (Shuell, T., 1992). This is followed by a consideration of how the framework operates in practice as a dynamic mechanism reflecting student learning and social needs as well as supporting the management their expectations.

Reflections on the significance of the role played by the modules' online elements in cultivating a learning community (Wenger, et al., 2002) in order to better support students who are predominantly busy professionals are endorsed by student feedback and analysis of their online contributions. Emerging patterns of communication and behaviour within a networked learning environment together with associated implications for the roles of both tutors and learners are discussed. Learning within a networked environment is new to students who have yet to become encultured (Blackler, 1995, p.1024) in the practices of online learning communities. It is therefore suggested that learners need guidance in how to behave, customs to be observed, what to expect and what is expected from them. The paper concludes that successful learning depends on the robustness of the underlying pedagogical framework and the development of a supportive learning community whilst acknowledging that the development of successful networked learning takes time and effort due to the changes required in cultural and professional practices.

– 259 –

C 20 24 August 2005 08:30 - 10:30 Room A018

SIG Invited Symposium

Writing

METHODOLOGY IN WRITING RESEARCH: A CONDITION FOR THE-

ORETICAL INSIGHT AND FOR EFFECTIVE LEARNING

Chair: Joachim Grabowski, PH Heidelberg, Germany

Thierry Olive, CNRS & University of Poitiers, France

Organiser: Joachim Grabowski, PH Heidelberg, Germany

Thierry Olive, CNRS & University of Poitiers, France

Discussant: Pietro Boscolo, University of Padova, Italy

Linda Allal, University of Geneva, Switzerland

This is the Invited Symposium of SIG Writing. Within EARLI, there is a rather stable international writing research community across Europe and beyond Europe that encompasses a wide diversity of research interests, disciplines, methods, and approaches related to the phenomenon of writing as a human skill, its processes and products, its mastery and development, and the affiliated conditions and educational supports. However, theoretical insights, empirical results and practical experiences will only accumulate and lead to mutual progress if the community members, in spite of their different disciplinary backgrounds, continue to mutually explain the ways they reconstruct and theoretically as well as practically approach the questions at issue and the needs and problems they encounter in their respective fields. The symposium is also aimed at concisely informing the learning-and-instruction community about a relevant part of the state of the art on educational writing research.

Thus, the presentations of the symposium will provide an overview of recent important methodological approaches in writing research, from theories of working memory (Olive, Piolat and Kellogg: Investigating the role of working memory in writing: Ways, uses, and conclusions) and their applications to writing instruction at school (Bourke & Adams: Applying working memory research to writing instruction within the Primary School classroom) to the study of graphic and ocular motor behavior in skilled writing and its development (Alamargot: What on-line analyses of the eye and the pen reveal about the writing process and the ways to foster it) and the ways in which classrooms, the main loci of writing instruction, are considered and theoretically reconstructed from different educational perspectives. Together, the symposium tries to show and delevop ways how to get -From writing processes

– 260 –

to writing instruction, and back againì, a question that, raised by Linda Allal in her opening presentation, guided the 2004 SIG Writing Conference.

Investigating the role of working memory in writing: Ways, uses, and conclusions

Thierry Olive, CNRS & Universitede Poitiers, France

Annie Piolat, Universite de Provence, France

Ronald T. Kellogg, Saint-Louis University, United States

Writing involves several resource-demanding cognitive components that have to be orchestrated when composing a text (McCutchen, 1994). Understanding how writers compose a text thus means explaining how activation of these writing processes is orchestrated in working memory whose main characteristic is its limited capacity in simultaneously maintaining and processing information. Working memory is the cognitive structure (or function) that temporarily stores and processes information during the realization of complex cognitive activities. In writing, the role of working memory in writing was investigated with two theoretical perspectives and with two kinds of methodological approaches. On one side, based on McCutchen's (1994,

1996) capacity theory of writing (the more efficient the writing processes, the less they require resources from working memory and the more resources are available for activating other processes and for coordinating goals), Berninger and Swanson (1994; Swanson & Berninger, 1996) studied individual differences in working memory on writing performance with a correlational approach. This approach led them to propose a theory of writing acquisition in which the emergence of the writing processes is determined by the constraints of working memory. On the other side, Kellogg (1996) analyzed, at the light of Baddeley's multi-component model of working memory, the relationships between the writing processes and the different components of working memory (the central executive, the phonological loop and the visuospatial sketchpad). Methodologically, Kellogg's proposals were tested with the dual-task technique. Within these two lines of research, this presentation will describe some typical experiments and studies conducted in each perspective in order to review the recent findings about the involvement of working memory during writing. In parallel, we will illustrate how the different techniques have been fruitfully used to explore the role of working memory during writing.

– 261 –

Applying working memory research to writing instruction within the Primary School classroom

Lorna Bourke, Liverpool Hope University College, United Kingdom

Anne-Marie Adams, Liverpool John Moores University, United Kingdom

To investigate the relationship between working memory abilities and the development of Children's narrative writing skills. The nature and extent of this relationship is examined in two contexts. The first relates to the process used to instruct the children on the composite skills required to generate written narratives (word, sentence and text level, NLS, DfES, 2001). The second context relates to the use of appropriate statutory assessment measures of writing skill (i.e. Baseline; Key Stage

1 [KS1]). The implications of the research in terms of educational practice will be considered. Children were assessed on the quality of their writing at word, sentence and text levels at two age points (4-5 years & 6-7 years). Tasks measuring the visuospatial (corsi blocks, visuo-spatial pattern span), phonological (digit span, nonword repetition, word span) and central executive (verbal fluency, sustained attention to response, dual-task co-ordination) and integrated phonological storage and processing components (complex listening span task) of working memory were adopted.

The written texts were analysed in terms of diversity of vocabulary, mean length of sentences in morphemes, and overall coherence (age 6-7 years only). They were also assigned an attainment level according to educational assessment guidelines for both ages (i.e. Baseline assessment at 4-5 years and Key Stage 1 assessment at

6-7 years). Hierarchical regression analyses revealed that individual differences in central executive functioning predicted performance on baseline assessments (age

4-5 yrs) and vocabulary diversity, total number of sentences, text coherence and

KS1 attainment level (6-7 years). Individual differences in working memory capacity is associated with poorer quality texts. These findings support the characterisation of the writing task both in educational literature and in investigations of the cognitive skills underpinning writing as a resource-demanding task which requires the writer to successfully manage a number of processes in order to achieve a wellwritten text.

What on-line analyses of the eye and the pen reveal about the writing process and the ways to foster it

Denis Alamargot, University of Poitiers - CNRS, France

Since Hayes and Flower's (1980) seminal work, writing research has aimed at solving two important issues: (i) to further describe the processes involved in writing,

– 262 –

and (ii) to understand the dynamics of these processes. Among the different ways of conducting real time analyses of writing processes, recording the variation of graphomotor activity is particularly interesting because it is objective, non-intrusive and offers a continuous measure of the temporal aspects of processing. Stemming from research on oral production, this method can nevertheless be insufficient to assess some specific processes of writing. Based on the combined measurements of ocular

(the input) and graphomotor (the output) activities of the writer, the Eye and pen device (Chesnet & Alamargot, in press) provides a very fine-grained description of the temporal characteristics of written production and offers a new framework to understand the writing processes. Eye and Pen relies on two main devices: a digitizing graphic tablet (to record spatial coordinates and pressure of the pen on the tablet surface) and an eyetracker (to record eye movements). All these observations are stamped with a common base millisecond timing. From the text written by the participant, and digitalized by the tablet, one may rebuild forward and backward on the computer screen, the trace leaved by the pen and the eye position at the same time

(synchronized events). It becomes possible to improve investigations on processes engaged during the course of a pause as well as during a period of transcription.

At an experimental level, this device will allow advances in the study of the visual component engaged during writing and of the functioning and dynamic of writing processes. At a methodological level, it already allows the study of handwriting in a multimedia computer environment (like the upcoming screen-pad).

Writing research in the classroom: Strengths and restrictions of different theoretical orientations and pedagogical perspectives

Triantafillia Kostouli, Aristotle University of Thessaloniki, Greece

While writing has been traditionally associated with schooling, the shift of writing research into school classrooms has brought into the foreground a new set of issues with regard to the way classrooms, as a specific and rather complex kind of context, may actually affect the nature of Children's writing; of great importance in this regard is another, parallel-running research strand focusing on the pedagogical processes and practices through which students in specific classrooms learn how to create texts according to specific, school-valued ways and reflect upon the texts they or their classmates produce. Different research traditions may be noted in the literature on classroom writing, differentiated through the methodological frameworks adopted for describing the nature of classrooms as learning contexts and through the different kinds of data analyzed; reference can be made to cognitive and sociocognitive research that traces school Children's developing understand-

– 263 –

ings of writing, to discourse-analytic work on texts produced by school children in classrooms (usually seen as finished products), to ethnographic accounts of textual processes and practices, to sociocognitive and sociolinguistic work on writing contexts and to critical discourse analytic work on the nature of school genres and classrooms as specific types of learning communities. Rather than singling out the similarities and differences of these perspectives in detail, this overview sets out to delineate certain overarching dimensions through which these research traditions may be differentiated from one another. This is attained by attending to answers provided to two basic questions: How is the nature of classroom as a specific type of context conceptualized by pertinent research strands? What are the types of data

(texts and activities) through which writing ability is defined and assessed in classroom contexts? The research traditions developed are outlined and their pedagogical influence is delineated.

C 21 24 August 2005 08:30 - 10:30 Room A019

SIG Invited Symposium

Motivation

HOW TO CONCEPTUALISE AND MEASURE THE SITUATED NATURE

OF MOTIVATION IN CONTEXT-ORIENTED RESEARCH?

Chair: Simone Volet, Murdoch University, Australia

Organiser: Simone Volet, Murdoch University, Australia

Marold Wosnitza, University of Koblenz-Landau, Germany

Discussant: Susan Nolen, University of Washington, United States

This symposium addresses the topic of conceptual and methodological challenges in contextually-oriented research on motivation. The emergence of the person-incontext perspective within social-cognitive psychology combined with the growing interest in sociocultural conceptualisations of learning have lead to significant shifts in motivation research in recent years. Many scholars around the world have taken the challenge of studying motivation in real-life contexts and are now grappling with the difficulty of conceptualising and measuring the situated dimensions of motivation. The alternative positions, represented by the social-cognitive and the sociocultural traditions, are associated with different research methodologies and units of analysis. While distinguishing the fundamental assumptions of each tradition is necessary for conceptual clarity and interpretation, it has been argued that

– 264 –

from a pragmatic perspective combining approaches may do better justice to the complexity and multi-dimensional nature of motivation in real-life contexts. The specific aim of this symposium is to unpack and openly discuss these challenges.

The first paper by Turner & Patrick critically examines and clarifies the different theoretical conceptualisations and related methodologies, adopted by researchers who have invoked a contextually-oriented perspective in their research on motivation. Representative examples of recent research are analysed with a focus on their theoretical and methodological features. The second paper by Jarvela, Volet& Jarvenoja extends the discussion by examining the social processes of motivation in goal-oriented collaborative activities. The challenge of investigating motivation as a dynamic, dual psychological-social phenomenon is discussed and illustrated with the development of a dynamic instrument. The third paper by Minnaert outlines the conceptual and methodological threats faced by researchers conducting motivation research in ecologically valid contexts. A case for the the design of more longitudinal designs and multi-dimensional approaches but also more dynamic theories is made, with support from recent empirical studies. An open forum will follow the discussion by Nolen.

Mixed messages: The difficulty and challenge of defining and measuring situated classroom motivation

Julianne Turner, University of notre Dame, United States

Helen Patrick, Purdue University, United States

Recent attempts to apply contextually-oriented theories to understanding motivation have resulted in considerable changes to both concepts and methods used within motivational research. These changes are especially apparent in classroom research. Recent research has attempted to explain and measure motivation not simply as an individual difference, but also as situated within the classroom context. The growing acceptance of this position has required researchers to address several methodological concerns with respect to measuring situatedness - issues they did not previously have to attend to within a social-cognitive framework.

Thus, researchers invoking a contextually-oriented perspective find themselves with fewer established paradigms and procedures than those from other perspectives. Researchers have responded to the challenge of understanding motivation situated in classrooms from a rangeósometimes a combinationóof perspectives and with a mix of ecologically sensitive methods. The time is right to step back and examine differences and commonalities in this research. We will analyze and synthesize representative examples of international empirical research on motivation in

– 265 –

classrooms. The focus will be on three features:

1. Explicit or implicit theoretical conceptions of motivation. Some researchers characterize their conception as ìperson in context, others as ìsociocultural or ìsituated.

Yet others do not state a theoretical conception, but cite references, e.g.Vygotsky or Rogoff, implying a similar conception. Such conceptions can also be implied in choice of language.

2. Methods. Both the research design and the measures used to capture motivation in classrooms will be examine. The unit of analysis and which features of students and the context are studied and how will be noted.

3. Congruence and Discrepancies between theoretical conceptualization and methodological operationalization.

Following the description and analysis of different approaches to conceptualizing and measuring situated motivation, we will discuss the implications (theoretical, methodological, practical) of these approaches for the study of motivation in classroom contexts.

Motivation in collaborative learning: New concepts and methods for studying social processes of motivation

Sanna Jarvela, University of Oulu, Finland

Simone Volet, Murdoch University, Australia

Hanna Jarvenoja, University of Oulu, Finland

The aim of this paper is to discuss theoretical and methodological issues related to the study of contextual and social processes of motivation in goal-oriented collaborative learning activities. Three challenges towards a more social conceptualisation and investigation of motivation are identified. First, how to conceptualise motivation as a dual psychological-social phenomenon? Second, how to conceptualise the social nature and origin of motivation? These conceptual questions lead to the third challenge of how to study the social processes of motivation in dynamic, socially challenging collaborative learning activities. Studying the dynamics of motivation in socially challenging collaborative activities is needed since new pedagogical environments have stressed the importance of learners' active involvement in collaborative, social constructivist forms of learning. Such activities create new motivational challenges in the classroom and need to be better understood. A recent study that examined how university students collaborated on a study activity perceived as particularly motivationally and emotionally challenging will illustrate an attempt to collect questionnaire data on dynamic social processes of motivation.

We will introduce the idea of a ìdynamic questionnaire, which assesses each group

– 266 –

member's engagement in self-, other- and negotiated regulation strategies that are aimed at addressing what they perceived as the most socially challenging aspects of the situation, and eventually achieving their personal goals.

Beyond research on motivation in social learning contexts: Threats and challenges

Alexander Minnaert, University of Groningen, Netherlands

Empirical studies in the area of motivation in social learning contexts are often criticized about various conceptual and methodological issues. Many threats of construct, internal, external, and statistical conclusion validity are likely, also - and sometimes especially - when the research in conducted within an ecologically valid learning context. The objective of this contribution is to advance our understanding and awareness of the threats and challenges in the area of research on motivation in ecologically valid, social learning contexts. Three longitudinal, panel design studies, conducted in secondary vocational education and higher education within the

Netherlands, are used to illustrate these conceptual and methodological issues. On a meta-level, the findings of these studies compel for more fine-grained, multi-dimensional, and multi-level approaches to disentangle the recursivity among various motivational, social, affective, and cognitive variables in social learning contexts.

In reaction to the plentiful static designs and models, we need more ìcontextualized, dynamic longitudinal designs (e.g., cross-sectional, trend, intervention, time series, or panel studies) enriched by moderator/mediator variables and multiple outcome variables. Besides, the study dealing with theoretically incongruent or dissonant patterns of learning among first-year university students urges not only the need for more longitudinal designs and multi-dimensional approaches, but also the need for more dynamic theories able to model motivational orientations, learning conceptions, and learning and regulation strategies in a comprehensive way. Once the theory has been revised in this way, the phenomenon of dissonance, or at least some otherwise disturbing aspects of it, become part of the theory rather than something outside it. The threats and challenges in research on motivation in social learning contexts will be discussed.

– 267 –

C 22 24 August 2005 08:30 - 10:30 Room B107

EARLI Invited Symposium

Mathematics Education

TEACHING FOR A HIGH QUALITY OF MOTIVATION AND DEEP UN-

DERSTANDING: OUTCOMES OF A BI-NATIONAL PROJECT

Chair: Kurt Reusser, Institute of Education, University of Zurich,

Switzerland

Organiser: Kurt Reusser, Institute of Education, University of Zurich,

Switzerland

Discussants: Paul Andrews, University of Cambridge, United Kingdom

Kurt Reusser, Institute of Education, University of Zurich,

Switzerland

Which learning environments for mathematics education are optimally supportive, motivationally satisfactory, and effective? In a bi-national video study "Teaching quality and mathematical understanding in different cultures of instruction (a joint project guided by research groups located at the University of Zurich and the

German Institute of International Educational Research in Frankfurt; c.f. Klieme,

Reusser & Pauli, 2003), Swiss and German mathematics teaching has been investigated using both a longitudinal and a microgenetic design. The study is based on a multi-level (classroom, individual), multi-perspective (students, teachers, outside experts), multi-method (video data, questionnaires and tests), and multi-criteria

(maths achievement, maths interest and motivational profile) approach in order to examine the impact of instructional quality on the development of mathematical achievement and motivation of secondary school students. The study, which was conducted over the course of one year, combines video data (5 lessons per teacher) from two different mathematical topics with information gathered on a range of relevant dimensions of teaching and learning quality, incorporating self-reports from students, high-inference expert ratings of instructional patterns and content, and standardised achievement tests. The topics were anchored both in the Swiss and the

German math curricula, and consisted of the first three lessons of the introduction to Pythagorean theorem on the one hand and a lesson dealing with mathematical word problems on the other. The sample of the study consisted of 20 Swiss and

20 German secondary school classes from the two higher school tracks. Based on the videotaped lessons collected from the 40 teachers and the questionnaires and

– 268 –

tests of approximately 954 pupils of Years 8 and 9, the papers will use multi-level modelling to address differential relations (across different groups of students and different mathematical topics) between observational video data, student perception of learning environments, and student outcomes related to multidimensional measures and criteria.

Teaching patterns and learning quality in Swiss and German mathematics instruction

Isabelle Hugener, Institute of Education, University of Zurich, Switzerland

Christine Pauli, Institute of Education, University of Zurich, Switzerland

Kurt Reusser, Institute of Education, University of Zurich, Switzerland

Urs Grob, Institute of Education, University of Zurich, Switzerland

Teaching has an influence on the quality of the learning process and consequently on pupil achievement. The goal of this analysis is (1) to identify different teaching patterns, (2) to relate these to the self-reported, subjectively experienced learning quality of the pupils and (3) to compare this with a quality rating by trained observers with regard to cognitive activation in order to settle the question of whether a particular teaching pattern is especially conducive to learning qualities on the part of the pupils such as cognitive learning activity and understanding. From the research it is known that quality features such as "cognitive activation and "deep understanding are positively correlated with mathematics achievement. The problem-solving, discovery-based teaching pattern of Japan has therefore long counted as a model of cognitively activating teaching. As yet it has not been possible to empirically verify this link. Instead, previous research results point to the fact that a questioning-developing, direct instruction is particularly effective for pupil achievement. This question of the link between cognitive learning activity and different problem-oriented and questioning-developing teaching patterns is addressed in this contribution. This analysis emerges in the framework of the bi-national project. It is based on the 39 videotaped instructional units (117 lessons) on the introduction to Pythagorean theorem. In addition, pupil surveys (N = 954) on the experienced instruction and the quality rating of the videotaped lessons will be included. In the presentation both the methodological procedure and the results of the analysis will be presented.

– 269 –

Instructional quality, instructional content, and conceptual understanding: A micro-genetic study of geometry learning

Eckhard Klieme, German Institute for Int. Educational Research, Germany

Barbara Vetter, Institute of Education, University of Zurich, Switzerland

Frank Lipowsky, German Institute for Int. Educational Research, Germany

How should instruction be shaped to allow students to gain a deep understanding of domain-specific concepts and develop adequate, non-schematic epistemological beliefs? This paper reports on a study which tries to investigate content-specific teaching and learning from a micro-genetic point of view. 40 Teachers were videotaped in three lessons dealing with the introduction to the Pythagorean theorem, and they were asked to implement a mathematical proof at least once within that unit.

Thus, higher levels of "cognitive activation could be expected. Different kinds of cognitive outcomes (procedural and factual knowledge, geometrical problem solving, understanding of the Pythagorean theorem, and understanding of proof) were measured (a) at the beginning and the end of the school year (global math scores, long-term design) as well as (b) immediately before and after a three-lesson unit on the Pythagorean theorem (content-focused math, short-term design). Thus, conceptual development could be described both from a long-term (global) and a shortterm (thematically focused) perspective. General aspects of instructional quality, especially with regard to cognitive activation, were captured in student questionnaires as well as in high-inference ratings by trained video observers. Detailed content coding assessed which of a set of mathematical aspects related to the theorem had been covered within the lesson. Hierarchical linear models showed that the overall level of cognitive activation has a strong effect on learning gains, especially with regard to the understanding of proof. Analyses of the contingencies between content covered in the lesson and student's understanding of the Pythagorean theorem proved that content coverage has an effect on the kind of understanding which students generate. Four cases with specific profiles of learning gains were studied in detail by interpretative analysis of videotapes. The presentation will also address the issue of differential effects of instruction.

– 270 –

Learning environment and learning motivation in mathematics instruction - differential relations depending on the topic?

Katrin Rakoczy, German Institute for Int. Educational Research, Germany

Christine Pauli, Institute of Education, University of Zurich, Switzerland

Eckhard Klieme, German Institute for Int. Educational Research, Germany

According to self-determination theory (Deci & Ryan) and the person-object approach to interest (Krapp) the quality of learning motivation is a relational concept developing from the interaction between the individual and his or her social environment. As a result, the subject which is to be learned plays a crucial role for the students' quality of learning motivation. The present study takes a closer look at the motivational regulation in mathematics instruction by comparing different topics with regard to the question: Does the quality of students' learning motivation show topic-specific relations to the learning environment?

In order to answer this question data of the bi-national study "Quality of instruction and mathematical understanding in different cultures was analysed. The sample consisted of 20 Swiss and 20 German secondary school classes from the two higher school tracks.

The students' learning environment and their self-reported quality of learning motivation were assessed in the context of two different mathematical topics: the introduction to the theorem of Pythagoras on the one hand, and the dealing with word problems on the other hand. An objective perspective on the learning environment was realized by videotaping parts of these teaching units. The subjective experience of the learning environment was assessed by questionnaire.

In order to investigate the relationships between the different perspectives on the learning conditions and the students' learning motivation in both mathematical topics, hierarchical linear modelling was conducted. The results of the hierarchical linear modelling show that differential relationships between the perceived learning environment and the quality of students' learning motivation across the two different mathematical topics do exist. The discussion of results is going to focus on the question to what extent the observed learning environment confirms or explains these relationships.

– 271 –

The motivational importance of introductory lessons against the background of the expectancy-value model of achievement motivation

Alex Buff, Zurich University of Applied Sciences, Switzerland

Kurt Reusser, Institute of Education, University of Zurich, Switzerland

Christine Pauli, Institute of Education, University of Zurich, Switzerland

The first impression is (often) crucial!? As we have all experienced ourselves, our first encounter with an object often leaves a deep impression, which even in the face of later contradictory experiences proves to be fairly resistant. Introductory lessons can, in this sense, be seen as "first encounters, which means that from a motivation theoretical perspective they can be particularly important. To put it another way: If the introduction to a topic is "messed up in any way, this can have an unfavourable influence on further learning. In this contribution, the question of the motivational importance of introductory lessons (Pythagorean theorem), or specific instructional features of them, are discussed against the background of the expectancy-value model of achievement motivation by Eccles and Wigfield. Examined are (1) the importance of instructional features of the introduction for the affective experience of the pupils directly after the introductory lessons, (2) the importance of instruction and affective experience with regard to the expectancy and the task value components of the model prior to the test on Pythagorean theorem, (3) the effects of dispositional features (control beliefs and interest), which were measured at the beginning of the school year, and finally (4) the dimensions analysed in questions

1-3 are related to the test achievements, controlling among other things for the general ability in mathematics, which was also measured at the beginning of the school year. The analyses are based on the statements of 954 pupils from 40 classes, who participated in the bi-national study presented in this symposium over the course of one year. The special features of the study lie in the consistent longitudinal design in the recording of the relevant dimensions, incorporating different data sources as well as the measurement of the constructs on different hierarchical levels: dispositional vs. situational.

– 272 –

C 23 24 August 2005 08:30 - 10:30 Room E010

SIG Invited Symposium

Conceptual Change

SITUATED CONCEPTUAL STRUCTURES

Chair: Gunilla Petersson, Karolinska Institute, Sweden

Organiser: Gunilla Petersson, Karolinska Institute and Stockholm University,

Sweden

Kaarina Meerenluoto, University of Turku, Finland

Discussant: Erno Lehtinen, University of Turku, Finland

Stella Vosniadou, University of Athens, Greece

Knowledge is always situated. According to sociocultural analysis and situated cognition it is situated in practical settings and realised as management of cultural tools. In established constructivist theories knowledge is embedded in conceptual nets which form conceptual structures. Although context is a fundamental part of both approaches, it stands for quite different things; in one it refers to communicative settings and in the other to conceptual systems. The debate between these two approaches has given rise to a number of questions. Can we, for example, explain students' interpretations of a study task in an educational setting by condensing one theory into the other? Is it fruitful to try to unify the two approaches? How should we account for the situated aspects of cognitive function and for individual variations in a situation from a sociocultural perspective? This symposium aims to highlight these issues and is thus focussing on context and its meaning in sociocultural as well as in constructivist theories.

Situating conceptual structures in physics: What are we looking at?

Claudia von Aufschnaiter, University of Hannover, Germany

Within the last years, an increasing body of research in science education has addressed theoretical foundations of students' learning of science. Theoretical accounts often address either the social aspect of learning or individual cognition.

The paper argues that the theoretical framework chosen very much depends on the frame of reference. In order to understand how students develop their knowledge about a specific subject while acting and thinking in a learning community we might not describe and theorize how experts act nor should we refer to outcomes

– 273 –

of pre-post-testing. Rather, we may consider how a learner experiences the social and material world around him or her and makes use of this world for both developing his/her own understanding and creating opportunities for other learners to gain new knowledge. The framework used to investigate empirical data on learners approaching the domain of physics is based on individual experiences and individual learning in relation to the social and material world. Furthermore, it addresses some more general ideas of neurobiology in order to distinguish cognitive structures from cognitive processes as well as to describe how the notion of ìsituated may refer to

(mental) activity.

Studying the (con)textuality of explanations in the learning of science

Kristiina Kumpulainen, Unversity of Oulu, Finland

Marjatta Kangassalo, University of Tampere, Finland

Satu Vasama, niversity of Oulu, Finland

In the traditional conceptual change research tradition, explanations are often investigated as monological acts of language and cognition, witnessing the nature of conceptual understanding the individual holds of a particular scientific phenomenon.

Typical to this strand of research has been to de-contextualize explanations from the social and physical contexts in which they are created. Less attention has been paid to the sociocultural contexts of explanation activity. In following the sociocultural and discursive approach to conceptual thinking and learning, this presentation discusses the potential of intertextual analysis to illuminate the contextual nature of explanation construction in the learning of science. The presentation draws on an empirical study that examined the intertextual elements of Children's explanations in a technology-enriched early years science classroom (N=22). The overall goal of the study was to investigate how the inquiry-based science unit, including its tools and activities, created the children social spaces to engage in the activity of explaining. Micro-level analyses of video-taped and transcribed data covering a 5-month period reveal that inquiry-based science learning activities enriched by a multimedia science learning tool, PICCO, promoted learners' active explanation construction. The Children's explanations were found to draw on multiple contexts, namely on textual and material links, hands-on explorations as well as on recounting events. These intertextual linkages functioned as important tools for the children (a) to share and validate previous experiences as sources of knowledge, (b) to establish reciprocity with each other, (c) to define themselves as learners of science and as individuals with specific experiences and background, and (d) to construct, maintain and contest the cultural practices of what it means to do and learn science

– 274 –

in the classroom.

Communicating understanding in a primary mathematics classroom: Towards a community of practice

Raymond Brown, Griffith university, Australia

Peter Renshaw, Griffith university, Australia

This study investigates the emergence of a community of practice within a primary classroom. Through employing detailed analyses of video/audio-taped transcripts, teacher/student journal entries, and students' work samples, the study investigates a Year 7 classroom's ways of coming to know, do, and value mathematics over the course of one school year. One of the concerns of the study was to provide insights into whether students use cultural tools such as mathematical notation systems to communicate meaning and to promote understanding when they operate in a classroom culture that provides them with the support necessary to positively experience the ways in which mature communities of practice communicate. The communicative activities of mature communities of practice may be said to be characterised by the expectations that (a) the products of members' efforts will undergo critical comment, (b) individuals will communicate with each other as equals, and (c) correctness and plausibility are to be found through members engaging in the discourse practices of the group. The study found that students do use mathematical notation systems to communicate meaning and to promote understanding when they are engaged in classroom talk that encourages them to change the relationship between themselves and the tools that they use to communicate their mathematics. It was found that the culture of the classroom influenced students' moves toward more mature forms of numerate behaviour by enacting norms that encouraged them to present their ideas and opinions, to accept that their presentations may not always be adequate, and to give truthful feedback and reports. It was found also that student use of mathematical notation systems to communicate understanding required the teacher and students to co-construct skills (e.g., listening, critiquing, evaluating) over time and to challenge each other to extend the quality of teaching and learning that takes place in the classroom.

Reasoning, categorising, and conceptualisation in situated practices

Asa Makitalo, University of Gothenburg, Sweden

Roger Saljo, University of Gothenburg, Sweden

Categorising is fundamental to human activities. Our perception, our memory, our

– 275 –

conceptual knowledge, and all our cognitive activities are grounded in distinctions of a categorical nature. Categorising is obviously one of the most important mechanisms through which human experiences are shaped and communicated. But we also need to recognise that categorising is fundamental at the collective level. Institutional practices such as those taking place in research, in courts of law, bureaucracies, and in many other settings are coordinated by means of categories. We would argue that categorising, and the uses of conceptual knowledge, must be understood in terms of this tension between individual knowing and collective practices. In research on conceptual learning and conceptual knowledge, concepts are generally studied as ready-made entities that people are supposed to acquire and master.

The concepts appear as finished and closed entities, and they preferably should be grounded in scientific discourse in order to be legitimate. The extent to which people master these normatively sanctioned concepts, then, serves as a measure of conceptual learning. In our perspective, conceptual knowing is a natural element of human practices. Such knowing emerges in situated activities precisely because it corresponds to local needs and because it has the coordinating and rhetorical functions that are productive. Furthermore, we argue that conceptual knowledge should not be understood primarily as abstract linguistic entities that somehow represent reality but rather be conceived, and studied, as parts of the concrete practices of various activities. In the present study, we show how conceptual tools simultaneously serve as elements of individual reasoning, of collective communicative practices, and of the computerized technology that has is used to coordinate activities.

C 24 24 August 2005 08:30 - 10:30 Room E111

Symposium

Metacognition

THE NATURE OF METACOGNITION

Chair: Marcel V. J. Veenman, Leiden University, Netherlands

Organiser: Marcel V. J. Veenman, Leiden University, Netherlands

Discussant: Anastasia Efklides, Aristotle University, Greece

Metacognition appears to be one of the most powerful predictors of learning (Wang,

Haertel & Walberg, 1990). Since Flavell coined the term in the seventies of the last century, however, consensus about definitions, about constituents, and about measurement methods of metacognition has been far out of reach (Veenman, in press).

– 276 –

This symposium is intended to contribute to a more comprehensive view on metacognition and its relation to other learner characteristics, such as developmental processes, compliance to demands of the task, intelligence, and motivational processes. Moreover, the undertow of these issues is the intricate relation between cognitive and metacognitive processes. Metacognition regulates cognitive activity, but at the same time it needs cognitive activity as a vehicle. For instance, checking the outcome of a mathematical procedure requires the cognitive activity of recalculation. In the presentation of Panaoura and Demetriou metacognition, the relation between cognitive and metacognitive processes is studied from a developmental perspective with very young children. Broekkamp et al. present a review study on how metacognition regulates the adaptation of students to self-imposed and externally enforced task demands. Meijer et al. investigated the relation between intelligence and the mere quantity of various metacognitive activities. Finally, Bendorf tries to uncover the relation between metacognitive and motivational processes through an intervention study. In various ways, all presentations deal with the relation between metacognitive and cognitive processes.

The interplay of the development of processing efficiency, working memory and self-representation with the development of mathematical performance

Areti Panaoura, UCY, Cyprus

Andreas Demetriou, University of Cyprus, Cyprus

The human mind includes three fundamental levels of organization: the processing system, the environment oriented level and the self-oriented level which involves metacognitive abilities. Although recent research on mathematics education has focused on metacognition and its influence on problem-solving ability, we know little about the relationship between cognitive and metacognitive processes. The focus of the present study was the investigation of the interplay of the development of processing efficiency, working memory and self-representation with the development of mathematical performance. A series of three repeated waves of measurements were done at 126 pupils (3-5 grades). Dynamic modelling was used to specify the nature of change in the main aspects of the mind and the possible interrelations in the patterns of change in these aspects. After testing several models we found that growth in each of the abilities was affected by the state of the others, especially the state of processing efficiency. Specifically, the existence of significant correlations among different cognitive abilities, especially between processing efficiency with working memory and cognitive performance in mathematics suggest that growth in each of the abilities was affected by the state of the other variables, especially the

– 277 –

state of processing efficiency at a given point of time. From the analysis of dynamic models it was quite clear that the processing efficiency had a coordinator role on the cognitive and the metacognitive system. Additionally results of the dynamic models indicated the coherence of the metacognitive system. The lack of relations between self-image and cognitive abilities suggested that growth of self-image was not directly affected by the state of the processing efficiency or working memory at a given point in time. Actually the advancement on self-image depended on the advancement of mathematical performance.

Strategic flexibility: A special case of metacognition

Hein Broekkamp, University of Amsterdam, Netherlands

Bernadette H. A. M. van Hout-Wolters, University of Amsterdam, Netherlands

Marcel V.J. Veenman, Leiden University, Netherlands

Strategies are at the heart of metacognition. By choosing, constructing, executing, monitoring, evaluating and revising plans of action, students regulate and become aware of their task performance and learning. Engaging in these metacognitive processes, however, is not enough. It is also the quality of these processes that determines the effectiveness of task performance and learning. A crucial quality criterium is goal directedness. The degree to which metacognitive processes are directed towards task goals will determine the likelihood that these goals are actually met.

A distinction can be made between goals that learners set for themselves (personal goals), and goals that are defined by others or the environment in which the task is performed (task demands). We present a review study about students' attunement of study strategies to the particular information and types of processing that they should emphasize while studying (i.e., content and processing demands). For instance, students may rehearse information or create mnemonics to meet verbatim reproduction demands, while they may self-generate questions or reorganize learning materials when integration or deep comprehension is required. Similarly, students may pay special attention to text parts that are relevant to the task. Such task attunement draws heavily on metacognition. Part of the metacognitive knowledge and skills is needed to execute the strategies properly. Other metacognitive knowledge and skills are needed to analyze task demands and choose or tailor strategies according to these demands. We refer to the latter type of metacognition as students'

ìstrategic flexibility. So far, strategic flexibility has been examined in limited ways.

It is clear, though, that effective task attunement is not only dependent on the student but also on the environment. Teachers can foster students' task attunement by being clear about task demands and by providing metacognitive instruction aimed

– 278 –

at the development of students' strategic flexibility.

Metacognitive activity, intelligence, and learning results

Joost Meijer, University of Amsterdam, Netherlands

Marcel Veenman, Leiden University, Netherlands

Bernadette Van Hout-Wolters, University of Amsterdam, Netherlands

In this study into the relation between metacognitive activity, intelligence and learning outcomes, metacognitive activity was operationalised by nominal descriptions of mental behaviour observed in thinking-aloud protocols rather than by quality judgments. A taxonomy of metacognitive activities that was developed earlier, was revised in order to achieve sufficient interrater agreement. The framework of the taxonomy is built around the superordinate categories orientation, planning, execution, monitoring, evaluation and elaboration. Thinking-aloud protocols from 43 thirteen-year olds reading texts about history and physics were analysed using the taxonomy. Before the thinking-aloud sessions, participants had been tested for their intelligence and prior knowledge about the subjects of the history text and the text about physics. After the thinking-aloud sessions, posttests about the history and physics subjects were administered. Analysis of the data revealed that for the history text and tests, intelligence, metacognitive activity and test performance were unrelated. Posttest performance was only determined by pretest performance. For the physics text and tests, it appeared that intelligence determined posttest performance indirectly, through the pretest. Apart from executive activities, metacognition did not show any effect, neither on the pretest, nor on the posttest.

Fostering metacognition and learning strategies in commercial education

Michael Bendorf, Georg-August-University, Germany

Self-regulated learning becomes more and more important and, meanwhile, is regarded as one of the most popular objectives in commercial education. The question is how the constitutive components of self-regulated learning - learning strategies, metacognition, and motivation - can be promoted in learning situations. We prefer a direct and domain-integrated approach, which simultaneously promotes motivation, metacognition and learning strategies. In our project ìPromoting self competence, domain competence and functional competence in the commercial subjects at the

Fachgymnasium Wirtschaft we aim for enabling eleven graders to act adequately in complex economic situations. It is assumed that motivation and strategic learning behavior play an important part in the development of these required competencies.

– 279 –

This contribution gives an insight in our approach to promote learning strategies and metacogntion through learning diaries; thereby the influence of motivation is taken into consideration. We assume that motivation and strategy trainings are necessary to support intrinsic kinds of motivation and the use of metacognion and strategies in their contribution to promote competencies. To investigate the effects of motivation and strategy/metacognition training on learning performance we conduct our study in 15 classes at five schools; thereby, we vary both kinds of training as independent variables. In this context, two critical aspects have to be mentioned. First, the use of strategies does not automatically lead to better learning performance.

Secondly, the relationship between the subjectively estimated learning behaviors in questionnaires and the actual context-specific strategy use is not yet sufficiently clarified; therefore, we conduct summative and formative assessments to highlight this relationship. This contribution has its focus on the development of strategies and metacognition - under particular consideration of the influence of motivation in learning. Our study started in August 2004. First results are reported.

D 1

CIT Sessions

24 August 2005 11:00 - 12:20 Computer Labs

EVALUATION OF THE IMPACT OF A TOOL FOR COMPOSING PER-

SONALISED STUDENT RATING QUESTIONNAIRES IN COMPARISON

WITH A STANDARDISED QUESTIONNAIRE

Marjoleine Breda, University of Leuven, Belgium

Kim Waeytens, University of Leuven, Belgium

Mieke Clement, University of Leuven, Belgium

Validity and reliability are considered important aspects of student ratings when the results are used in summative evaluations, but less so when student ratings have no other goal than giving feedback to the faculty. In a formative context, the quality of student ratings can be expressed in terms of its power to provoke reflection, change in teacher thinking and teaching practice rather than psychometric qualities.

Questionnaires gain power to change teaching according to the results of the rating when they are in line with teachers' conceptions of teaching (Johnson, 2000). Optikwest is an on line tool that helps faculty construct a questionnaire that is in line with their subjective theories and educational practice. The tool is designed to provoke reflection. The tool introduces frameworks that need to be used by the faculty

– 280 –

to describe their educational practice and subjective theories as a first stimulus for reflection. As a second stimulus for reflection, feedback on their subjective theories and educational practice is given. The student feedback is the third stimulus for reflection. Opti-kwest is evaluated by comparison to a standardised questionnaire

(EVALEC; Janssen, 1993) on the criteria of power to foster change en provoke reflection. Structured interviews were conducted by phone with users of either one of the instruments. Both qualitative and quantitative information was gathered. Preliminary results show that the design of Opti-kwest effectively provokes reflection when the questionnaires are composed. There does not seem to be a large difference between Opti-kwest and EVALEC with regard to the power to provoke change. It is suggested that contextual factors might influence the perception of the users and the impact of the instruments and that embedding a questionnaire tool in a larger offer for faculty improvement might augment its impact.

ConceptOrganizer: a Multifunctional Tool for Structuring Content

Lassi Nirhamo, Univ. of Turku, Finland

Constructing structured models of concepts and their interrelationships requires usually a full platform or specialized software. These models can be used in learning to facilitate computer supported collaborative learning, to help in developing argumentation skills, and to support knowledge building. Using a full platform in common every-day learning situations especially in educational institutions can cause problems because its seamless integration to working practices is not easy to achieve. Specialized software works well as an individual tool but the collaborative use of individual tools has proven to be a real challenge. To address these problems

Educational Technology Unit from University of Turku has developed ConceptOrganizer, a multifunctional tool for structuring content. ConceptOrganizer is a small

Flash application which has a very simple but powerful user interface. It provides means to place ìcontainers to application's desktop and define their relationships.

Types of relation lines and the number of relation types are fully user definable.

ConceptOrganizer's input information about available containers and output information about container locations and relations are in XML format. By using XML as communication format the data to and from ConceptOrganizer application is easily processed or analyzed with other applications. ConceptOrganizer's Flash source code is freely available to everyone and additional language support is easily added by only tagging appropriate terms in external language text file. This flexibility makes it possible to use ConceptOrganizer in various settings. In the University of

Turku ConceptOrganizer application is integrated in learning platform WorkMates.

– 281 –

To give users different view to content, ConceptOrganizer application is used in

WorkMates as a graphical interface to materials. This makes it easy to teachers to give the students e.g. wanted view to course materials. There are some research projects planned to gather information about actual usage and to guide further development of ConceptOrganizer.

D 2 24 August 2005 11:00 - 12:20

Thematic Poster Session

Assessment and evaluation

Discussant: Constantinos Papanastasiou, University of Cyprus

D 1 The Diary as a methodological tool for accessing the learner perspective on a learning environment

Margaret Brady, UEA, United Kingdom

This paper looks at the diary as a means of gathering data on the learning environment from the perspective of the learner. The diary helps the inquirer to gain insight into motivational and emotional aspects of learning and to capture the immediacy of a learning experience. The added value of using a diary approach is the learning that may be gained by encouraging the learner to take a reflective stance towards their work. Time spent on looking forward in a reflective frame of mind may reduce anxiety and maximise the learning that may be gained from the experience ahead. The paper will discuss the experience of the author of using the diary as a research instrument in a pan European Research Project in order to gain the learners' perspective on an ICT learning environment which was implemented in a crosssection of schools through the action of the project. A comparison is made between the merits of two forms of diary used to collect evidence. The main themes that emerged of significance for learners were collaborative work with cross-national partners and issues in relation to the telematics environment. It is concluded by the paper that the diary is instrumental in gaining access to the perspective of the learner and in providing insight into emotional and affective aspects of learning. As a method of investigation it can filter evidence for more focused inquiry and may be triangulated with other methods of investigation in order to increase the credibility and dependability of findings. Besides providing information about the student perspective, it stimulates reflectiveness in individuals and among groups which helps future learning, opening the mind to new ways of looking at problems to be encountered.

– 282 –

D2 The role of time of assessment in collecting expectancies in process-oriented course evaluations

Katrin Kahmann, University of Regensburg, Germany

Regina Mulder, University of Regensburg, Germany

It is a characteristic of the -integrative evaluation to collect expectancy rates of participants in addition to evaluations (Henninger & Balk, 2001). Measuring expectancies has proven to be a helpful source of information. The question arises if there is a suitable moment to collect expectancy rates. In order to find an answer to this question two argumentation lines were taken. The first argumentation was found in the action-theoretical aspect of evaluation and in the process-oriented evaluation

(Frese & Zapf, 1994; Henninger, Balk & Mandl, 1998). The basis for the second line of argumentation was if expectancy rates of participants may change over the time of a course. The first line of argumentation implied, that expectancies should be measured at the beginning of the course. The theoretical outcome of the second argumentation showed, that adaptive processes of expectancies can take place during the course, which can lead to changes in the expectancies. The empirical verification of this thesis gave no clear results. There was no significant difference of expectancy rates at three different moments of collecting expectancies. But the participants surveyed at the end of the course stated, that they could not remember exactly the expectancies they had at the beginning of the course. Comprising it was the result of the study, that expectancy rates should be collected as early as possible in a course. This inquiry is a starting point for further research in course evaluation.

As a next step an ìadaptive online evaluation toll is due to be designed. We will elaborate on that during the presentation.

D3 MAPS - Measurement & Assessment of Problem Solving

Eveline Wuttke, Johannes Gutenberg-Universitaet Mainz, Germany

Karsten D. Wolf, Otto-Friedrich Universitaet Bamberg, Germany

While researchers and educators consider problem solving as basic skill needed by today's learners, very little is still done to help students to acquire those skills. This is partly due to the fact that - apart from not knowing how to teach them - teachers dont know how to assess complex problem solving skills. Our work within this field is based on a theoretical framework (Sembill 1992) comparable to the IDEAL framework (Bransford and Stein 1993). It is called Analytical Ideal Type (AIT) of planned action and provides in a first step design criteria for complex learning environments were students solve problems. In the second step it can be used as an

– 283 –

instrument to evaluate quantity and quality of students' solutions to problems. The instrument measures how they:

1. analyze the given situation,

2. specify goals,

3. develop adequate strategies and

4. (mentally) control, if the solutions are adequate and the problem is solved.

Unfortunately the AIT system is extremely time-consuming and therefore not really usable for teachers. Another critical point is that the results of the tests are highly dependent on the student's motivation to try hard in the problem solution and their ability to put down a solution in writing. Therefore we are currently developing the

AIT further into an instrument called MAPS (Measuring and Assessing Problem

Solving). In contrast to the AIT method the MAPS test supplies detailed questions along with the problem description. These questions are intended to help students with the problem solution and "trigger" answers evens from students who are not highly motivated to provide written solutions. Results concerning the quality of the instruments and first results about problem solving skills from various pilot studies will be reported at the conference.

D4 The utrecht early mathematical competence test in a spanish sample

Manuel Aguilar-Villagran, University of Cadiz, Spain

Jose Navarro Guzman, University of Cadiz, Spain

Concepcion Alcalde, University of Cadiz, Spain

Esperanza Marchena, University of Cadiz, Spain

Gonzalo Ruiz, University of Cadiz, Spain

A large number of reports linked with early mathematical competence in pre-school children were published in the nineties. Some studies considered skills and abilities of young children with numbers using longitudinal procedures. Results show huge mathematical competence differences for school beginners. They have important consequences in mathematic teaching: if we know differences, an accurate learning program would be feasible. Within the Spanish educational context, early mathematical competence tests are not common. In this presentation an Spanish adapted version of The Utrecht Early Mathematical Competence Test (EMTC) (van Luit, van de Rijt, & Pennings, 1994) is presented. A total of 128 children aged between

4 and 5 participated in this study. They attended kinder garden school, came from middle-class family backgrounds and were socially adapted. The Utrecht Early

Mathematical Competence Test (Version A) was individually administered to each participant in one session, with the prior agreement of their parents. EMTC assess

– 284 –

eight aspects of mathematical competence: Concepts of Comparison of quantitative and qualitative characteristics of objects; Classification of objects in class o subclass; Correspondence one to one relation; Seriation of objects in class or subclass based on criteria; Using counting words, forward and backward; Structures counting, synchronous counting, shortened counting from the dice structure; Resultative counting, structured and unstructured quantities as well as counting hidden quantities; and General knowledge of numbers, being able to use knowledge of the number system in simple problem condition. Each subtest has five questions. After the administration, the given answers were judged on their correctness with the help of the EMTC scoring key. Then, the children test scores were transformed in a competence score. This hematical ability. Finally, in order to check the EMTC fiability, results were compared with the experimental version of the Basic Mathematical

Competences Assessment Test (TEDI-MATH) designed for Spanish children.

D5 Diagnosis and treatment of developmental disorders with computer based targeted exercises

Gabriella Blenessy, ELTE TTK, Hungary

In Hungary some 7 to 17% of children in school age are affected by some kind of developmental disorder. If we compare this number with the low number of available trained therapists we understand the need for a new approach in this field. One solution would be to develop tools for therapy which could be used for independent work with only little assistance from the part of a trained staff. These tools should be accessible for schools, therapeutical institutions and other organizations which target children. If parents could help their children in using these toolsóbased on the advice of expertsóthat would greatly enhance the situation.

D6 Teacher education students’ evaluations of their problem-based learning experiences

Rosalind Murray-Harvey, Flinders University, Australia

Helen Askell-Williams, Flinders University, Australia

This paper reports the application of a conceptual framework to teacher education students' evaluations of the effectiveness of problem-based learning (PBL) environments for teaching and learning. The framework permitted comparisons between students' evaluations and the aims of their teacher education program.

Much PBL research has concentrated on medical and health-related fields, and on

– 285 –

student achievements in content knowledge measured by test scores. Many less easily measured qualities that PBL purports to foster, such as collaborative skills, critical thinking and problem-solving, self-directed learning (SDL), and developing professional identity, are generally not researched. They are, nevertheless, desirable outcomes of university study and are explicit aims of many programs, including teacher education. To evaluate the effectiveness of a PBL environment for developing these less easily measured qualities we developed a conceptual framework to analyze students' written evaluations of their PBL experience in their teacher education program at a South Australian university. The framework contains five categories that reflect key aims of both the B.Ed program and concepts in extant

PBL literature: knowledge; critical thinking; theory-practice relationships; collaboration; and self-directed learning. Using the conceptual framework we examined students' manuscripts with NVivo to map the scope and complexity of their ideas about teaching and learning. The framework proved to be an informative tool for mapping and interpreting students' perspectives of structural (PBL tutorial format, group work) and functional (personal, interpersonal, professional skill development) contributions of the PBL environment for their learning. Key findings include that students (1) perceive the value of case studies for engaging with subject content, motivating learning and connecting theory with practice, (2) develop awareness of reflective practice and professional identity, (3) recognize links between PBL and

SDL, and (4) build collaborative group learning skills. The paper elucidates ways in which desired outcomes of professional education programs may be realized through PBL.

D7 Teacher Credentialing Reform and Candidate Assessment: Challenges and

Dilemmas

Anne Hafner, California State University Los Angeles, United States

Many states in the USA have passed teacher credentialing reform laws in response to a national accountability push. This paper reports on the impact of one state`s credential reform on candidate assessment. A mixed methods approach was used, with a web survey and site visits. The study found that the reform influenced major changes in assessment. Most programs embedded candidate assessment into their classes and gave a summative evaluation of teaching. Although respondents were positive about having candidate outcome data, some were negative about the prescriptive nature of the mandate. There was great concern about the resources needed to implement assessment. One lesson is that although programs view candidate assessment as valuable, sustaining high quality assessment requires considerable

– 286 –

resources of time, personnel and training.

D8 Peer group assessment of teaching competency

Erica McAteer, University of Strathclyde, United Kingdom

Tracy Maluleka, University of Glasgow, United Kingdom

Teaching competency can be defined, and assessed, in various ways in Higher Education institutions. The Dearing recommendations influenced UK policy throughout the mid 1990s and led to the development of the Institute for Learning and

Teaching in Higher Education, now re-formulated as the Higher Education Teaching Academy, and the growth of accredited courses in Higher Education teaching. This paper reports on research into the application and use of generic criteria to peer-assess teaching competency in Higher Education, querying other research positions (Neumann: 2001, McLean: 2002) which argue against this practice. The

New Lecturer Programme at the University of Glasgow provided a case for specific examination of the use of generic criteria to assess the teaching competencies of participants. The intra and interdisciplinary peer relations of the cohort during the period and the resultant social learning claimed by members of the group was an additional outcome of the study and indicated success of the programme in converging the various disciplines and facilitating a unified conception and achievement of teaching competency among the participants. A major focus of the study was the efficiency and effectiveness of peer assessment procedures across disciplines traditionally known to have diverse educational orientations. The use of generic assessment criteria in the NLP resulted in the generation of a common conception of teaching competency throughout the various disciplines that differ in their educational orientations and discouraged disciplinary disunity as far as teaching competency is conceived while simultaneously making provision for disciplinary relevance.

Culture and Education

Discussant: Helen Ftiaka, University of Cyprus

D9 Colour Coded? How well do learners of different race groups really mix in

South African schools? Eight graders share their views. Telling it like it is

Saloshna Vandeyar, University of Pretoria, South Africa

This study sought to interrogate the quality of contact between learners in desegregated schools in South Africa and to examine ways in which these schools have

– 287 –

elected or omitted to adopt certain strategies with the opening of racially exclusive schools. These strategies were probed in light of the different discourses in the debate about race and desegregation. Semi-structured interviews with Grade 8 learners from three secondary schools as well as school observation and field notes provide the evidentiary base of this paper. There are three major research findings emanating from this study: first, the institutional culture of the school has remained relatively intact; daily operational principles of these schools are strongly embedded within the existing hegemonic culture. Second, learner identities are shifting and there is an emergence of new self-identities and third, the polarities of the status quo versus transformatory reform are currently being contested by learners in the schooling environment.

D10 Comparing Chinese and Western: the shifting cultural categorisation work of two parents in an interview

Juliet Choo, Griffith University, Australia

Peter David Renshaw, Griffith University, Australia

Helena Austin, Griffith University, Australia

This paper is based on a wider study on cross-cultural collaboration between professionals and parents in an Australian special education context. Rather than invoke and rely upon a notion of cultural dissonance between the Western trained professionals and parents of diverse cultural background as some sort of explanation for low parental participation, we employ sociocultural approaches and Membership

Categorisation Analysis (MCA) to examine the dynamic ways in which parents and professionals deploy cultural categories in the interview-talk. Further, the analysis seeks to find out what work is accomplished in deploying cultural categories and what attributes are assigned to these cultural categories. In accord with the tenet of ethnomethodology to avoid pre-theorisation in analysis, only those cultural categories that participants made relevant in the talk were taken for analysis. This paper is drawn from the case example of the Chinese parents of a child with autism spectrum disorder (Timothy). Analysis of the interview shows the parents to report culture as irrelevant in their relationship with professionals. However they deployed the cultural categories Chinese and Western when talking around the theme of parenting and education. The parameters of what they talked about change and slide through the interview, revealing the layering of complexity in which they manage their situations and, also, the importance of context in their moment-to-moment categorisation work. For instance, when describing their non-disabled daughter, the parents deployed the cultural categories Chinese and Western to account for their parenting

– 288 –

practices as effective. This category pairing changes as their category work became more delicate and refined when they moved on to comparing Timothy and his sister.

The analysis shows the complex and fluid ways in which the participants assigned traits to cultural categories, shifting among a general-trait account, a cautious and situated trait explanation, and an orientation to hybridity.

D11 Chats behind the Chig - How Bedouin teenagers perceive ICT and online

Chats

Gadi Alexander, Ben Gurion University, Israel

Adnan Grebea, Ben Gurion University, Israel

The present study attempted to find out how and why the online chat is used in schools by Bedouin teenagers and how they perceive it in relation to the expectation of their traditional community. A questionnaire was administered to 203 young

Bedouin high school students. Some of the topics covered were: common uses of the internet, criteria for selection of chat portals and communities, presentation or false presentations of ones identity, etc. In addition, they were questioned about their perception of the norms of their home culture, and about possible matches or mismatches between these norms and the chat activity. It was found that the chat is perceived by these teenagers as playing a significant role in their socialization process. They use chats to find virtual friends, seek opportunities to communicate with the other sex, etc. However, a repeated claim was that chats can provide a bypass allowing them to take part, at least virtually, in a kind of discourse that would not be acceptable in their own community. The chat is a potential arena in which norms and rules of the elders can be stretched or even violated without having to pay an immediate social price. In addition, it expands the social network of the teenager, and creates a space were various identities of the Bedouin youth can be virtually experienced. This study has many implications for educators who attempt to learn about the social life of their student and create a bridge between the culture of the web and the school culture. It sheds light on unintended uses of ICT in school which have to be reevaluated and screened on the basis of a changing reality in which technology is used as both an outlet and a trigger for many social and educational changes.

– 289 –

D12 Testing a Model of the Adaptation of Mainland Chinese Postgraduate Students to the Universities of Hong Kong

Min Zeng, University of Hong Kong, Hong Kong

Recently the People's Republic of China has become an increasing source of international students for universities all over the world. However, it is surprising that little research has been directed to the adjustment of such students. This lack is particularly true at the postgraduate level at which more and more of these students are studying. The same lack of research applies to Hong Kong whose universities have attracted large numbers of Mainland Postgraduate students in the last ten years.

Their adjustment to studying in Hong Kong may be different to studying at Western or Mainland Chinese universities. Based on the previous theory that the academic integration and social integration are the major predictors of student persistence and satisfaction, this study proposed and tested a model on the adaptation of MPS in a sibling cultural setting--Hong Kong. Unlike the most previous research, which focused on either institutional factors or cultural factors, this research explored the influences of both factors on student adaptation. The instruments were modified according to MPS' situation. The participants were 222 current students from four universities in Hong Kong. Results indicated that both academic integration and social integration were strongly correlated with persistence while academic integration was more strongly correlated than social integration to student satisfaction.

Among the background variables measured, motivation, Cantonese proficiency and self-evaluated English language skills showed significant correlations with students' academic integration, social integrations, satisfaction and persistence. The general support of the data on the model and hypotheses presents implications for the faculty and university administrators that, in this sibling cultural setting, Mainland Chinese students' adaptation may be similar to that of domestic undergraduate students in previous American studies. However, as research students, academic integration may play a more important role in student satisfaction. Other implications from the study were also discussed.

D13 Factors Influencing the Academic Achievement of Chinese High School Students

Guifang Fu, University of Jilin, China

Kurt Reusser, University of Zurich, Switzerland

This study aimed to investigate the effects of control expectation, task value, selfefficacy, achievement goals, learning strategies, effort and perseverance on achieve-

– 290 –

ment by means of covariance structure analysis in a sample of Chinese high school students. Our hypothesized structural model is a conceptual framework that stems from the literature on self-regulated learning, task value, self-efficacy, goal orientation, and learning strategies. A total of 1,968 students in grades 7, 8, 9, and 10 from

5 high schools in Changchun, China were randomly selected to complete a Chinese version of the Motivated Strategies for Learning Questionnaire (MSLQ, Pintrich &

De Groot, 1990). Confirmatory factor analysis was conducted to verify the structure validity of the inventory. Academic achievement consisted of Chinese, mathematics, and English scores on end-of-semester exams. The relationships among the factors were analyzed with LISREL. The results indicated that learning strategies directly influenced achievement. Achievement goals had not only direct effects, but also indirect effects on achievement through learning strategies. Task value directly influenced learning strategies as well as effort and perseverance. Task value, self-efficacy, effort and perseverance all had an indirect influence on achievement through learning strategies. Control expectation had no effect on learning strategies, effort or perseverance. The sequence of significance of the effects of the factors on

Chinese high school students' achievement was task value, effort and perseverance, self-efficacy, achievement goals, and learning strategies. Task value, self-efficacy, achievement goals, effort and perseverance, and learning strategies were able to explain 58% of the variance of achievement. Task value, self-efficacy and achievement goals accounted for 82% of the variance of learning strategies, and 52% of the variance of effort and perseverance. The results suggest that providing training of task value, effort and perseverance, and learning strategies might foster the academic achievement of Chinese high school students more efficiently.

D14 Gender Differences in Language Learning Strategies of College Students

Kai S. Cortina, University of Michigan, United States

Kelsea Lane, University of Michigan, United States

Research on language learning has led to an increased awareness regarding language learning styles and ways to improve the language learning process. As it is true for other domains, e.g. mathematics, one means of improving language learning is through a better understanding of effective learning strategies. Language learning strategies are of particular interest to strategy researchers for a better understanding of the acquisition process of foreign languages. Native language learning is, under normal circumstances, not a conscious process. Yet, when learning a foreign language the process is effortful and planned, and requires the implementation of strategies. This purpose of this study was to explain the gender differences in language

– 291 –

learning strategy use of college students learning German as a foreign language.

In line with the work of Oxford and her colleagues (Oxford, Nyikos, and Ehrman,

1988; Oxford and Nyikos, 1989; Ehrman and Oxford, 1989; Green and Oxford,

1995; and Oxford, 1996), we expected to find gender differences favoring females.

Furthermore, we expected gender differences to decrease significantly when controlling for variables reflecting the language learning history. The variables we hypothesized as contributing to the sex differences are exposure to family members or relatives that speak a foreign language as well as informal and formal experience with foreign languages. Significant gender differences favoring females were found using the Strategy Inventory for Language Learning (SILL) (Oxford, 1996). These gender effects were examined by controlling for exposure at home, informal exposure as measured by understanding, or formal foreign language exposure. Contrary to our hypotheses, these factors did not account for a significant amount of the gender differences noted. Additionally, post hoc analysis addressing a student's desire to continue studying German did not significantly explain the gender differences.

Findings are discussed against the backdrop of gender socialization theory.

D15 The use of information technology and gender: A survey of 8th grade students in Croatia

Iris Marusic, Institute for Social Research, Croatia

Branislava Baranovic, Institute for Social Research, Croatia

Ivana Batarelo, Institute for Social Research, Croatia

Despite the increasing awareness about the importance of computer literacy, the current primary curriculum in Croatia offers ICT only as an elective subject, and up to date there are no studies offering reliable information on the actual use of ICT in the educational settings. Therefore the aim of this research was to investigate the level of computer and Internet use among Croatian elementary school students.

Special reference was made to gender differences, since a number of studies indicate the existence of gender differences in the use of ICT among school students in various countries. The study was carried out on a representative sample of 121 school in Croatia. The sample of pupils consisted of 2674 pupils, 1322 girls and

1345 boys whose average age was 14. The administered questionnaire comprised items related to the use of computers at school and elsewhere, and to the assessment of the optional school subject of ICT on the following five dimensions: interestingness, comprehensibility, difficulty, usefulness for current life and importance for the future life. Our research findings show that of all places school is the last place where pupils use computers. About three quarters of Croatian pupils never

– 292 –

or rarely get to use computers at school, and even 80% either never use Internet at school or do so rarely. Data reveal some gender differences in the frequency of the use of computers outside school. The assessment of ICT as an elective subject reveals that boys find it easier and more comprehensible than girls, although both genders are equally aware of its usefulness and importance for the future life, and find it equally interesting. These findings have an implication for the creation of future educational policy regarding the inclusion of ICT as an integral part of the school curriculum.

D16 Teaching of lutheranism: what and how do teachers teach and why?

Elina Hella, University of Helsinki, Finland

Paivi Tynjala, University of Jyvaskyla, Finland

In Finland, religious education (RE) as a school subject is called Evangelical Lutheran Religion, because the majority of 85% of students belong to Evangelical

Lutheran Church. Lutheranism can be seen as a complex, multidimensional phenomenon which has traditionally been intertwined with the culture, Christian religiosity and education in Finnish society. However, the role of Lutheranism in religious education is still unclear. This presentation focuses on the question of how teaching of Lutheranism is constituted by RE teachers. What meanings do the teachers assign to teaching of Lutheranism: what do they say that they teach about

Lutheranism, about the way they teach it and about their reasons and grounds for those ways to teach it? The object of the study was approached from phenomenographic research perspective, because of its focus on both the content and structure of meanings in the experience. Thus, phenomenography was considered the most appropriate research approach to explore the qualitative variation in teachers' ways of arguing for their ways of teaching Lutheranism. Data collection and analysis were based on the phenomenographic methodology. 20 RE teachers, 6 male and 14 female from 15 different regions, were interviewed according to semi-structured procedure. Teachers' experiences of their teaching were identified in terms of differences and similarities in both content and structure of meanings. The preliminary results will be presented as the outcome space of categories of description and discussed to develop religious educational practice. Acknowledgement of teachers' descriptions of their teaching of Lutheranism serves as a tool for teachers to reflect and focus on relevant aspects for student learning of Lutheranism according to curricular goals. The results also reveal the role of Lutheranism as a subject matter of religious education as seen by the teachers.

– 293 –

D17 Educational beliefs of teachers of sacred subjects

Shira Iluz, Bar Ilan University, Israel

Yisrael Rich, Bar Ilan University, Israel

This research investigated the educational beliefs of teachers of sacred subjects to girls in Israeli religious high schools. Sacred subjects are unique because they have behavioral, emotional and moral purposes alongside academic goals. This study initiated scientific investigation in this field. We also explored effects of debates in

Orthodox Jewish communities regarding womens' roles in family and society on educational beliefs of teachers of sacred subjects. Perceptions of Bible and Oral

Law teachers were compared to those of literature and biology teachers followed by comparisons between Bible and Oral Law teachers. 255 teachers responded to questionnaires comprising three variable groups: Perceptions of discipline (e.g., well-defined); perceptions of student (e.g., importance of religious development and academic achievement); and teacher as life guide. 15 teachers were interviewed using the CQR method of analysis. Compared to teachers of secular subjects, teachers of sacred subjects perceived their academic disciplines as more static and holy.

They stressed non-academic student features including religious affect and behavior and emphasized teaching for shaping student character. Additionally, teachers of sacred subjects highlighted their role as spiritual guide to female students whereas their secular subject colleagues emphasized academic achievements. Reflecting the debate on the role of women in Orthodox society, Oral law teachers differed on preferred topics of instruction and desired instructional methods. Some protested that Oral Law was intentionally taught at low conceptual levels to girls to maintain male scholarly superiority regarding sacred subjects. Although this study examined educational beliefs of teachers of Jewish sacred subjects to girls, implications are relevant for instruction of sacred topics in other faiths. Furthermore, secular subjects having prominent affective, moral or behavioral components might also be informed by this study (Osler & Starkey, 2002).

Development of expertise in specific domains

Discussant: Csikos Csaba, University of Szeged, Hungary

– 294 –

D18 Cognitive and motivational strategies in learning music theory and students achievement

Barbara Fritz, Elementary Music School Krsko, Slovenia

Cirila Peklaj, University of Ljubljana, Slovenia

The aim of the present study was to find out how students regulate their learning in

Music Theory Learning (MTL). The research is based on social-cognitive theory of learning, which emphasises the importance of integration of metacognitive, cognitive and motivational processes in learning. The aim of our study was twofold: first, to construct the instruments for measuring cognitive and motivational processes in learning MT and second to examine the relationship between these processes and achievement in MTL. 457 fifth and sixth grade students from ten different elementary music schools in Slovenia (153 boys and 303 girls) participated in the study.

Two questionnaires were constructed for the purposes of this study; the Cognitive

Strategies Questionnaire (CSQ) for measuring cognitive processes in MTL and the

Motivational Questionnaire (MQ) for measuring motivational processes in MTL.

The factor analysis of CSQ revealed three different factors: strategies for solving convergent tasks, strategies for solving difficult tasks and strategies for solving divergent tasks. Factor analysis of MSQ revealed four different factors: perception of applicability and importance, anxiety, competence and inner interest, and lack of self-efficacy in listening comprehension. Students general and music abilities were also assessed. Further analysis showed significant correlation between achievement

(results on Music Theory Achievement Test (MTAT) and final grades) and almost all motivational and cognitive factors. The best predictors of results on MTAT were abilities, anxiety and strategies for solving divergent tasks. On the other hand, the best predictors of final grades were competence and inner interest, abilities and anxiety. The implications for educational practice as well as future research were discussed.

D19 Gender and music education: Differences in the attitudes among students in compulsory education

Barbara Muzzatti, University di Padova, Italy

Several studies have appeared on gender- based choices of music instruments, on various music activities, and on the motivations for either studying music or playing a specific instrument, while little research has been published on boys' and girls' attitudes toward music education as provided in Italian compulsory education . The purpose of the present study is to check whether both the gender-based and age-

– 295 –

based attitudes towards musical instruments and activities found in previous studies, are also present in music education as taught in Italian schools. Six-hundred and fifty-two elementary and middle school Italian students took part at the study.

The procedure consisted in administering a questionnaire about pupils' attitudes, beliefs, and feelings toward music education. Results are congruent with the existence of a female characterization of music education as early as the third grade, and seem to document a progressive decline in interest for this subject matter.

D20 Developing a 'big picture': Collaborative construction of multi-modal representations in history

Maaike Prangsma, Utrecht University, Netherlands

Carla van Boxtel, Utrecht University, Netherlands

Gellof Kanselaar, Utrecht University, Netherlands

Research has given us insight into conditions for effective use of pictorial representations in addition to verbal ones. Positive effects were established within the domain of science and technology and in the context of individual use of presented multi-modal representations. Current trends in the field of learning and instruction stress the importance of active knowledge construction and collaborative learning.

The focus of our research is on the active construction of multi-modal representations in collaborative learning tasks in history. We address the question whether collaborative construction of multi-modal representations can contribute to the acquisition of a chronological-conceptual frame of reference that can be used as a basis for historical reasoning. Studies have shown that many pupils have difficulties with the acquisition of a coherent overview of significant historical events and developments and confuse phenomena and concepts. In an experimental study we compared the learning processes and outcomes of pupils who co-construct textual representations, textual-visual representations and textual-visual representations integrated in a timeline. The participants were 100 eleven to thirteen-year-old pupils in pre-vocational secondary education who worked in gender neutral dyads on a series of four tasks on significant historical developments in the Early Middle Ages.

All participants took a pretest, posttest and retention test and peer interaction was audio taped and transcribed. We predicted that pupils who constructed the visualtextual representations would show higher scores on the posttests than pupils in the textual representation condition and that the pupils that integrated the visual-textual representations in a timeline would outperform the pupils in the other two conditions. We present the results of our analyses of pupils' performance and the quality of peer interaction in the three conditions.

– 296 –

D21 The role of knowledge development and ontological change in the development of competence in legal reasoning.

Fleurie Nievelstein, Open University of the netherlands, Netherlands

Els Boshuizen, Open University of the Netherlands, Netherlands

Jan van Bruggen, Open University of the Netherlands, Netherlands

Frans Prins, Open University of the Netherlands, Netherlands

Some domains in law - especially private law - are very hard to learn and to teach.

Problem solving in private law is based on a knowledge base that is not often readily applicable. Although rules are available, the application and applicability of the rules is never a matter of straightforward verification. Law is a system that changes continually. Not only does the law itself change, the interpretation of rules varies with changes in society as well, (e.g., threat of terrorism, globalisation, individualisation, economic change) which makes it even more difficult to determine the applicability of rules. Findings in expert-intermediate-novice studies suggest that the underlying reasons of the difficulties might be the required knowledge structure and its ontological qualities that are intricately linked to domain specific reasoning skills. A first analysis of what makes private law difficult suggests that on top of the development of the right ontology, scripts and concepts, script negotiation and adversarial reasoning must be mastered. In this study we only focused on the question in which way domain knowledge in the field of private law differs between participants with different levels of expertise. The aim of the study is to gain more insight into the way knowledge of private law is structured in participants' minds to get a broader view of the difficulties students experience during legal reasoning. To answer this question a cognitive approach will be taken using methods used in expertise development and expert performance research. In our study we will perform a qualitative analysis on basis of thinking aloud protocols.

D22 Help-seeking as a self-regulatory skill and motivation in learning statistics with a web site

Nathalie Huet, University of Toulouse II, France

Fabrice Noury, University of Toulouse II, France

Paul Chotin, University of Toulouse II, France

– 297 –

This research supports the view of adaptive help-seeking as a self-regulatory skill

(e.g., Puustinen, 1998). As noticed by Aleven et al. (2003), this issue has been mostly studied in traditional school contexts but not in computer-learning environment.

This paper aimed to examine, on a web site in statistics learning, the impact of help-seeking and motivation on learning during problem solving according to the difficulty of the problem. Sixty seven students in psychology enrolled in a distance education had to solve easy and difficult statistics problems on a web site designed for their course. They could use spontaneously helps such as work-out problems, on-line course of statistics. In case of failure, they were presented automatically the helps for correcting their errors as a feedback of assistance (Huth & Narciss,

2001). Motivational variables were achievement goal orientations and self-efficacy and were respectively assessed by using the P.A.L.S (Migdley et al. 2000) and by asking after reading the problem, how sure they feel they could solve the problem.

Results showed that more motivational differences were found on help seeking for difficult problems: (1) both performance-approach goals and performance-avoidance goals were negatively associated to the frequency of help used; (2) Mastery goals were positively associated to performance of problem solving although both performance-approach and performance avoidance goals were negatively related to performance; (3) Self-efficacy was positively associated to the amount of helps requested on difficult problems and performance for both easy and difficult problems.

Besides, high performance-approach goals and high work-avoidance performance goals were positively related to a high number of giving up after a feedback of failure for both easy and difficult problems. Unexpectedly, no relationship was found between mastery goals and help seeking. Future research should investigate the effects of mastery goals on self-regulation in learning with computer environments.

D23 The Evaluation of the Web-based ARTIST, Assessment Resource Tools for

Improving Statistical Thinking

Ann Ooms, University of Minnesota, United States

Joan Garfield, University of Minnesota, United States

Robert delMas, University of Minnesota, United States

As technology expands in educational settings, so does educators' interest in using web-based educational resources. The increased integration of online resources into the curriculum resulted in a growing need for high quality educational resources.

Since there is no quality control in place to determine the credibility, quality, and accuracy of published websites (Branch, Kim, & Koenecke, 1999), educators, and

– 298 –

other users, are in need of a tool to guide them in evaluating the quality of information available on the Internet (Oliver, Wilkinson, & Bennett, 1997). In this paper we present a model for evaluating online assessment resources. We use the Web-based

Assessment Resource Tools for Improving Statistical Thinking (ARTIST) as our example.

D24 The development of conceptual understanding of global warming

Tiina Nevanpaa, Institution for Educational Research, Finland

Conceptual development and structure of existing knowledge have been an issue of debate within theoretical frame of conceptual change. This paper introduces a research project that aims to clarify the development of conceptual understanding in case of abstract and complex phenomena, global warming. Thus the aim of this paper it to examine 1) how pupils` pre-instructional ideas develop on a two-year time scale (from grade 7 to 9), 2) what kind of changes in pupils` ideas take place during a learning period (grade 9) and 3) how these changes can be interpreted within frame of conceptual change? Pupils` pre- instructional ideas were examined by using a questionnaire with free-response items (7th grade, n= 415) and by essays

(9th grade, n=45). Pupils` responses were analysed by using a phenomenographic approach in a practical manner. As a result, distinct categories reflecting qualitatively different ways to conceptualise global warming were generated. Based on this information and literature a 7 hour learning period was constructed and implemented at grade 9. After the learning period pupils` ideas were examined again. All the lessons were videotaped in order to get insight into the learning process of pupils.

Results of this study raises interesting questions about development of conceptual understanding in case o f abstract and complex scientific phenomena. It seems that younger pupils` pre-instructional ideas are based on connection of separate scientific facts to every day reasoning. Proportion of scientific elements increased with age, but still lack of scientific elementary knowledge leads to faulty connection between concepts. However, it seems that older pupils` ideas of global warming were more coherent and theory-like compared to younger pupils conceptions. Thus it can be hypothesised that conceptual understanding of global warming starts from more or less separate facts followed by theory-based reasoning.

Learning and cognitive science

Discussant: Georgia Panagiotou, University of Cyprus

D25 A trialogical approach to learning

– 299 –

Sami Paavola, University of Helsinki, Department of Psychology, Finland

Kai Hakkarainen, University of Helsinki, Department of Psychology, Finland

It has been maintained that a strict confrontation between constructivism and sociocultural perspectives should be given up (Vosniadou & Hallden 2003; Anderson,

Greeno, Reder, & Simon 2000). We also try to alleviate a strict dualism, but rather than analyzing similarities between two approaches, we present them as ideal types with a third approach. A basis for our ideal types is a distinction between two main metaphors of learning (Sfard 1998) and our own suggestion that a third one is needed (Paavola & Hakkarainen in press; Paavola, Lipponen, & Hakkarainen in press;

Hakkarainen, Palonen, Paavola, & Lehtinen 2004). According to the acquisition metaphor of learning (a ìmonological approach), learning happens basically in an individual's mind. In contrast to this, the participation metaphor of learning (a ìdialogical approach) emphasizes communities, and participation to their activities and practices. Rather than epistemological issues, learning concerns identities, practices, social norms. A third approach, the knowledge-creation metaphor of learning

(a ìtrialogical approach), emphasizes mediation, and mediating artifacts. We will present our project for developing the idea of a trialogue as a central feature of the third metaphor. The conception has its basis on L. S. Vygotsky's notion of mediated activity and Charles Peirce's ideas concerning triadic (object- and future-oriented) nature of sign-processes. Our presentation will concentrate on theoretical issues but we will also briefly analyze some ways schooling can represent the knowledgecreation approach. We present a project with one teacher and her grade 5 (11-yearold) students, entitled ìArtifacts where the aim was to learn to understand the role of artifacts in the past, present and in the future (Seitamaa-Hakkarainen et al. 2004). A central characteristic of this project is the involvement of students in sustained work for generating knowledge of artifacts where both conceptual and material artifacts were emphasized; as we maintain, in a trialogical way.

D26 An intentional analysis of students' descriptions and explanations of the seventeenth century

Liza Haglund, department of education, Sweden

The present paper deals with part of a study that investigated students' (age 11-12) reasoning about history. The issue at hand in this paper is students' explanations and description of the seventeenth century in Sweden, as presented to them by the Vasa

Museum. A total of fifty-five students worked individually with three tasks on three

– 300 –

different occasions. Their tentative result challenges the idea that students' historical reasoning is best described in terms of levels. Some students used different types of explanations to explain the same phenomena. There was also a variety in how the students encountered the tasks and actualised their correspondingly different conceptual contexts. Within the same responses, there were shifts between conceptual contexts and explanations and students gave descriptions both generally and with reference to particular objects and circumstances, as if they were answering different questions. This raises the need for a new description or metaphor that accounts for complexity in historical reasoning, as well as further research that departs from a theory that takes this into account. Pointing to the difference between a teleological explanation and a causal one, may be enlightening for many students. But it seems that such a discussion must parallel one about different conceptual contexts, and how explanations can be valid in one context but miss the point in another.

D27 Effect of reading aloud and simple arithmetic as a vehicle for activating prefrontal area in children on academic attainment

Hajime Yoshida, Ritsumeikan University, Japan

Mariko Ishikawa, Ritsumeikan University, Japan

Recent studies indicated that performing very simple arithmetic such as 3 + 6 nd reading aloud activated prefrontal area by using fMRI or PET. Prefrontal area is regarded as control center of thinking, memory, control, or other higher cognitive functions. Activation of this area would lead to improvement of these higher cognitive functions. Thus, it would be assumed that activating prefrontal cortex by performing simple arithmetic and reading aloud lead to improvement of cognitive functions and then to one of academic attainment in pupils. The goal of this study was to confirm this hypothesis by interventional investigation. 103 fifth graders in the experimental (E) group were given one of three kinds of tasks in a day; simple arithmetic, reading aloud, and writing Japanese characters. This intervention lasted for three months. Before and after intervention, assessment test was administrated to check pupil's attainment on arithmetic and Japanese language. Pupils in the C group were given the pre- and post-tests. Total change score from the pre- to the post-test in arithmetic for the E group was significantly greater than one in the C group. In language, total change score from the pre- to the post-test in language for the E group was not different from one in the C group.

D28 The relationship between motor skills and reading performance among 7 years old children

– 301 –

Giorgos Daloukas, University of Thessaly/Special Education, Greece

According to recent studies 10-15 percent of the school population faces oral and/or written language difficulties, which are more severe in a 4-5 percent. On the other hand, 5-15 percent of school children seem to face movement control problems.

Previous recearch found that children with specific reading disabilities appears also serious deficits in a large spectrum of movement skills such as balance and generally sharp kinetic ability. More recently has been suggested that motor impairments are restricted to a subset of the dyslexic population. The aim of this study was to investigate the possible relationships between motor skills and reading performance of typical seven years old children, in order to compare the two skills at the within-individual level. The sample consisted of 300 children. Children's reading performance evaluated with the Test of Reading Performance, which is constructed of six tasks: letter knowledge, phoneme blending, word identification, morphology knowledge, syntactic knowledge and passage comprehension. Movement evaluation was carried out using the motor test of Movement Assessment Battery for children which covers three major motor domains: manual dexterity, ball skills, static and dynamic balance. The results showed that there are significant correlations between reading and motor scores. Such results give additional support to recent views according the role of cerrebellum in both motor and cognitive functions, including reading. Our findings create a fruitful ground for the further study of movenent assessment as a means to predict later reading performance.

D29 Working at home for school: The significance of individual competences and scholastic stress factors

Petra Wagner, University of Vienna, Austria

Barbara Schober, University of Vienna, Austria

Christiane Spiel, University of Vienna, Austria

The issue "homework" is discussed from various perspectives throughout the twentieth century. Previous studies have shown an obvious variability in time students spend working at home for school, and very controversial results concerning the relationship between time investment and achievement were found. To clarify this unsatisfying research situation, a simultaneous look on different relevant determinants is necessary. So we analyzed the amount of time students invest at home for school in relation to other school related determinants. The basis of including these determinants was the "model of time spent on learning" (Helmke & Schrader,

1996). Among other issues this meta-model shows the relation of psychological

– 302 –

variables, learning time and school achievement. Regarding psychological variables the present study focused on academic self-concept, text-anxiety, depression, and pressure to perform. In sum, 595 Austrian students (312 girls, 283 boys) from various general secondary schools (74 girls, 43 boys), academic secondary schools

(235 girls, 107 boys) and vocational schools (78 girls, 98 boys) were investigated.

A diary was used to measure the amount of time invested in work for school, the individual competences and scholastic stress factors were measured with a questionnaire. To identify learning types we used cluster analyses and found three types of students according to their time investment: the minimalist, the homework-type, and the test-type. They differed in their individual competences and scholastic stress. Implications of these findings for explaining the often found very low correlation between time investment and achievement will be discussed.

D30 Learning styles: A longitudinal study in higher education

Vincent Donche, University of Antwerp, Belgium

Peter Van Petegem, University of Antwerp, Belgium

Studies concerning student learning in higher education indicate that learning styles can be distinguished using the diagnostic inventory of learning styles (ILS) of Vermunt (1998). This inventory is based upon a model of learning styles in which processing strategies, regulation strategies, mental models of learning and learning orientations are integrated. Few studies investigate the stability and change of these learning style components within a longitudinal research design. Some of these studies indicate that over time students become more meaning oriented between the first and third year of a study at the university. The former findings gave raise to a new study aimed to (1) test the generalisability of Vermunt's model of learning styles within a sample of 146 students from a Flemish institute of higher education in Belgium and (2) to test the stability and variability over time of learning styles.

The ILS-inventory is found a reliable instrument to measure most of the distinguished learning style components of Vermunt. Learning style components significantly change over time. First year students who reached the third year of their study reported to carry out more concrete processing strategies and their learning is less externally regulated than in the first year. Learning conceptions of third year students shift also over time as can be indicated by a decrease of the conception of learning as an intake of knowledge and an increase of the conception of learning in which the use of knowledge is important. However these results will be analysed more in depth with several analysis techniques. The paper describes the empirical findings concerning these two main questions and discusses implications of this

– 303 –

research for both theory and practice.

D31 Individual learning patterns in virtual reality environments

Elhanan Gazit, Tel-Aviv University, Israel

Virtual reality learning environments (VE's) provide direct learning experience

(Dede et al., 2000). The Virtual Solar System (VSS) is a non-immersive 3D virtual environment based on high-resolution spacecraft images of the solar system's planetary objects which revolve in their orbits around the Sun. The learner uses the computer mouse to change his/her viewpoint while "flying" in 3D space (Yair et al., 2001). This offers a new visual learning experience, which has not been systematically studied yet. The study main objectives were to describe and analyze learners' learning interactions during a free exploration task of the VSS and to study their conceptual development of the basic astronomical phenomena during pre-plan tasks. A systematic examination of ten students (6 boys, 4 girls, age 15-16) realtime learning interactions reviled that during a free exploration task, all students established ìanchoring-dock points. Three different exploration patterns emerged: a superficial learning pattern, an object-based learning pattern in which the learner study in-depth the various celestial objects, and a flexible learning pattern in which the learner moves between a global observation pattern and a local observation pattern. Moreover, results showed that the conceptual development is a non-linear process which includes transitions between and within at least five dimensions: (1) a cognitive-local dimension, (2) a cognitive-global dimension, (3) a navigation dimension, (4) an interface dimension, and (5) an affective dimension. To conclude, the virtual reality environment affords the emergence of individual differences which are manifested in the multidimensional and emergent nature of the learning interactions patterns. Thus, this kind of learning calls for suitable scaffolding and guided reflection design. This multi-dimensional conceptual framework of the learning interactions might serve other related research areas which study learning interactions in various complex virtual environments.

D32 Classic logic taught to 5th grade children: effects of a training

Lucia Bigozzi, University of Florence, Italy

Beatrice Accorti Gamannossi, University of Florence, Italy

The aim of the present study is to evaluate the efficacy of a formal logic teaching process for children attending 3rd to 5th grade in primary school. Hypotheses:

We assume an effectiveness of the training on formal logic competence, on logical

– 304 –

mathematic abilities, on mathematics, on logic comprehension in reading, on narrative structure. We also assume that the training effect is independent from the person who administers it. Participants: 196 children (104 males and 92 females, mean age: 9 years and 8 months) participated to this study. Children were divided in three groups: two experimental groups and one control group. Procedure: This is a longitudinal study. All the groups were given objective pre-tests and post-tests.

The experimental groups followed a formal logic training: in one group (1) the trainer was an external experimenter, in another group (2) the trainer was the class teacher. The training (4 months, 2 hours a week) was about explanation, discussion, execution and collective correction of 90 teaching cards concerning the main areas of formal logic. The control group followed the standard teaching program. Data analysis: To verify the training efficacy a comparison among the initial and final means of the two experimental groups and the control group has been conducted through a repeated measures ANOVA. Results and conclusions: The teaching training is effective to enhance competence in formal logic, in logical mathematic, in narrative structure, in arithmetic, geometric, statistic and problem solving abilities.

In each variable the training is effective independently from the trainer. As far as the task of comprehension of aesthetic-poetic language is concerned there are not significant differences among groups. Results will be discussed in relationship with their educational and teaching worth.

D33 Students Intuitive Beliefs About Practice

Virginia Hill, Fordham University, United States

Mitchell Rabinowitz, Fordham University, United States

Despite the common belief that expert performance is due to innate abilities, recent research in varied domains of expertise has shown that expertise is developed by daily practice over an extended period of time. Ericsson, Krampe, and Tesch-

Romer (1993) posits that the salient component in developing expertise is deliberate practice, defined as highly structured, effortful activity, the explicit goal of which is to improve performance in any domain. In this study, a 22 item Practice Beliefs instrument was developed to assess students intuitive beliefs about the structure of practice. The 22 items were anchored in Ericsson's definition of deliberate practice

(11 items that were consistent with the literature; 11 items that were polar opposites) with a four-point, Likert-scale: 4-Strongly agree, 3-Agree, 2-Disagree, and

1-Strongly disagree. The survey was completed by 34 graduate students who were enrolled in a beginning course in cognitive psychology. The results suggested that the sample population had an understanding of the efficacy of practice methodol-

– 305 –

ogy but were not clear on what constitutes deliberate practice. They also retained, to a great extent, popular, erroneous beliefs that people are born with talent, some people dont have to practice to achieve mastery, and good learners will be able to determine their own practice needs. These findings are relevant to curriculum development and the design of teacher or instructor training programs.

Mathematics Education

Discussant: Athanassios Gagatsis, University of Cyprus

D34 The Effect of Self-Explanation on Solving Mathematical Word Problems

Hidetsugu Tajika, Aichi University of Education, Japan

Narao Nakatsu, Aichi University of Education, Japan

Hironari Nozaki, Aichi University of Education, Japan

The purpose of the study was to examine how a metacognitive strategy known as self-explanation influences word problem solving in elementary school children.

Participants were 73 sixth-graders. They were assigned to one of three groups, the self-explanation group, the self-learning group, or the control group. Students in each group took two kinds of test. The results showed that the self-explanation group and the self-learning group outscored the control group on a transfer test. In addition, high explainers who generated more self-explanations relating to deep understanding of ratio word problems outscored low explainers on ratio word problems and transfer tests. The self-explanation effect is discussed.

D35 The relation between computational skill and arithmetic word problem comprehension

Jose Orrantia, University of Salamanca, Spain

Santiago Vicente, University of Salamanca, Spain

In this work, and from an information-processing perspective, we examine the hypothesis that problem comprehension and computational processes interact during the solving of an arithmetic word problem. It is clear that both computational processes and problem comprehension can be expected to have an effect on performance in problem solving. However, little is known about the manner in which these factors contribute to differences in performance. We tested subjects with high and low skill in computation on a series of word problems, the content of which varied

– 306 –

on the basis of problem type (simple-complex) and on the numbers type included in the problems (positive integers-fractions). Performance data are presented and analyzed in terms of solution reaction times. The results revealed an effect of problem complexity. Overall solution time was significantly shorter for simple problems than for complex ones. There was also a main effect of number. Overall solution time was significantly shorter for problems that included positive integers than for problems that included fractions. The more important findings, however, concern the relation among complexity, number and skill. There was significant two-way interaction between skill and complexity when the problems included positive integers, but this effect there was no significant when the problems included fractions. When the problems included positive integers, differences between simple and complex problems were shorter for high skill subjects than their low skill peers.

Our results not support the hypothesis that the processes involved inthe comprehension of arithmetic word problem text and the proces of computatation are independent of each other. The results suggest an alternative hypothesis: that the cognitive processes involved in solving the problems can occur in cascade, this is, one process begins before the other is completed. However, there is no cause to expect that automatized retrieval of computations will improve problem solving.

D36 Achievement and interest development in Swiss 8th grade mathematics classrooms: the role of different teaching features in traditional and reform-oriented classrooms

Monica Waldis, University of Zurich, Switzerland

Kurt Reusser, University of Zurich, Switzerland

In the context of the Swiss participation in the cross-cultural TIMSS 1999 Video

Study (Hiebert et al. 2003) a representative sample of 140 videos of 8th grade mathematics lessons have been collected and analyzed. As a national extension, video data were complemented with a national teacher and student questionnaire and the

TIMSS achievement test. The wide data corpus allows us to combine the perspective of students' and experts' (high inference video ratings) on features of instructional quality. The present paper aims to highlight the relationship between different teaching features - either associated with direct teaching or indirect teaching

- and multiple educational outcomes as achievement and interest development. In a first, cross-sectional study, we explored the effects of distinctive teaching features known from literature to be positively related with either achievement or mathematical interest. Hierarchical linear regression analyses (Raudenbush & Bryk,

2002) of students' perceptions and video ratings showed rather small context effects

– 307 –

on achievement after the controlling of school type and students' socio-economical background. In contrast, students' perceptions of instructional quality explained a small amount of individual interest variance. Cognitive activation and instructional clarity seem to be of a certain importance. The second study shifts from a variablecentred to a group-centred research approach. Teachers' information about their use of instructional methods allowed a clear distinction between classes with rather traditional, teacher-led instruction (n=41) and classes with a more reform-oriented, student-centred teaching approach (n=43). Significant differences with a slight advantage for the reform-oriented teaching environment were found for the teaching features already examined in study 1. Students' perspective and experts' perspective converge in this point. However, longitudinal analyses of educational outcomes revealed a certain ceiling effect concerning interest development in reform-oriented classrooms between class 8 and class 9.

37 How do teachers' pedagogical content beliefs about teaching Pythagorean

Theorem interact with pupils' achievements and views on the teaching style?

Miriam Leuchter, University of Zurich, Switzerland

Kurt Reusser, University of Zurich, Switzerland

Christine Pauli, University of Zurich, Switzerland

Teachers' instructional action is closely related to their beliefs about teaching and learning. This analysis (1) uses concrete instructional situations to record teachers' pedagogical content beliefs in the context of a standardized videotaped three-lesson unit on the Introduction to Pythagorean theorem, (2) establishes links between teachers' beliefs, pupils' achievements and perceptions of the teaching style, and (3) relates teachers' beliefs as well as pupils' achievements and perceptions to instructional actions observable in the video data. The perceptions of the teaching style of a teacher on the one hand from the point of view of the pupils as participants in the instruction and on the other hand from the perspective of researchers as video observers demonstrate the extent to which the teachers' beliefs are reflected in the perceived instructional practices. The influence of teachers' pedagogical content beliefs on pupils' achievements indicate the importance of the integration of beliefs for successful measures of teacher training and further development. Within a Swiss-German video study on teaching and mathematical understanding (N=40 classes, Year 8/9 (Klieme & Reusser, 2003), a semi-structured interview concerning a concrete, thematically standardized, videotaped instructional unit recorded teachers' pedagogical content beliefs. Pupil perceptions of the teaching style (N=980) were measured with questionnaires, their achievement with two tests. Statistical

– 308 –

analysis enabled the linking of the pupil, teacher and video data and an examination of the congruence of the three perspectives.

D38 The role of students' age, problem type and situational context in solving mathematical word problems

Vesna Vlahovic-Stetic, University of Zagreb, Faculty of Phylosophy, Croatia

Zeljka Mendek, 6th Elementary school Vukovar, Croatia

Daria Rovan, University of Zagreb, Faculty of Phylosophy, Croatia

Daria Rovan, University of Zagreb, Faculty of Phylosophy, Croatia

The aim of this research is to verify hypotheses regarding the importance of understanding situation in mathematical word problems that follow from Reusser's

SPS (Situation Problem Solver) model by exploring efficiency of Children's mathematical word problems solving as a function of age, problem type and situational context of the problem. Children of three age groups participated in the study: preschool kindergarten group, first grade and second grade students. The final sample was formed of 67 kindergarten children, 79 first grade students and 85 second grade students. Testing was conducted by 20 specially trained senior psychology students.

Two categories of word problems were used: change problems and compare problems. Every child was tested twice, one time with neutral context problems and other time with familiar context problem. Repeated measures analysis of variance with age, situational context and problem type as independent variables was performed with Children's performance as a dependent variable. The main effects of age, situational context and problem type were significant, as well as the interaction of age and problem type. Our research results indicate that older children are more successful than younger children in solving mathematical word problems and that the Children's performance on the change problems is better than on compare problems. Results also showed that Children's performance on the problems with familiar context was better than performance on the problems with neutral context.

These results confirms Reusser's essential hypothesis that adding additional sense to the problem text would facilitate designing the situational model of the problem.

Implications for teaching mathematical word problem solving are discussed.

Science Education

Discussant: Kostas Korfiatis, University of Cyprus

D39 Attitudes and conceptions of teachers and students toward two lab methods: virtual versus conventional

– 309 –

Naftaly Dov, Hebrew University of Jerusalem, Israel

Moti Frank, Technion - Israel Institute of Technology, Israel

The purpose of this study was to examine attitudes and conceptions of teachers and students toward learning and instruction in two lab environments - a computerized simulation (by means of EWB - Electronic Work Bench software) and under conventional methods (learning and instruction with students performing hands-on experiments). The participants were 12th grade students (N=248) and their teachers (N=67) in 22 groups in electricity and electronics laboratories in Israeli high schools. Qualitative and quantitative methods of research were combined in this study. Questionnaires were given to students and teachers. An anonymous attitude questionnaire was given to students to determine their views and grasp of laboratory methods. Data from teachers was gathered through an anonymous survey. The teachers' responses indicate that most of them conducted lab lessons through an idiosyncratic combination of both methods. They used the EWB simulation program for a number of purposes, the main one of which was having students do virtual experiments as preparation before performing them in the traditional method. The findings gathered from students' and teachers' questionnaires focus on the main advantages and disadvantages of the EWB computerized simulation versus the conventional method in the electricity and electronics laboratories. In our presentation we will detail the advantages and disadvantages of each method according to our findings.

D40 Students' understanding of primary science concepts, and their perceptions of classroom interactions

Bruce Waldrip, University of Southern Queensland, Australia

Jeff Dorman, Australian Catholic University, Australia

Darrell Fisher, Curtin University, Australia

John Green, University of Southern Queensland, Australia

The overall aim of this study is to investigate relationships among students' perceptions of cultural factors that affect their primary classroom learning environment, teacher-student interactions, students' understanding of science concepts and attitude towards science in upper primary school science classes. The instruments to

100 primary school classrooms across three Australian states. This list of concepts were those concepts that are common to teaching science in each of these states for years 5-7. The internal consistency/reliability ranged from 0.62 to 0.83 suggesting that each QTI scale has acceptable reliability, especially for scales contain-

– 310 –

ing a relatively small number of items and for the Perceptions of Cultural Factors affecting Learning Environment they ranged between 0.74 and 0.85. Female students were more likely to have more positive attitudes towards science, rated equity higher as well as teamwork being more desirable than individualisation. As well they perceived higher levels of leadership, helpingfulness, understanding, and student responsibility, and less dissatisfied, admonishing and strict behaviours. As the year level increased, student attitudes decreased. Where students displayed greater understanding of science concepts, they were more likely to be male, and to have teachers as displaying more leadership. Students who had better understanding of the science concepts tended to perceive greater equity, teamwork, congruence of learning between home and school, perceived more leadership, helping, student responsibility and less uncertain interpersonal behaviours. All of the scales of the QTI were found to be associated with students' attitudes. This study examined actual science content knowledge. Those students who had lower science understanding were more likely to display less favourable interactions. The study adds to present learning environment research by examining whether students' perceptions of culturally sensitive factors of their learning environment and teacher- student interactions is influenced by their understanding of science concepts.

D41 Crystallized or dynamic conception of teaching Science. Some reflections after the implementation of an EU co-operation project in rural areas of El Salvador

Alberto Nagle Cajes, Universidad ORT-Uruguay, Uruguay

The aim of the work is to explore rural teacher's change of conception of Science education in primary schools in eastern countryside areas of El Salvador. Traditional teaching practice in basic education has often stressed out the development of technical skills which keep learners ìon task creatingìteacher centered classroom for the transmission of the teacher's knowledge. The new paradigm on learning sees the role of the teacher in another way. The new teacher structures the course content and the classroom environments in order to facilitate the engagement of student's activities which promote learning and deep understanding. A comprehensive intervention in a sample of schools was carried out, in order to bring some methodological changes in teachers practice. The following questions guided the present investigation: What are the conceptions of teaching science held by participants? How

– 311 –

do these conceptions change as participant's progress through the implementation of an innovation in their schools? How do these teachers see the relationship between learning and understanding? How do the teachers describe their usual way of teaching? Primary sources of data were verbatim transcripts of 15 semi-structured formal interviews with participants. The analyses reveal that the teachers have an almost crystallized conception of teaching science. The analysis shows two phenomenographic categories of description. In one category the teachers stressed out their purpose to foster student appreciation of scientific knowledge and scientific phenomena in their daily lives. Another group of teachers consider teaching as a mission whose ends lie beyond scientific literacy. The ultimate goal of teaching is to help student's become responsible citizens and human beings.

D42 A case study of utilization and evaluation of the effectiveness of the ìi tool in an inquiry based biology laboratory course

Nilgun Tatar, Gazi University, Turkey

Fatma Sasmaz Oren, Gazi University, Turkey

Dilek Erduran, Gazi University, Turkey

Nowadays, one of the most significant problems is the quality of learning in educational settings. Students should learn how to do inquiry, how to find and interpret data and be able to apply their learning in different situations. In this point, inquirybased teaching, which is a student centered method, is assessed as the teaching method A tool ìI was developed by Phillips and German (2002) to help teacher and student in the classroom where used inquiry-based teaching method. In this study the inquiry ìI tool was used. In this study; we apply inquiry-based learning method by using a tool ìI and evaluate efficacy of this tool in biology laboratory course on osmosis diffusion subject. This study is conducted a case study research.

Study was executed in Gazi University, Department of Science Education Teaching. Students selected randomly in the second grade classroom. In the first group used ìI tools based on inquiry-based learning. In the second group used traditional teaching methods. Each group consist of 25 students. To determine effectiveness in academic achievement, after the study 5 open-ended questions were asked on both experimental group and control group. Furthermore, to evaluate effectiveness of this tool we performed interview with 6 students. In conclusion of the study, according to the students who applied inquiry-based teaching method, this tool rather effective. We observed that this tool facilitates following of experimental process, increases students' participation of the course and academic achievement. According to interviewer results, students have positive attitude against the tool. Since

– 312 –

ìI tool in front of the students, they have a chance study as a scientists in all of the phase scientific process in laboratory course. This provide probability them to develop their inquiry and investigation abilities. Simultaneously, the tool facilitate guiding and evaluating for teacher.

D43 Mistakes and Negative Knowledge - Idle Capacities in Teaching Science

Concepts? Findings of a Video Study in Physics Instruction

Lena Meyer, Leibniz-Institute for Science Education (IPN), Germany

Tina Seidel, Leibniz-Institute for Science Education (IPN), Germany

Manfred Prenzel, Leibniz-Institute for Science Education (IPN), Germany

For students, mistakes in the classroom often go along with individual disappointment and low graded achievement. From the teachers' point of view, student mistakes are often connected with disruptions in the planned sequence of teaching and are negatively associated. However, mistakes can be used to reveal prior knowledge, to outline concepts and to develop negative knowledge. Within the scope of a mistake culture in the classroom, negative knowledge terms declarative and procedural knowledge about matters and processes respectively, in order to determine outside elements or procedures. By exposing those knowledge's relation to science concepts, it fulfils the task to encompass the concept by developing negative knowledge about what are not corresponding components and how procedures do not work. It is assumed that negative knowledge utilized for constructive concept formation, and a mistake culture on closer examination of the blending of learning and test situations in the classroom play a dominant role in the quality of concept formation and perceived learning conditions. The sample includes 22 German physics classes and combines students' questionnaire data with high and low inferent video analysis. The findings show that the utilization of negative knowledge and a positive mistake culture is associated with a more positive perception of supportive learning conditions and learning motivation, as well as with the long-term development of students' scientific literacy.

D44 Determining the relationship between basic science process skills and attitude toward science in primary school

Dilek Erduran, Gazi University, Turkey

Nilgun Tatar, Gazi University, Turkey

Fatma Sasmaz Oren, Gazi University, Turkey

The purpose of this study is to investigate whether there is a relationship between

– 313 –

basic process skills and attitudes toward science in second grade students of primary school.

The research was conducted in 2004-2005 fall semester a sample of 250 students.

The sample consist of 150 students at Erdemir primary school in Karabuk, 50 students at Akpinar Primary school and 50 students of Nenehatun Primary School both settled in Ankara. In this research, an attitude instrument (Attitude toward Science in

School Assessment) was implemented to measure students' attitudes toward science by designed Germann (1988). In order to evaluate basic scientific process skills ì

Science Process Skills Assessment (SPSA) instrument was used. The science process test adapted by the Mason City Schools, August,1993, in the Iowa Assessment

Handbook (Enger and Yager, 1998) was used. For using these instruments given permission by Germann (to Attitude toward Science in School Assessment) and

Yager (to Science Process Skills Assessment). These instruments (Science Process

Skills Assessment, Attitude toward Science in School Assessment) was used to assess the relationship between attitude toward science and basic scientific process skills. All student data collected have been confidential. Student data were analyzed using quantitative methods. Quantitative data was gathered in order to answer the research questions (SPSA) and Attitude toward Science in School Assessment

(ATSSA) instruments posed in this inquiry. It was examined correlation relationship by using SPSS package program on analyzing data. Mid-level correlation was detected between attitudes toward science and basic scientific process skills. According to the results, it was found that conclusion skills of many students were low in contrast to classifying skills. Students' who have negative attitude toward science, science process skills lower than others. Results tabled, interpreted and appropriate suggestions have been presented as given percentages.

D45 Development of Australian High School Students' Understandings of, and

Attitudes About Biotechnology Processes

Vaille Dawson, Edith Cowan University, Australia

One of the essential outcomes of science education is to enable students to develop a deeper understanding of the world around them, and to be able to engage in relevant discourse about science in everyday life. There has been significant emphasis placed on the importance of scientific literacy in science education. A high level of scientific literacy can help young people to question the claims of the scientific community, weigh up evidence about science issues, use critical thinking skills and enable them to use their understanding of science to make well informed decisions. One area of science that will increasingly impact on our society is the

– 314 –

field of biotechnology. The aim of this study was to determine how understandings and attitudes about biotechnology processes develop as students progress through high school. In a cross sectional study, data was obtained from written surveys of students in Years 8, 10 and 12. The results indicate that students' ability to provide a correct definition and examples of biotechnology, cloning and genetically modified foods was very poor amongst Year 8 students but improved in Years 10 and 12. Students in all year groups had a better understanding of cloning than biotechnology and GM foods. Year 12 students studying biological science had the best understanding. Most students approved of the use of biotechnology processes involving micro-organisms, plants and humans and disapproved of the use of animals. However, Year 8 students' attitudes were less favourable than Year 10 and 12 students.

An awareness of the development of students' understandings and attitudes may lead to a more appropriate use of biotechnology curriculum materials and thus improved biotechnology education in schools.

D46 Classroom talk and science education

Zilda Fidalgo, ISPA - Instituto Superior de Psicologia Aplicada, Portugal

Filipa Ferreira, ISPA - Instituto Superior de Psicologia Aplicada, Portugal

Joana Sequeira, ISPA - Instituto Superior de Psicologia Aplicada, Portugal

Isabel Matta, ISPA - Instituto Superior de Psicologia Aplicada, Portugal

The aims of this study are: a) To observe how teachers of young children developed classroom talk on a scientific theme, and b) To adapt to Portuguese the method of classroom talk analysis proposed by Wegerif and Mercer (1997), Wegerif, Mercer and Rojas-Drummond (1999).

Four preschool and primary school teachers and 63 children (4-7years) participated. Teachers were invited to give a 3 session lesson about the properties of the air, which were videotaped and transcribed verbatim.

The following key-words coded exploratory/non exploratory, were selected from the transcripts: let's see what is going to happen/what happened; what happens Ö; because; why; as (comparative/causal); if; so. Cohen's K values ranged from 1 to

.62. The discourses was segmented in Spiral and Loop IRF exchanges and discursive actions identified (13), inspired from Dimension/Action matrix of Edwards and Mercer (1987) and Wegerif, Mercer and Rojas Drummond (1999). Loop IRF exchanges were negatively correlated with the key word exploratory use let's see, and action a) building knowledge from one to another; b) asking questions that explore pupils' levels of understanding, and positively correlated with a) making explicit the ground rules or demands of the task. Concerning the exploratory use of

– 315 –

the key-words, differences between teachers and between children were registered.

Concerning Loop versus Spiral IRF exchanges, discourse with the seven years old children showed significantly more Loops, associated to more direct questions, explicitly links of prior knowledge to current activity, and recapped learning with pupils. These results are encouraging both concerning the validity of coding systems used and information obtained. However, Loop versus Spiral IRF exchanges was essential to characterize classroom discourse with 7 years old children as the more away from socio-cultural model. Further developments of coding system are needed to enable to characterize teaching and learning practices.

D47 Giving priority to evidence in science teaching:A first year elementary teacher's specialized practices and knowledge

Lucy Avraamidou, University of Cyprus, Cyprus

Carla Zembal-Saul, The Pennsylvania State University, United States

The purpose of this qualitative case study was to examine the nature of a first-year elementary teacher's specialized practices, knowledge and beliefs for giving priority to evidence in science teaching and explore the possible sources from which this knowledge was generated. Data included three audio taped interviews, six video taped classroom observations, lesson plans and samples of students' work. The findings of this study revealed that: a) Jean gave priority to evidence in her teaching practices by constantly engaging the students in collecting evidence through observations and tests, recording and representing evidence and using that evidence to construct explanations; and b) critical experiences during her preparation to teach and specific university coursework acted as sources through which this aspect of

PCK was generated. This study adds to the value of the concept of PCK within the domain of research on science teaching (Van Driel, Verloop, & de Vos, 1998) by illustrating how to study and what to look for when studying this specific aspect of PCK. At the same time, this study underscores the need for further research and larger scale studies in the area of teachers' specialized practices, knowledge and beliefs for giving priority to evidence in science teaching that will contribute to a better understanding of its nature, sources and development.

Motivational Social and Affective Processes

Discussant: Monique Boekaerts

D48 Is math something scary?Attitudes and beliefs toward math and math anxiety in secondary school students

– 316 –

Vesna Vlahovic-Stetic, Faculty of Philosophy, Dept. of Psychology, Zagrel,

Croatia

Ana Severinac, Faculty of Philosophy, Dept. of Psychology, Zagre, Croatia

Lidija Arambasic, Faculty of Philosophy, Dept. of Psychology, Zagreb, Croatia

In today's society a smaller proportion of women, in comparison to men, are engaged in the professions connected with mathematics and similar subjects, and they are less frequently enrolled in the faculties where math is an obligatory subject.

According to some authors, the mathematics is the reason why there is a relatively small number of women in the well paid, prestigious careers (Hyde, Fennema, Ryan,

Frost and Hopp, 1990.). Because of that it seemed interesting to investigate what general attitudes do secondary school students have toward mathematics, do they see it as a male domain, do they think that math abilities are inborn and do they experience math anxiety. Participants in the study were students from all four classes of the two language oriented and one science oriented secondary schools (N=531).

Two scales were used in the study: Scale for measuring attitudes and beliefs toward math and Scale for measuring math anxiety. Results show that science oriented students have more positive attitudes toward math and they believe more that math abilities are not inborn. Science oriented students and girls do less believe that math is a male domain. Language oriented students and girls had more intensive math anxiety. There was no significant interaction effect between educational orientation and gender on any variable. Obtained results were commented in regard to students' educational orientation and different gender role socialization process.

D49 Are parent's beliefs, practices and personal characteristics linked to negative self-appraisal of competence in children?

Valerie Dubois, Universite du Quebec a Montreal, Canada

Therese Bouffard, Universite du Quebec a Montreal, Canada

Mathieu Roy, Universite du Quebec a Montreal, Canada

Marie-Eve Vaillancourt, Universite du Quebec a Montreal, Canada

Children's appraisal of competence is a motivational resource that is strongly linked to their functioning and achievement (Bandura, 1986). Young Children's appraisal of competence is thought to be usually positive, but Phillips (1984, 1987) showed that some children tend to underrate their competence. The present study's aim was to investigate the factors that may be involved in causing this self-defeating phenomenon. Since, in the constructivist perspective, children build their perceptions of competence through interactions with significant social agents, parental variables

– 317 –

like the types of educational practices used, sense of efficacy, appraisal of their own competence when they were young and their appraisal of their child's competence were examined. 560 parent and child from fourth and fifth grade agreed to participate in this study. Results showed that parents are somewhat capable of perceiving how accurate their child is at evaluating their competence. This was done by confronting the parent's evaluation of their child's self-judgement of competence with a standardized score of regression of perceived competence by the child on

IQ. As for the effect of parental variables on the distribution of children in groups that overestimate, underestimate or estimate accurately their competence, we found that, while parental practices were not linked with the child's accuracy of perceived competence, some parental beliefs and personal characteristics of the parents were linked. Parent's belief in an incremental conception of intelligence was higher in the group of children that underestimates their competence, while parent's self-efficacy was higher in the group of children that overestimates their competence. It was also noted that parents of children in the underestimating group had a lower perception of their competence when they were children. The discussion will focus on the mechanism through which parents' beliefs may impact on their child's sense of competence.

D50 Mothers' Expectations and Family Climate Before and After Cochlear Implantation

Amatzia Weisel, Tel Aviv University, Israel

Tova Most, Tel Aviv University, Israel

Rinat Michael, Tel Aviv University, Israel

The purpose of the present study was to examine functioning and stress among families with hearing-impaired children as a function of both the time passed since the cochlear implantation and parents' expectations from the implant. The participants were 64 mothers of hearing impaired children in different time ranges of implantation: candidates, up to 3 years after implantation, and more than 3 years after implantation. Three questionnaires were used: 1) ìIn My Family, for assessing family functioning (Olson, Russell & Sprinkle, 1979; 1980; 1983). 2) The Parental Stress

Index that examines stress among mothers of children with special needs, and, 3)

Expectations from the Cochlear Implant Questionnaire which measures parental expectations regarding communication, self image, social adjustment, academic achievements, family, and rehabilitation demands. The main results of the study were (a) As the time from the implantation increased, the levels of expectations related to communication and academic achievements decreased. (b) Families with higher level of cohesion reported less stress than families with less coherent cli-

– 318 –

mate. (c) Positive correlations were found between parents' expectations regarding communication, social adjustment, and family climate and the current performance of the child in these domains. (d) Contrary to our hypothesis parents' expectations in the areas of self-image and academic achievements were not related to the child's current performance in these domains. (e) The level of stress of the families did not change as a function of the time from the implantation. (f) Mothers with higher level of education expressed less stress than mothers with lower level of education.

The implications of these results to the rehabilitation process will be discussed with special emphasis on the relatively high expectations before the implantation and on the need to include the family perspective in the rehabilitation process.

D51 peer interaction and social identity dynamics: Workings of the epistemic triangle

Charis Psaltis, University of Cambridge, United Kingdom

While the productive role of social interaction between peers in promoting cognitive development has been clearly established, the communicative processes through which this is achieved have not been clearly identified. This paper reports two studies in which 226 Greek-Cypriot pupils 6.5 to 7.5 year (first graders) and

256 pupils of the same age were presented correspondingly with the Piagetian task of conservation of liquid and a spatial perspective taking task. Both male and female non-conservers worked with a conserving partner in either same or mixedsex dyads, thus creating four different pair types. The pairs were asked to discuss their conflicting answers and agree upon a joint response. Sociometric measures of popularity and academic reputation were collected through peer nominations and teacher's evaluations. Cognitive progress was assessed by pre to post-test gains in pre-test-interaction-posttest (immediate and delayed)design. The results indicated that the type of conversation established during the interaction was strongly related to the outcome and that the gender composition of the pairs and the academic reputation and popularity of the pupils influenced the type of conversation which occurred. These results are discussed in relation to the general model of socio-cognitive conflict and extend Chapman's (1991) theoretical framework of the epistemic subject-other-object triangle by unravelling the role of social gender identity dynamics and other sources of asymmetries (academic reputation and popularity)in mediating the way in which the conflict is jointly expressed and resolved and individual cognitive progress takes place.

D52 Comparing two methods for investigating peer relationships: conceptual

– 319 –

and methodological issues

Katja Kosir, Department of Psychology, University of Ljubljana, Slovenia

Sonja Pecjak, Department of Psychology, University of Ljubljana, Slovenia

In sociometric research tradition, popularity is defined as the degree to which children are liked or accepted by their peers. However, it has been suggested recently that popular students may represent a less homogenous group that has been commonly assumed. Research indicates that it should be distinguished between two definitions of popular students: (1) popular students as those students who are well liked by many and disliked by few peers and (2) popular students as those students who are described as popular by their peers. The main purpose of the present study was to examine the relationship between sociometric and peer perceived popularity in students of different grades of elementary and secondary school. Additionally, the age differences in the relationship between sociometric and peer perceived popularity were examined. Another purpose of the study was to investigate the differential relationships between concepts of popularity and some students' characteristics. The participants were 321 boys and 329 girls who ranged from fifth grade of elementary school (the mean age approximately 11 years) to third grade of secondary school

(the mean age approximately 11 years). Data were collected using sociometric test, peer interpersonal assessments, and self-concept questionnaire. The results of this study confirm previous findings that peer perceived popularity is a construct that is distinct from sociometric popularity. There are some substantial differences in relations between indices of perceived popularity and sociometric indices between students of primary and secondary students, i.e. between early adolescents and middle to late adolescents. It seems that perceived popularity and sociometric popularity are rather similar constructs in elementary school students, whereas in secondary school students they become almost unrelated to each other.

D53 Comparison between co-peer tutoring and near-peer tutoring

Jinlian Hu, The Hong Kong Polytechnic University, Hong Kong

Yeung Lapyan, The Hong Kong Polytechnic University, Hong Kong

Patrick Lai, The Hong Kong Polytechnic University, Hong Kong

Effectiveness of Co-peer Tutoring (CPT) and Near-peer Tutoring (NPT) in university study is investigated. CPT scheme was implemented in a class of nine students.

Having involved in CPT, six of them volunteered to be peer tutors of thirty five tutees in the NPT scheme held in the semester that followed. Results of in-class observation, pre and post-test questionnaires and focus group interview were used to

– 320 –

evaluate the effectiveness of the two different models of peer tutoring scheme. The results suggested that most of the students have a positive attitude to the scheme.

About half of the students first joined CPT with negative attitude but they changed their mind after participating in the scheme. Tutors in NPT generally had positive attitude in peer-tutoring scheme, they find that it is a valuable experience for their personnel grown as take part in designing the course content. However, significant change of attitude to the scheme does not appear again on the tutees in NPT. This result was owing to reduction of course contents, ability of peer tutors or relative lower participance of the tutees to the tutors. We find that NPT can benefit the tutors much more than the tutees in their ability of management and responsibility, however, it take huge amount of administrative work in monitoring the scheme.

In the other hand, CPT request less resources and it benefits on both the tutors and tutees as the participants were requested to act both of the roles, which increased their participance.

– 321 –

D54 Intersubjective consciousness: Learning computer game making in companionship

Christina Aderklou, University of Gothenburgs, Senegal

Berner Lindstrom, University of Gothenburgs, Senegal

The need to study what happens when children on their own create games and learning material in computer game programs has becoming more urgent as computers and computer games have become every-day artefacts in Children's worlds.

What can we learn from children playing and creating software that could generate new pedagogical approaches in school? We used material, tools and processes that were more familiar from the Children's every-day living, rather than from the teacher's perspective. The pedagogical situation was based on acceptance of Children's sense of self-agency (Stern 2000). Three Swedish classes were involved in the studies one with 8 year olds, and two with 11 year olds. In the latter group we listened to Children's self-observations and reflections. In the eight year olds group we set up an experimental situation. The task in both groups was to learn advanced complex digital tools for the purpose of making digital school material (as well as and have fun). The concept of learning in companionship (Trevarthen 2002) and shared endeavour was used instead of instruction in complex digital learning situations at school. Intersubjective consciousness was defined to have three dimensions in the two studies. The first dimension of intersubjectivity concerns the present moment of being-with-someone, here-and-now; the second concerns the changes in role, and the third involves sharing with a not present unknown other. Results show an increase in self-agency and intersubjective consciousness was expanding to children-to-become-pupils. This was evident in game production, understanding of the educational theme and learning idea, aesthetic sensibility, future playability, altruistic dialogues and changed roles as a result of new thoughts and experiences.

The study indicates that computer game playing should be taken seriously as a context for different forms of learning and for finding new strategies for teaching, learning and transmission of knowledge account at school.

– 322 –

Teacher Professional Development

Discussant: Niki Tsaggaridou, University of Cyprus

D55 Differences in teachers' perception of instructional events depending on their school experience

Edgar Krull, University of Tartu, Estonia

Sirje Sisask, University of Tartu, Estonia

Kaja Oras, University of Tartu, Estonia

Kaarel Haav, University of Tartu, Estonia

This paper introduces a study of differences in teachers' perceptual capabilities as reflected in teachers' comments of observed lesson events. The aim of the study is uncovering and defining indicators of differences in novice and expert teachers' perceptual capabilities. Another aim of this study is comparison of its results with findings of other similar studies. The elaboration of research methodology for this study was based on the implementation of four consecutive tasks: developing a procedure for selecting teachers with different and identified levels of professional development as research subjects; producing a stimulus material for eliciting reactions reflecting differences in research subjects' perceptual capabilities and fixing conditions for the experimental study; developing a reliable methodology for recording and analyzing the elicited reactions by the stimulus material; and carrying out a comprehensive analysis of research outputs leading to the definition of indicators of teaching expertise and to the ways of their identification. A special focus was given to the provision of reliability of the categorization procedures of the research subjects' transcribed comments used for the qualitative content analysis. To this end the entire research design was thoroughly analyzed and adjusted in terms of structuring the stimulus material used for eliciting comments of research subjects, of the meaning and size of units (ideas, themes or frames) used for the qualitative content analysis, and of criteria used for the categorization of these units. The achieved relatively high reliability of categorization procedures of teacher comments justified a following cross-cases content analysis of comments by categories and by experience groups of research subjects. The preliminary findings of confirmed that there are clear qualitative differences in novice and expert teachers' perception of instructional events and other features characterizing teaching in the classroom.

– 323 –

D56 Teacher conceptualisations of quality in adolescent fiction: multiple perspectives, multiple practices

Rosemary Hopper, University of Exeter, United Kingdom

Debra Myhill, University of Exeter, United Kingdom

Set in the United Kingdom context of a National Literacy Strategy which focuses closely on analytical approaches to reading a variety of fiction and non-fiction; and alongside concerns that the private reading of fiction by adolescents may be diminishing, this paper explores secondary school (11 - 18) teachers' views on the fiction available to adolescents for use in the classroom and for private enjoyment.

Perceptions of quality in this fiction are examined, referring to earlier findings by

Whitehead (1977), Sarland (1994) and Benton (2000) and considered in relation to a notion of a canon of literature for the young reader. The research which the paper draws upon is a pilot study, the first stage in a longer proposed study, into perceptions of quality in fiction for adolescents. The findings of this research draw on semi-structured 26 interviews with teachers, librarians and classroom assistants

(TAs) from secondary schools in the UK in which they spoke about fiction for adolescents and how they interpreted quality in these texts. This paper reports on the multiple perspectives of quality expressed by teachers and illustrates the difficulty in identifying clear criteria of quality. The findings of this study are related to teacher education and the need identified by the Office for Standards in Education

(OFSTED) UK for teachers to have a wider knowledge of fiction for adolescents to support the development of reading within a school context.

D57 Video-Based Lesson Analysis as a Vehicle for Improvement of Teacher Pedagogical Content Knowledge

Rossella Santagata, LessonLab Research Institute & UCLA, United States

Giulia Angelici, Universita` Bicocca Milano, Italy

Research on teaching and teacher knowledge emphasizes the importance of teachers knowing and understanding subject matter in ways that enable its use in teaching (i.e., pedagogical content knowledge). The complexity of pedagogical content knowledge makes it challenging for educators to convey and for teachers to learn.

A recent approach to teacher education and professional development sees the engagement in video-based analysis of practice as a promising way to foster teachers' pedagogical content knowledge. Despite this growing interest in video-based analysis of practice, there is little empirical research on the effects of the use of video analysis on the improvement of teacher knowledge. Lacking are also studies

– 324 –

that document the teacher learning process. This study aims at documenting the learning process of a group of 50 pre-service mathematics teachers and a group of 16 in-service mathematics teachers who participated in a video-based lesson analysis course. Teachers were asked to complete a series of written analysis tasks focused on the integration of content and teaching strategies. Participants learning process was documented through their written responses to analysis tasks. Their improvement in lesson analysis abilities was assessed through a pre- and post-test.

The qualitative analysis of teachers' responses to course tasks unveils teacher thinking process while engaged in lesson analysis. Pre-service teachers' analysis abilities improved significantly. Initial analyses of the in-service teachers' pre-test show a better ability to reflect on the videotaped lesson than observed in the pre-service teachers' pre-test. In-service teachers provide comments and alternatives that are more elaborated. The data collected will inform the design of future video-based lesson analysis courses. The analysis of teachers' learning process will assist in the development of courses that are tailored to different levels of teacher analysis abilities.

D58 The Use of the Research Lesson Approach for Professional Development involving ìa Knowledgeable Other

Chris Dowson, Hong Kong Institute of Education, Hong Kong

There has been recent interest in active ways of raising teacher effectiveness through techniques such as teacher coaching and direct guidance. (Staub, 2004; Fischler,

2004). One methodology for examining and improving teaching in schools that is emerging in western settings is the use of research lessons, also known as lesson studies. The research lesson approach has its origins in Japan. It is essentially a process of professional development for groups of teachers who engage in critical analysis of the teaching of a topic or education aspect of a subject through observation with an objective of creating the most effective way of teaching the identified critical elements of the topic or education aspect. In this study of the teaching of

ESL to deaf Cantonese speaking learners, data gathering includes direct observation, video recordings, discussion, interviews and open-ended survey. There is extensive involvement of the researcher in this study. This has been referred to as the involvement of a ìknowledgeable other. In research lessons the functional roles of facilitator, teacher and lesson observer are shared by the researcher and teacher.

Following on from a description, the initial barriers, solutions and findings, use of

ìa knowledgeable other and the research lesson approach, as a research instrument and professional development device with groups of teachers teaching English as a

– 325 –

second language to deaf Cantonese speaking students are reported.

D59 Teacher Efficacy, Perceived Organizational Support, and Commitment of

Science Teachers Participating in a Teacher Support Program

Anastasia Kitsantas, George Mason University, United States

Donna Sterling, George Mason University, United States

Wendy Frazier, George Mason University, United States

Jennifer Hansen, George Mason University, United States

Juanita Matkins, George Mason University, United States

Molli George, George Mason University, United States

Charalambos Vrasidas, Cardet, Cyprus

The purpose of the present study was to determine the impact of a program for new provisionally licensed teachers, the New Science Teachers' Support Network

(NSTSN), on their self-efficacy to teach science, perceived organizational support, and commitment to their profession. The NSTSN provided a multi-faceted support system (coaching, mentoring, science methods courses, and a website) to twelve

(N=12) provisionally licensed science teachers. Teachers were asked to complete three online surveys, 2 self-efficacy scales and the school support scale, before and after exposure to the NSTSN. It was hypothesized that following the intervention, teachers would report significantly higher self-efficacy beliefs to teach science effectively and that they would be more committed to teaching. It was also expected that teacher's self-efficacy would be highly correlated with their commitment to teaching. Partial support for most of the hypotheses was found. The results suggest that the six elements of support provided by the NSTSN significantly enhanced teachers' self-efficacy to teach students of diverse backgrounds. In addition, teachers felt more efficacious in enlisting community involvement following the intervention. Although no significant differences emerged for perceived organizational support, continuance, affective, and normative commitment, there were significant correlations between these subscales and teacher self-efficacy suggesting that efficacious teachers are more likely to be committed to their school and the profession.

Overall, the findings of the present study show that educational interventions such as the NSTSN may be helpful in assisting novice teachers, particularly provisionally licensed teachers to improve their instruction skills and remain in the profession. The results are discussed from a social-cognitive perspective.

D60 Implementation of the Early Reading and Early Math Strategies

Marie Jose Berger, University of Ottawa, Canada

– 326 –

Renee Forgette-Giroux, University of Ottawa, Canada

The study of the Implementation of the Early Reading and Math Strategies is a multifaceted research investigation including both quantitative and qualitative research methodologies. Sponsored by the Ontario Ministry of Education (Canada), it has been designed to gather data over a two-year time period in order to provide a comprehensive picture of, and feedback on, the implementation of the Early Reading and Early Math Strategies Initiative Towards Professional Development. The research project explores many dimensions of the initiative. To begin, the study seeks to provide a greater understanding of current practices in reading and mathematics instruction and the leadership role of Principals in the context. Collecting and examining evidence of changes in teacher knowledge, classroom practice, and school leadership as a result of participation in this initiative is another dimension of the study. Finally, the effect this initiative has had on student achievement is explored.

Teacher professional development and vocational education

D61 Change to PBL in Danish Engineering Education

Annette Kolmos, Aalborg University, Denmark

During the last ten years, Higher Education has undergone tremendous changes. A lot of these changes are caused by institutional and external factors, as government policy concerning resources, educational and quality assurance policies. The framework set by government, is rather simple to describe. However, at the institutional level, institutions have developed many different pedagogical models, using very different strategies for development. Nearly all Danish Engineering institutions have implemented elements of Problem Based and Project Based Learning (PBL). Particular 5 Engineering University Colleges have undergone changes towards PBL.

Pedagogical Network for Danish Engineering Education (IPN) has been one of the central agents in the change processes for engineering education in Denmark. IPN has been responsible for staff and faculty development at the engineering university colleges and has been running the co-ordination for exchange of experiences among all Danish engineering institutions. However, it is not the same PBL-model which has been developed at the 5 different institutions - it is very different PBLmodels, developed on the basis of very different development processes. In this article, results concerning the change processes at 3 different institutions will be presented. The research results are based on interviews with rectors, central change

– 327 –

agents and staff developers at the institutions. The result is clearly underpinning the theory, that only top-down decisions at institutional level together with a pool of motivated staff, will cause changes at a system level. A bottom-up approach with decentralised development at departments leads to a variation within the institution, but it might be difficult to develop curriculum models at system level.

D62 Processes of change of teachers' conceptualisation and practice

Teresa Guasch i Pascual, Open University of Catalonia, Spain

Montserrat Castello, Open University of Catalonia, Spain

The aim of this research is to analyse the processes of change of teachers' conceptualisation and practice in teachers who were participating in a training process about how to teach and learn a learning procedure (note-taking). The participants were four teachers from three Secondary Schools. They participated in a training process in teachers' learning context, with the aim of facilitating professional knowledge construction and to help teachers to make more explicit the representations that support their practices. During the training sessions, they shared the sense and meaning of note-taking, knew a methodological proposal about the teaching and learning of note-taking, and designed and carried out a didactical sequence to teach learning procedures. The methodological proposal was different for each teacher: a) strategic teaching of note-taking procedures; b) technique teaching; and c) usual teaching.

The process consisted of weekly between one of the researchers and a teacher. The results from the analysis of resistances and agreements shared with the researcher during the training process, and the analysis of the recorded sessions, allow us to conclude that: 1) the initial conceptualisation of note-taking and its teaching and learning have an effect on the way of interpreting the training process, and therefore on the assimilation of the contents studied in the various sessions; 2) we can infer that the reflection and expounding of different viewpoints through the training and the teaching-learning process influence the change of the teachers' mental representations, and consequently their practice; 3) teachers' dominance theory is linked to the teaching methodology, and therefore directly affects the type of interaction in the classroom.

– 328 –

D63 Opportunities for teachers' professional development in three countries

Raimo Rajala, University of Lapland, Finland

Maria Flores, University of Minho, Portugal

Aki Tornberg, University of Lapland, Finland

The issue of teachers' professional development is becoming increasingly important for the national teachers' inservice training systems. In this comparative study teachers' opportunity for learning at school is regarded as a key factor to be capable to develop professionally. Also, the type of teacher community is seen to be related to opportunities to develop professionally. Leadership plays an important role in realizing policy promoting development and growth of organization members. And, finally, meaningful job content itself allows for workers development in great many respects. The research questions are as follows:

(1) What is the situation of teachers' professional development in Finland, Serbia and Monte Negro, and Portugal?

(2) How factors bearing on opportunities for learning at work, teachers' professional orientations, school leadership and job contents are related to professional development in the three countries?

The data of the research will be collected via closed ended questionnaires. The main research variables are professional development, opportunities for learning at work, school leadership, teachers' professional orientation and job content. Teachers represent primary and secondary levels of teaching in the focal countries. At the moment, the final data collecting procedure is being completed. The pilot study data is however available (N=89). Descriptive analyses of the pilot data suggest that the opportunities for regular professional development and motivation for it varied countrywise. Analyses into the relationships among the key variables were carried out using multiple regressions. The analyses suggest that both opportunities and motivation for regular professional development were predicted by good opportunities for learning at teaching work, high community orientation in teaching staff and effective leadership at school.In the final data, these tentative finding will be are revalidated. The results will be discussed with respect to the differences and the similarities between the three counties in teachers' professional development.

– 329 –

D64 The professor’s multimodal explanation and lecturers’ training

Naykiavick Rangel de Torres, Universidad de Carabobo, Venezuela

Marina Castells LLavanera, Universitat de Barcelona, Spain

This investigation arises motivated to the necessity of the Lecturer's Formation and it is developed in the line of Professor's discourse analysis. We select the electromagnetism like study topic taking account that, for its nature, it facilitates the development of the capacity of engineering student's abstraction, But, how to analyze the discourse of a professor and what theory to select for the analysis of this? In this investigation, the opportunity is presented, to study the theoretical mark of Perelman and Olbrechts, and apply it in the discourse analysis of the Physics Professors, being that an environment different to the juridical one accustomed. This research is oriented to develop a doctoral dissertation at the University of Barcelona, and is carried out in Venezuela, in the Faculty of Engineering of the University of Carabobo, taking like reference the expositive discourse of a professor in Electromagnetic

Physics. The instrument of gathering of data is based on the direct observation, non participative and using like support the recording in video of the classes and notes of the investigator's field. The theoretical framework has proven to be effective in capturing key aspects of physics explanations from a rhetoric-argumentative perspective; in fact, it is useful to identify the specific characteristics of the expositive discourse of physics professors and the resources used by them when looking for the adhesion of students to the theses that are being presented, what communicative modes the professors use and what functions they have in the coherent construction of scientific meanings. It is sought to guide the results of discourse analysis towards the professional improvement of the lecturers.

D65 The PISGA in Israel: A new model of professional development for in-service teachers

Anat Raviv, Haifa University, Israel

R. S. Paterson, Bringham young university, Canada

A major change in the in-service education of teachers was introduced in 2003 in Israel. Former teacher developed centers, jointly operated and administrated by ministry of education and local municipalities were replaced by a new organizational phenomenon called 'PISG"A', the new name signaled a new conception of teacher education for practicing teachers. The PISG"A centers aim is to offer an answer to the frequent changes in the education field, country priorities and the educational agenda of the Israeli ministry of education. In order to develop teacher's knowledge

– 330 –

and building new practical ideas and experience the teacher's training will be based on professional growth, experience and learning from the materials and methodological tools that are offered combined whit reflection on the learning/ experience action (Zeichner, 1993; Schon, 1988; Zilbershtein, Ben-Perez, Ziv 1998). Studies suggest that professional development experiences that include most of these characteristic can have a substantial positive influence on teachers' classroom practice and students' achievement (Kroath, 1990; Nespore & Barlyske, 1991). The paper critically explores the experience of the "new" frame of in-service for teachers in the

PISGA centers in Israel from dual perspective (an instructor/lecture in the PISGA and a visiting teacher educator from Canada and the U.S.A who has spent nearly

25 years as a Dean of education in large teacher education institutes). It presents teachers views and experience that participate in the training program and as well as offering suggestions for change.

The authors of the proposal and paper have identified a number of principles of success which have relevance not just for the Israeli PISGA, but for teacher education elsewhere.

D66 Professional development: an Activity-Theoretical study of expansive learning in good practices

Wietske Miedema, University of Amsterdam, Netherlands

M. Stam, Polytechnic of Amsterdam, Netherlands

G.T.M. Ten Dam, University of Amsterdam,, Netherlands

J.H.A.M. Onstenk, Polytechnic InHolland, Netherlands

W.L. Wardekker, Free University of Amsterdam, Netherlands

Educational innovation requires teachers to be lifelong learners and change agents: without teachers who are willing to change their day-to-day practice and simultaneously change the context for this practice, educational innovation cannot take place.

Whether educational innovation will in actual fact take place depends to a great extent on whether teachers are capable of further developing their competencies and whether they are capable of creating a challenging and nurturing learning environment for themselves as well as for their students, with the help of the broader context of the school. This paper focuses on the development of the competencies of teachers who work within the context of good practices in innovative vocational schools in The Netherlands. The good practices were selected because they were developed in a bottom-up, rather than a top-down way. We report some tentative results of four qualitative case studies that look into the expansive learning, individual and team-learning, that takes place in these good practices'. The activity theory is

– 331 –

the theoretical framework that helped us analyse our data. The tentative conclusions deal with the tensions and contradictions in the activity system, or between activity systems, as the motor of the ongoing development and growth of competencies, by teachers and by teams of teachers. The study reports on factual developments, on experiences and on the developments of attitudes and visions of teachers. It ultimately focuses on the development of meaning within teams and personal sensemaking by individual teachers with regard to the innovation process.

D67 Is the teachers' understanding of learning reflected in their preparation of lessons? A project to foster self-regulated learning in vocational education

Astrid Elke, University of Basel, Switzerland

Sandra Grieder, University of Basel, Switzerland

Corinne Tiaden, University of Basel, Switzerland

Gerhard Steiner, University of Basel, Switzerland

The complete research project is designed to optimise learning processes in vocational education. Therefore teachers are trained in workshops to encourage students to use learning strategies. This part of the project focuses on the teachers' influence on the students' learning processes. 32 teachers were asked in a standardized inter-view about their cognitions of learning and their procedure when preparing lessons. The data were processed by TAMS Analyser. We wanted an answer to the questions: a) What are the teachers' conceptions of learning? b) Are the teachers' conceptions of learning reflected in their conceptions of the lessons? c) Do learning strategies play any part in the teachers' beliefs about learning and lesson preparation? d) Can any changes be found after the intervention? The teachers' concept of learning was divided into surface processing (content focus and competence focus) and deep processing strategies (meaning focus teachers and growth focus). One third of the teachers mentioned more than one learning concept. With regard to lesson preparation for one third of the teachers the curriculum was the only important orientation. Other important structuring procedures were development according to content, learning process, starting phase of the lesson, and aim of the lesson. There is no systematic connection between learning concept and lesson preparation. With regard to learning strategies, they were rarely mentioned. In con-text with the learning concept some teachers referred to cognitive strategies and hardly any to motivational ones. Metacognitive strategies could not be found. In the lesson part of the interview only very few teachers mentioned cognitive strategies. The results show that learning strategies are not considered often when describing what learning is or how to prepare a lesson. Furthermore the understanding of learning is not consist-

– 332 –

ently transferred to the development of lessons. The effect of our workshops will be analyzed in due course.

Learning and Instructional Technology

Discussant: Charalambos Vrasidas, Cardet - Intercollege, Cyprus

D68 Overcoming the tyranny of distance via an Interactive Satellite System in university education in remote Australian outback

Juhani Tuovinen, Batchelor Institute of Indigenous Tertiary Educati, Australia

This research sought to assess the value of a satellite-mediated interactive distance learning (IDL) system for University programs in remote northern Australia. The

IDL system has used in two Australian states for school education. 167 remote sites can be reached simultaneously in the Northern Territory (NT) from two studios.

Each of the two studios can broadcast and be in two-way communication with all of the sites in NT. This is one of its key attractions for education at any level. The system allows a clear full-sized video of the teacher and good quality sound to be displayed to all students. The students' pictures are not available to the teachers, but they can hear the students. The system also includes an Internet connection between the students and the studio. Many different systems are used to provide interactivity over distances, with variation in quality of and the numbers of sites linked. Many factors that could influence the teachers' and students' experiences of interactive video teaching have been studied, such as participants' prior experiences, the availability of support staff, etc. Overall, it appears students learning in interactive video-mediated contexts learned as well as in normal face-to-face teaching situations or better. The university community indicated varying amount of interest in the use of this system. Some programs with up to 1000 scattered students were interested.

Others felt that there were too many resource, time or training limitations for the system to be worth using. This study highlighted the importance of the interacting technical, organisational, support, pedagogical, discipline-specific and human relational aspects of the satellite-mediated interactive programs. The significance of this study demonstrates how these issues can be fruitfully tackled in planning, designing and implementing distance courses mediated by a live interactive audiovisual communications system over huge distances under extremely demanding conditions.

– 333 –

D69 An equitable postmodern curricular praxis for internationalised web-based distance learning in higher education

Henk Eijkman, Monash University, Australia

While universities are increasingly offering their courses transnationally, and via internationalised curricula, little research attention is devoted to equity considerations despite the increasing participation of socio-culturally diverse student cohorts.

This study addresses this under-researched aspect of internationalised web-based curricula by asserting that all students in transnational programs need to be able to engage, on an equal footing, with issues of difference, conceptual dissonance, hypercomplexity, and competing value systems. Moreover, they increasingly need to be able to do so in web-based distance learning environments which provide the most cost-effective delivery platforms for internationalised curricula. The study uses a comprehensive review of learning theories in conjunction with a qualitatively-oriented case study of a web-based access program to reveal how a critical postmodernist curricular praxis is better placed to promote the equitable educational engagement and learning outcomes for socio-culturally diverse student cohorts in internationalised curricula. This study is of particular significance because its critical theoretical review of dominant learning theories in web-based programs and empirical case study focused specifically on equity considerations. The analysis of the review and of the data obtained from staff, students and the content analysis of the access program's curriculum indicate that curricular practices framed by cognitivist and even constructivist assumptions continue, in the final analysis, to be tied to individualist curricular practices which inhibit students' ability to respond adequately to issues of difference, diversity, hypercomplexity, conceptual dissonance and competing value systems that are inherently social rather than individual in nature. Instead, the analysis points to the efficacy of a distinctly social and critically pragmatic framework for curricular praxis which enables students to collectively and critically explore and compare knowledges, cultural values, and practices embedded in their own and their academic discourses and thereby engage equitably with a world that is not just diverse but multi-aspectival in its presentation.

– 334 –

D70 O

Ilaria Mancini, University of Rome La Sapienza, Italy

Barbara Maroni, University of Rome La Sapienza, Italy

Moving from possible application of some aspects of Conversational Analysis

(Sacks, Schegloff, Jefferson, 1974) and of Sequential Analysis (Bakeman, Gottman,

1986) to forum on line for collaborative knowledge building, in a previous work we examined on line discourse development in a community (Mancini, Maroni, 2004).

We analysed turn-taking referring to modalities of selection for the first referent and to modalities of opening note (Ligorio, Maroni, Mancini, 2004). Aims of this study are:- To single out specific aspects of this CMC asynchronous context; - To extend the analysis to forum differently marked for participants and aim of the discussion.

We chose forum so that they would be homogeneous referring to participants and activity. Participants are grown up people, discussing about a theoretical topic.

Data are forum on line notes from Synergeia, a web based platform. We use two category systems: Receiver and Opening. Categories of Receiver are Community or Specific Participant. To single out Opening as unit of analysis we consider the syntagm and categories are: Nominal, Verbal, Prepositional, Adjectival, Adverbial,

Wh, Connection and Interjection. We expect to single out specific aspects of this

CMC asynchronous context and to find out preferences in selecting receiver and using syntagm modalities.

D71 Identity for learning in the web: A methodological proposal

Paola Francesca Spadaro, University of Bari, Italy

Vito Fabio Fraccascia, University of Bari, Italy

The main core of this study is the relationship between identity and learning, in particular, digital identity and computer-based learning. The relationship between identity and learning was developed in the past (Bruner, 1997), and now is a field of interest for many researchers studying the specific effects in digital environments

(Ligorio, Pugliese, Spadaro, 2004; Talamo, Ligorio, 2004). In their current studies they use concepts as ´dialogical selfª (Hermans, 1996), ´self positioningª (Harre' e Van Langhenove, 1991) and ´participationª (Wenger, 1998) that because of their complexity are difficult to be operationalized for the research. In this study we suggest an analysis method we call ìintegrated, by which we think it is possible to study the features of such complex concepts within collaborative learning contexts mediated by asynchronous digital environments. This approach is a combination of qualitative and quantitative methodologies through a three-step-procedure. Such

– 335 –

methodology will be supported by some examples of preliminary researches.

D72 Online Mentoring to Enhance Authentic Learning

Irja Leppisaari, Central Ostrobothnia Polytechnic, Finland

Riina Helenius, Central Ostrobothnia Polytechnic, Finland

This paper deals with the Online Mentor Project, which aim is to develop a new leverage model for the needs of online education. The model is expected to strengthen the relationships of higher education, working life and meaningful learning. As representatives of the working life, online mentors provide a link of more authentic problems and experiences as well as new models of thinking and working to distance learning. The development of expertise in mentoring takes place in AVERKO's online courses, which are organised in co-operation with students, tutors and representatives of the working life for this project. Learning tasks, theme discussions and practical examples connected with the real working life seem to enhance authentic learning. Meaningful learning connects knowledge into situations, where theory and its elucidation are connected. At its best, mentoring promotes active and critical reflection, where both students and mentors examine their experiences and views, and doing so, construct new knowledge together. Through the project, students will be given an opportunity to enrich their learning through accounts of authentic problems experienced by experts. Online mentoring makes online teaching and learning in higher education more working life oriented than before.

D73 Integrating different aspects of the learning and teaching process. How understanding, dialogical teaching and pedagogical sensibility interrelate creating dynamic learning environments

Kyriaki Doumas, Department of Education, Lund, Sweden

This study investigates how teachers, students and object of learning interrelate in a classroom situation within the institutional framework of formal schooling, when dynamic, complex, meaningful and joyful learning environments are created.

Further more some other issues are considered such as the role of the school subject, (humanistic and natural sciences), and the level of education. A qualitative approach was adopted exploring the experiences of students as they were expressed in semi-structured and in-depth interviews. The recent years themes concerning the construction of the space of learning have been explored by many scholars in the educational research field. It has been pointed out the distinction between the intended object of learning and the the ìreal, the enacted, object of learning which

– 336 –

can be quite different as it is constituted jointly in the interaction between teachers and students in the particular classrooms. The space of learning, is investigated here in a wider form as the pedagogical teacher-students interrelation is explored as possible constitutional element of the learning process. The following aspects were distinguished through the analysis of students' interviews: In-depth understanding as a result of dialogical teaching carried out with pedagogical sensitivity and sensibility. However, these three aspects are conceived as interwoven elements, all essential for the enhancement of the learning potentiality of the teaching process. The space of the enacted object of learning which is constituted in the teacher-students interrelation integrates cognitive, emotional and ethical aspects. Learning then is not only an intellectual dynamic but an empowerment of the whole person which brings even meaning and joy.

D74 Designing and Implementing a Technology-enhanced Inquiry-Based Learning Environment: Developing an integrated understanding of geology

Xornam Apedoe, University of Georgia, United States

The quality and characteristic of students' science knowledge, and the methods of teaching science have been under question for many years (e.g. Duschul & Hamilton, 1998) and new initiatives have been taken to redefine and re-conceptualize what science education should look like. The innovation explored in this study was the use of inquiry-based learning activities supported by access to, and use of, resources in a digital library. Of primary interest here was how a technologyenhanced inquiry-based learning environment could be structured to support the appropriation of, and engagement in, scientific practices and discourse, helping students develop an integrated understanding of geology. Preliminary findings suggest that a number of factors within the technology-enhanced inquiry-based learning environment may have a significant influence on student development of integrated understanding of geology. For example, differing levels of instructor knowledge for not only geology content, but also for pedagogical knowledge of inquiry-based teaching techniques appears to play a significant role in fostering and hindering student desire to appropriate and participate in inquiry processes. Implications of these and other results for the design and implementation of technology-enhanced inquiry-based learning environments in undergraduate geology and/or other science courses will be discussed in further detail during the presentation.

D75 Spatial Framewokrs in imagined navigation

Sofronis Sofroniou, University of Cyprus, Cyprus

Marios Avraamides, University of Cyprus, Cyprus

– 337 –

Past research on spatial frameworks (e.g., Franklin & Tversky, 1990) has shown that a set of orthogonal axes centered on our bodies is used to encode and remember the locations of objects that surround us. Work by Avraamides & Carlson (2003) has extended the spatial framework research to situations in which we imagine ourselves moving within a perceptually available space even in the absence of any objects. The present work examines the conditions under which the spatial framework model can be used to characterize imagined navigation. The results of our experiments suggest that an embedded perspective is critical for the spatial framework model to apply. Instead, the occasional updating of the imagined facing direction

- proposed by Avraamides & Carlson (2003) as a precondition for spatial frameworksódoes not seem necessary.

D76 Fostering the Comprehension of Diagrams by Interpreting and Constructing Diagrams in Multimedia Environments

Jessica Phillipp, University of Education, Freiburg, Germany

Rolf Ploetzner, University of Education, Freiburg, Germany

The goal of an empirical study was to investigate how learners gain knowledge about representational systems and how learning activities like interpreting and constructing diagrams can foster their comprehension. For learners it is often difficult to understand the affordances and constraints of representational systems and to use them flexibly in different domains. From an instructional design perspective, multimedia learning environments offer a wide range of possibilities for teaching the use of external representations like graphs and diagrams. They facilitate many forms of presenting, constructing and manipulating them. But only little is known about the effectiveness of such activities like constructing or interpreting diagrams.

Our study therefore analysed whether self-construction of representations or the interpretation of presented representations is more successful in supporting learners to understand representational systems and to apply these representational systems as a reasoning tool. Taking into account the research about learning with workedout examples and several transfer paradigms, different combinations of interpreting and constructing diagrams were considered. We compared four groups of fifth graders regarding learning and transfer outcomes. In a treatment phase they had to solve kinematic problems of increasing difficulty using distance-time-graphs. On each level of task difficulty the students had to solve two tasks with structurally identical diagrams. The first group solved both tasks interpreting presented diagrams. The students of the second group had to construct the diagrams in both tasks themselves.

– 338 –

In a third group, the students were given a diagram in the first task and had to construct their own diagram in the second task. Finally, the students of the fourth group had to construct their own diagram in the first task and solved the second task with a presented diagram. We compared the four groups regarding their performance in two post-tests considering short and long term effects.

24 August 2005 12:30 - 13:30 Conference Center Room A

Keynote Address

Chair: Anastasia Efklides, Greece

When effective becomes meaningful and when learning becomes participation-

Conceptual shifts in research on social interaction in learning and instruction

Paivi Kristiina Kumpulainen, University of Oulu, Finland

This keynote address will provide a historical overview of research on social interaction in learning and instruction. Whilst using our own experiences and empirical research work as points of departure, the presentation will illuminate conceptual shifts in social interaction research across time. Specific attention will be directed to changes in research purposes and related questions, conceptual frameworks, as well as data collection methods and analysis. These theoretical and methodological choices are suggested to be linked with the ontological and epistemological assumptions each strand of research has followed in a given time and historical context. This is also reflected in the ways in which the concepts of social interaction, learning and instruction and their interrelationships have been defined. Via the review, the keynote address shows how social interaction research in the field of learning sciences has become a rich and vivid area of scholarly investigation. In the presentation, introductions will be given towards studies of social interaction embedded in a process-product paradigm, cognitive perspective, as well as in sociocultural and situated perspectives. The methodological approaches applied by these studies have often drawn on psychological and domain-specific analyses, discourse analysis, ethnomethodology as well as on ethnography. A unifying trend in recent studies is seen in the acknowledgement of the dynamic interplay between multiple agents and contexts mediating the opportunities for individual and communal action and learning. The recognition and a wider comprehension of the dynamics of these agents and contexts has become an important research agenda for recent investigations. In sum, via the illumination of conceptual shifts in social interaction

– 339 –

research across time, the keynote address aims at furthering the dialogue about the potential and challenges each strand of research has brought towards understanding the psychological, social, linguistic and cultural elements that shape the practice of learning and instruction in diverse contexts.

24 August 2005 12:30 - 13:30 Conference Center Room B

Keynote Address

Chair: Andrea Karpati, Eotvos University, Hungary

On warm conceptual change: The role of personal epistemology in knowledge restructuring

Lucia Mason, University of Padua, Italy

In their classic 1993 article, Pintrich, Marx, and Boyle called for research on ìhot conceptual change, that is, for affective, motivational, and situational factors that may affect knowledge changes to be considered. Until then -- but even now to a certain extent -- research on conceptual change was focused on students' prior knowledge structures, developmental changes, and teaching strategies. This presentation introduces a line of research on warm conceptual change by analyzing the role of personal epistemology in knowledge revision processes, which was included by Pintrich (1999) among the motivational beliefs acting as resources or constraints. As it comprises beliefs about the nature, source, and justification of knowledge, personal epistemology is necessarily influential in knowledge restructuring. Starting from a theoretical perspective with the paradigmatic approaches to the investigation of personal epistemology, the outcomes of the few studies which have examined its effects on conceptual change, will be then reviewed. They suggest that students who believe in knowledge as immutable, absolute and certain, and made up of compartmentalized facts cannot engage in conceptual change to the same extent as students who believe that knowledge is hypothetical, evolving, and interconnected in its structure. To this regard the open question is how epistemological beliefs facilitate or impede the knowledge restructuring process. The mechanisms implied in these processes, such as systematic processing, will be discussed with reference to a ìdual process model of conceptual change in which intentionality is a key aspect. Examples from my own research will illustrate the argument. In addition, the significance of the development of epistemological thinking and the orchestration of powerful learning environments which enhance beliefs about knowledge and knowing will

– 340 –

be outlined in the light of outcomes from intervention studies.

24 August 2005 12:30 - 13:30 Conference Center

Keynote Address

Chair: Roger Saljo, Gothenburg University, Sweden

Room C

Hands and minds: Possible educational implications

Demetris Natsopoulos, University of Cyprus, Cyprus

Handedness as an index of functional brain asymmetry has stimulated great research interest for the last three decades, since Levy's work in 1969. This has culminated in a number of theories and a vast literature spanning many areas, such as neuropsychology, neurology, genetics, biology and educational psychology. One of the more prominent is Annett's theory of human balanced polymorphism (1985).

According to this theory hand preference and hand skill are associated with cerebral function (speech lateralization, mathematical thinking, spatial cognition, arithmetic and reading ability and higher general intelligence). Both extreme left- and righthanded individuals are at risk to develop deficits in the areas of cognition in contrast with those individuals, left-and right-handers, who are mildly biased to dextrality.

A second school of thought, also biologically oriented, posits that handedness need not involve cerebral function, and individuals less biased to the right are not intellectually advantaged (McManus et al., 1993; 2002). A third theory, which draws upon neuropsychological literature, distinguishes between pathological left-handers, as the result of brain injury, and normal left-handers in the general population

(mainly Bishop, 1990; Coren, 1992). Finally a theory by Geschwind (Geschwind and Galaburda 1985) claims that left-handedness is only partly genetic and partly a result of the intrauterine influences, such as excessive levels of testosterone, which causes delay of development in left-handers, learning disorders, and sometimes yields left-handers talented in space cognition and language. This invited talk is intended to give an overview of our current understanding and the speaker's empirical work (Natsopoulos et al., 1998, 2002). In addition, I intend to demonstrate:

(i) whether handedness is linked to cognitive deficits in general, (ii) specifically, if there exist subgroups of handedness that may be at risk to develop such cognitive dysfunctions, and (iii) whether intervention programs can be devised to help those individuals with difficulties.

– 341 –

E 1 24 August 2005 14:30 - 16:30 Room B107

Symposium

Mathematics Education

DEVELOPING PROCEDURAL FLEXIBILITY IN ELEMENTARY

SCHOOL ARITHMETIC

Chair: Lieven Verschaffel, University of Leuven, Belgium

Organiser: Lieven Verschaffel, University of Leuven, Belgium

Discussant: Elsbeth Stern, Max Planck Institute for Human Development,

Germany

One of the most important recent issues in (research in) mathematics education is how students can be taught curricular subjects so that they develop adaptive expertise. In his foreword of the book The development of arithmetic concepts and skills. Constructing adaptive expertise edited by Baroody and Dowker (2003), Hatano (2003, p. xi) describes adaptive expertise as ìthe ability to apply meaningfully learned procedures flexibly and creatively and opposes it to routine expertise, i.e.

ìsimply being able to complete school mathematics exercises quickly and accurately without (much) understanding. Although the constructs of adaptive and routine expertise were introduced by Hatano already more than two decades ago (Hatano,

1982), and although terms like adaptivity and flexibility have been used with increasing frequency by researchers and practitioners in the field of mathematics education, only few systematic and deliberate attempts have been made to rigorously study (a) adaptive expertise as a competence, (b) the acquisition of adaptive expertise, and (c) the cultivation and (d) assessment of adaptive expertise in the academic domain of mathematics. The overall aim of this symposium is to present four new studies dealing with teaching and learning to make use of arithmetical strategies in the domain of elementary school arithmetic in an adaptive or flexible way, and to reflect on the psychological and instructional implications of these studies for different groups of pupils, including mathematically weak and disabled children.

– 342 –

Strategy development in elementary school Children's simple addition

Dietmar Grube, Georg-August-Universitat Goettingen, Germany

Elementary school children's strategy use in solving simple addition and subtraction tasks seems to shift from counting strategies to direct retrieval from knowledge base and to decomposition strategies (also based on knowledge retrieval) very slowly. This view is supported by two different lines of research, the chronometric approach (e.g. Ashcraft & Fierman, 1982) and the observation/interview approach

(e.g. Carpenter & Moser, 1983). In the present study, the technique of ìarticulatory suppression (AU) was adopted from research related to Baddeley's working memory model in order to gain further evidence of the course of strategic change. It was argued that if children solve simple addition tasks using verbal counting strategies, continuous articulation of an irrelevant word during calculation should reduce addition performance (Hitch, Cundick, Haughey, Pugh, & Wright, 1987).

In two experiments, children were continuously articulating (experimental condition) or were tapping the space bar of the computer keyboard (control condition) while performing a simple addition verification task (sums < 20). In Experiment 1, first-, second-, third- and fourth-graders were tested at the end of the school year.

In Experiment 2, second- and fourth-graders were tested at the beginning of the school year. Here additional observation/interview methods were implemented. In both experiments, the effect of AU occurred only with the youngest age group and only with problems involving sums passing over ten. Therefore, this study did not demonstrate any use of counting strategies by children at the end of the second grade and older. The results of the observation/interview method implemented in

Experiment 2, however, indicated ample use of verbal counting strategies for both age groups. It is proposed that the distinction between use of (indicated by observation/interview) and dependence on verbal counting strategies (indicated by AU) contributes to a more complete view of strategy development in simple addition.

– 343 –

Split or jump? Children's strategy choices in the number domain up to 100

Joke Torbeyns, University of Leuven, Belgium

Lieven Verschaffel, University of Leuven, Belgium

Pol Ghesquiere, University of Leuven, Belgium

This study aimed at analysing the adaptive nature of children's strategy choices in the domain of adding and subtracting up to 100, starting from Lemaire and Siegler's theoretical and methodological framework. In their model of strategy change,

Lemaire and Siegler (1995) distinguish four parameters to characterise children's strategy competencies, namely strategy repertoire, distribution, efficiency, and adaptiveness. To gather unbiased data about strategy efficiency and adaptiveness,

Siegler and Lemaire (1997) propose the choice/no-choice method, which requires testing all subjects in two types of conditions, namely (a) a choice condition, in which subjects can use their preferential strategy on each item, and (b) one or more no-choice conditions, in which they have to solve all items with one strategy.

Sixty-nine second-graders of different mathematical achievement levels solved a series of additions and subtractions up to 100 both before and after they had received systematic instruction in the topic. At both measurement times, the series of items was offered in one choice condition and two no-choice conditions. In the choice condition, children could choose between the split and the jump strategy on each item. In the two no-choice conditions, they were instructed to solve all items with, respectively, the split and the jump strategy.

Our analyses revealed a surprisingly low number of adaptive strategy choices both before and after instruction in the number domain up to 100. Only high-achieving children took into account item characteristics during the strategy choice process; neither high- nor low-achieving children fitted their strategy choices to their individual strategy performance characteristics. These results might be explained by the strong instructional focus on the jump strategy, which might have prevented children to develop the necessary knowledge and skills to flexibly apply diverse strategies in the number domain up to 100.

Development of young children's strategic competence in numerosity estimation

Koen Luwel, University of Leuven, Belgium

Robert Siegler, Carnegie Mellon University, United States

Lieven Verschaffel, University of Leuven, Belgium

This study used the microgenetic method to investigate the discovery and development of a smart numerosity estimation strategy. The task consisted of estimating

– 344 –

numerosities of blocks presented in a 10 x 10 grid. This task allows for the use of two strategies: the addition strategy whereby estimates of different subgroups of blocks are added to a running total and the smart subtraction strategy wherein the number of empty squares is subtracted from the total number of squares in the grid.

Thirty-nine second graders, who had not discovered the subtraction strategy yet, were assigned to two conditions differing in terms of the frequency of items eliciting the subtraction strategy (10% vs. 90%) in the practice sessions. Pupils ran eight sessions (three test sessions, four practice sessions and one transfer session). Data were analysed in terms of: the rate of discovery and generalization of the subtraction strategy, overall task performance, and adaptiveness of strategy choices. Results revealed that participants discovered the subtraction strategy earlier in the subtraction than in the addition condition, in terms of the number of sessions but not in terms of the number of critical trials. Second, the generalization of the subtraction strategy increased with increasing session number, and was larger in the subtraction than in the addition condition. Third, during the study we observed an improvement in overall task performance on the trials with the highest numerosities. Finally, the adaptiveness of strategy choices tended to increase during the course of the study and was somewhat larger in the subtraction condition than in the addition condition.

These results suggest that a small, well-selected number of critical items might be sufficient to promote the discovery and the subsequent development of a strategy.

Traditional and realistic strategies used for solving division problems in Dutch national mathematics assessment at the end of primary school

Cornelis M. van Putten, Universiteit Leiden, Netherlands

Gabrielle Rademakers, Hogeschool InHolland, Netherlands

Meindert Beishuizen, Universiteit Leiden, Netherlands

Jan Janssen, Citogroep, Netherlands

In this study we collected and categorized the strategies used to solve division problems in the Dutch national mathematics assessment of 1997 at the end of primary school. Traditional long division algorithms were used in 43% of the written solutions; realistic strategies were seen in 22% of these solutions, and mental solutions

(only an answer, no working) in 31% of the cases. Most students were consistent in their strategy use over the set of problems: 46% could be characterized as using a traditional strategy either exclusively or combined with mental solutions; 29% as using a realistic strategy exclusively or combined with mental solutions; 19% as exclusive mental problem solvers. On average students correctly solved 5 of the

10 problems. Division word problems without decimal fractions were more often

– 345 –

solved correctly than word and number problems with decimal fractions. Student achievement in the traditional strategy group was significantly better than in the realistic strategy group. The decrease in Dutch national achievement for written arithmetic over the 1987 - 1997 period might therefore be attributed to a certain extent to the assumedly increased number of students using realistic strategies since the introduction of realistic mathematics textbooks at the end of the 1980s and in the 1990s.

E 2 24 August 2005 14:30 - 16:30 Room A108

Symposium

Teaching and Instructional Design

MAKING LEARNING POSSIBLE: THE ROLE OF PEDAGOGY FOR EN-

HANCING LEARNING

Chair: Jonas Emanuelsson, Gothenburg University, Sweden

Organiser: Ulla Runesson, Gothenburg University, Sweden

Discussant: David Clarke, Melbourne University, Australia

The overall aim of the symposia is to discuss constraints to and possibilities of learning from the point of view of the object of learning. A common theme in the presentations, although not always explicitly advocated, is: pedagogy matters! These papers all relate to teaching and learning practice and, are illustrated by examples from different subjects from primary to university level. The papers have another thing in common, they are grounded on the same theoretical premises learning means to be aware of certain aspects of that which is learned. In her paper Vikstr?m reports that differences in students' understanding of life circle were related to those aspects they discerned and were aware of. If learning implies the awareness of critical aspects, consequently, to provide learning implies sensitising the learner's attention to these aspects. In their paper, Watson and Mason discuss this in terms of the mathematical problems students encounter, particularly how these affords generalisations. In flour of the papers (Holmqvist et al., Ling & Ying, Pang & Marton, and Runesson) the authors demonstrate how teachers can systematically explore the critical aspects of learning in a learning study so they can be enacted effectively.

The results from these studies clearly show that, whether or not the learners had the opportunities to experience the critical aspects had consequences for the learning outcomes. One observation that be drawn from these studies is that it is important

– 346 –

to take the learners' acts of awareness into consideration, rather than the acts themselves.

The paradox of pedagogy

Ming-Fai Pang, University of Hong Kong, Hong Kong

Ference Marton, Gothenburg University, Sweden

One of the few principles that all educationalists seem to agree on is that learning takes an active involvement on the part of the learner. Moreover, it is generally assumed that the more active the involvement is, the better learning will result.

Pedagogy, on the other hand, is about helping the learners, making it easier for them to learn. Bringing these two principles together might then mean that the more powerful the pedagogy is, the less powerful the learning becomes. This sounds not only counter-intuitive, but clearly paradoxical. We will try to point out a way of solving this paradox, by means of an empirical study in which two groups of teachers set out for finding the best way to teach a difficult concept/phenomenon in

Economics. Both groups worked according to the learning study model (which is a combination of design experiments and Japanese lesson studies). One group varied one aspect of the phenomenon at a time, and brought the different aspects together subsequently, while the other group let all aspects vary at the same time. Following the above paradox, the second group would outperform the first group (because of higher active involvement), but in actual fact the opposite happened (the first group outperformed the second group, both on immediate and a delayed test of understanding after learning). These results could be interpreted within the framework of variation theory: Pedagogy is not so much supposed to make learning easier, but to make learning possible. When learning is possible and the learners want to learn

- they will. If pedagogy fails to make learning possible, it fails to support learning at all. But making learning possible is not enough. Only the learner can make it happen. By realizing this, our results can be understood and the above paradox can be solved.

LEARNING PATTERNS: How contrasts affect students' learning outcomes

Mona Holmqvist, Kristianstad University College, Sweden

Laila Gustavsson, Kristianstad University College, Sweden

Anna Wernberg, Kristianstad University College, Sweden

This study aims to describe the ways different patterns of contrasts used in educational settings affect the students' learning outcomes. Learning can be seen in

– 347 –

different ways by teachers. If learning in some way is seen as similar to building a house, a teacher can give the students a detailed description of how to build, without concerning if they have built before. They just build by the description, while the teacher has already decided exactly how this should be carried out. The students do not even take the risk to make mistakes; they always succeed if they follow the directions. In a second way, a teacher can focus upon the buildings the students have started to create by themselves, without giving them a detailed description. During the creation phase the building becomes even better than the students have planned by themselves. The initial plans my have been altered, because of the suggestions from the teacher in combination with a deeper understanding by the student. Finally, the teacher can recognize how the students have built models of a learning object but would like them to build knowledge different from what the student already has.

The student can be forced to tear down her/his own beliefs. In all three examples above the way contrasts are used to make the student see the critical aspects of the learning object differ and affect the learning outcome, both in a short and long time perspective. An understanding for which pattern of contrasts each learning situation needs is often an implicit knowledge that teachers create during experience. In this study we make this phenomena explicit by describing the patterns we have found in five different learning studies in three different school subjects, combined with an analyze of how the different patterns affect the students' learning outcome.

Sequence and variation: A learning study of the angle concept

Ulla Runesson, Gothenburg University, Sweden

In mathematics education, it is well recognised that it is necessary to see the angle as turning in order to understand the concept fully and, consequently, that the concept should be taught accordingly. The teachers in this study anticipated the significance of this on the basis of what they had got to know about their students' pre-knowledge on a diagnostic test. The teachers worked in an explorative process called the learning study. This is premised on variation theory. From a variation theory perspective learning implies being able to discern critical aspects of that which is learned. Promoting learning possibilities implies providing for the learners to attend to those aspects. A learning study is a cycle of planning and revising the lesson. It starts with one teacher teaching the first lesson planned on the basis of a pre-test of the pupils' prior knowledge. The lesson is videoed and after the lesson the pupils are tested. Based on The test results and on the teachers' reflections of the recorded lesson, the lesson plan is revised. Next, the second teacher teaches her class according to the new plan. This lesson is similarly videoed taped and the

– 348 –

pupils are tested. Then, if necessary, the lesson plan is further revised. To identify critical aspects of the space of learning afforded to the learners, the data from this learning study was analysed to reveal differences in the teachers' handling of the object of learning and how this was reflected in pupils' learning outcomes on a post-test. Although, angle as turning was brought out in all the lessons, the students' learning outcomes were different in the different classes. It is demonstrated that how the idea of turning was sequenced and varied was both a critical constraint and an opportunity for learning.

Enriching student learning: Structuring tasks which limit possible generalisations

Anne Watson, University of Oxford, United Kingdom

John Mason, Open University, United Kingdom

This paper presents issues and practices which address the effects of variation in mathematical questions. It is a theoretical paper based on our extensive experience working with learners of mathematics, beginning and experienced teachers and teacher educators. Learning mathematics is seen as becoming acquainted with generalisations of several types: concepts, techniques, classes of objects, properties, relationships and theorems. Through detailed consideration of examples from classroom practices we look at some typical classroom mathematics questions and ask what generalisation, and hence what learning, is afforded for the learner by the variations and in variances within the question sets. Paradoxically, the more constraints are imposed in sets of traditional mathematics questions, the more likely they are to afford appropriate generalisation for the attuned learner. We go further and claim that all learners generalise in all situations, and may be frustrated when it is not clear how to do so. They may not, however, be attuned to generalise in the ways expected by the teacher. We give several examples of the ways in which careful attention to dimensions of possible variation (a notion derived from Marton) and ranges of permissible change afford direct access to conventional mathematical structures through placing limitations on possible generalisations. It may seem at first as if we are advocating a return to textbook exercises as an alternative to problem-solving, activity-based teaching. On the contrary, we show that the learning skills, and the quality of learners' mathematical engagement, in each kind of situation are remarkably similar.

– 349 –

Using patterns of variation to teach electro-chemical series in Hong Kong

Mun Ling Lo, The Hong Kong Institute of Education, Hong Kong

Helen Hung Hoi Ying, The Hong Kong Institute of Education, Hong Kong

This paper describes how a team of teachers and researchers in Hong Kong worked together to develop a research lesson on electro-chemical series in a Learning Study at Secondary 4 level in the subject of Chemistry. Learning study is premised on a conceptual framework that builds on three types of variation: variation in students' understanding of a specific object of learning (V1); variation in teachers' understanding and ways of handling the object (V2); and the use of patterns of variation based on the Theory of Variation (V3). The study reported in this paper illustrates how the conceptual framework, in particular, the use of patterns of variation, helps to enhance student learning. In this study, the same teacher taught the research lesson in two cycles to two Secondary 4 Chemistry classes in the same school. The two cycles of research lesson were videotaped, analyzed and triangulated with the student learning outcomes in a diagnostic test administered before and after the lesson, and a post-lesson group interview with a sample of students. Whereas the theoretical framework of variation was employed to analyze what patterns of variation were created and hence what was made possible to learn in the lessons, the results of the pre- and post- test were compared to trace the students' progress in categories of questions that measured for their understanding. In the research lesson, patterns of variation which served the functions of ìseparation and ìgeneralization were used to help the students to understand how the electro-chemical series is derived, and to discern that the order of the metals in the series and the potential difference between any two given metals are constant and independent of the reference metal used. The student learning outcomes showed that there was substantial growth in the students' understanding of these concepts.

The characteristics of critical aspects and their role in constituting patterns of variation

Anna Vikstrom, Lulea University of Technology, Sweden

The aim of this study was to analyse the characteristics of critical aspects and the role they play in constituting patterns of variation and making certain kinds of learning possible. These were identified in an empirical study where the lifecycle of angiosperms was learned. The identified critical aspects defined different ways of experiencing the life cycle. Six teachers and their students (aged 7-12) participated in the study. The relationships between the characteristics of the critical aspects, the

– 350 –

patterns of variation that constituted the space of learning and the learning outcome were investigated. Before the lesson the teachers and the researcher established the object of learning. In order to understand how the students experienced the object of learning, the students were interviewed before and after the lessons. In order to get a good picture of how both teachers and students related to the object of learning, observations and video recording of lessons were made, as well as stimulated recall interviews with both students and teachers. All recordings were transcribed in verbatim. The result shows that critical aspects can be described in terms of being specific or general with respect to the object of learning. They were identified to be more or less dependent on relations to each other and that relationship were identified as hierarchical. For some aspects it was only the relation to general aspects that made it possible to constitute patterns of variation. The critical aspects were also identified to be characterised of the presence, or absence, of a physical reference, which here refers to a link to a visible object or process. A conclusion is that, for understanding the lifecycle of angiosperms the specific critical aspects are dependent on relations to general aspects and of the patterns of variation and physical references that relation offers.

E 3 24 August 2005 14:30 - 16:30 Room E003

Symposium

Instructional Design

INSTRUCTIONAL DESIGN IMPLICATIONS OF COGNITIVE LOAD

THEORY: SEQUENCING AND DESIGNING LEARNING TASKS

Chair: Liesbeth Kester, Open University of the Netherlands, Netherlands

Organiser: Jeroen van Merrienboer, Open University of the Netherlands,

Netherlands

Discussant: Peter Gerjets, Knowledge Media Research Center, Germany

Cognitive load theory has traditionally been involved with teaching problem solving. As an alternative for conventional problems that students had to solve independently, alternative instructional methods were developed that made a heavy use of worked examples, completion assignments, process models, etc. But the backbone of instruction is still seen as a succession of ìlearning tasks that students perform. Two issues are pertinent to current research: How those tasks are sequenced and how students are supported to learn from them. With regard to sequencing, it is

– 351 –

clear that learners cannot start with very difficult tasks. This would overload their cognitive system. Therefore, tasks are sequenced in such a way that ìintrinsic load is decreased in the beginning of the learning process. Furthermore, tasks should differ from each other on dimensions that also differ in the real world. Variability helps learners to abstract away from their concrete experiences and develop general ideas that allow for transfer of learning. Mindful abstraction causes ìgermane cognitive load. Well-designed instruction must thus find a balance between decreasing intrinsic load and increasing germane load.

With regard to learner support, worked examples, problem cues, and instructional aids all aim to decrease extraneous load, which is not relevant to learning, and increase germane load. The contributions to this symposium focus on sequencing and support. Ayres presents results for an isolated-to-interacting-elements approach to sequencing; Olina et al. present results on combining cued problems with randomization as a sequencing technique for redirecting learner's attention; Van der

Meij discusses results of using worked examples with or without conceptual information; Seufert and Br?ncken present results on the effectiveness of auditory and visual instructional aids, and van Merrienboer and Kester present results on dynamic task selection, where both the task sequence and the amount of support is determined on the fly.

Interactions between the isolated-integrated elements effect and levels of expertise

Paul Ayres, University of New South Wales, Australia

Pollock, Chandler, and Sweller (2002) found that a strategy that isolates interacting elements was an effective instructional method for some learners, but not all.

This present study investigated this strategy further by studying the impact of prior knowledge and an additional technique of increased practice on key computations.

Element isolation was achieved by subdividing specific mathematical tasks requiring consecutive computations into partial tasks requiring only individual computations. Experiment 1 demonstrated that the proposed element-isolation strategy was effective. Two groups of grade 8 students were given a mathematical problem-solving task. One group solved whole problems whereas the second group solved an equivalent set of partial problems (Isolated). The isolated group made significantly fewer errors and rated cognitive load lower. In Experiment 2 three groups received different learning materials. During acquisition one group followed an isolated format, a second group followed a fully integrated format, and the third group (Mixed) switched from an isolated to an integrated format. Test problems revealed a significant interaction between learner knowledge and instructional strategy. Students

– 352 –

with poor mathematical knowledge benefited the most from receiving the isolated treatment. In contrast, students with greater mathematical knowledge learnt more by completing fully integrated problems. The mixed mode was not found to be effective. In Experiment 3, computations that caused the most errors in the domain were targeted for extended practice. During acquisition, one group received an equal amount of practice on each computation, whereas a second group received the equivalent amount of practice overall, but three times the amount on targeted computations. On test problems a significant interaction was found: students with the highest mathematical ability benefited from the targeting approach, whereas students with the least mathematical expertise did not. The two interactions identified in this study are discussed in terms of cognitive load theory.

Applying cognitive load theory to teaching comma rules to us high school students

Zane Olina, Florida State University, United States

Robert Reiser, Florida State University, United States

Xiaoxia Huang, Florida State University, United States

Jung Lim, Florida State University, United States

Sanghoon Park, Florida State University, United States

This study was an attempt to test the concepts of cognitive load theory in a real classroom setting. It investigated the effects of (a) cued versus conventional and (b) blocked versus random practice problems on learner performance and perceived mental effort on tasks involving the correct use of comma rules. The participants were more than 300 US high school students, grouped by ability level (lower-ability and higher-ability). It was anticipated that cued problems would reduce learner extraneous cognitive load, while problems presented in a random order would increase learner germane cognitive load. The treatments yielded significant differences in learner perceived mental effort, but no performance differences were identified. Among lower-ability students, conventional practice problems presented in a random sequence yielded the highest perceived mental effort ratings. In contrast, there was a trend for higher-ability students to perceive cued practice problems as more difficult than conventional problems, regardless of the problem presentation sequence. This initial study shows promise for possible applications of cognitive load theory in a real classroom setting. However, our study also raised a number of issues warranting further investigation, including: the nature and extent of cues for decreasing learner extraneous load; the nature and extent of practice and relevant prior knowledge for schema acquisition; student perceptions and experiences with the different problem types and presentation sequences; and the need for an as-

– 353 –

sessment instrument that would help make more accurate distinctions between the difficulty of a given task, student motivation to perform that task, and student perceptions of the mental effort they invest in performing it. Discussion of the results of the study and the issues it raises should help set directions for additional studies examining cognitive load theory in classroom settings.

Worked examples in instructions for end-users of software

Hans Van der Meij, University of Twente, Netherlands

Background. Worked examples are concrete instructions that display a solution for a specific problem that represents a problem type. Research has shown that worked examples can positively affect learning. Most of the research on worked examples has been done in domains such as physics and mathematics in which students work on a limited set of problems. This study explores their role in instructions for endusers of software. Two instructional variants are pitted against each other: concrete instructions versus mixed instructions. The main questions are: (a) what are the critical design characteristics of worked examples in instructions for end-users of software? (b) Do concrete and mixed designs have different effects on the user?

Method. Sixteen participants worked with the Concrete Instructions, fifteen used the Mixed Instructions. During key moments in training users answered a question about cognitive load. Knowledge and skill was tested with an immediate and delayed post-test. Results. Participants with Mixed Instructions had a significantly higher score on the most complex items from the immediate knowledge test. Participants with Concrete Instructions, on the other hand, scored better on moderately complex items from the delayed knowledge test. Conclusion. In the designs of the user instructions three structural elements have been manipulated: (a) task explanations, (b) direct instructions to act, and (c) screen shots. The findings for these variations have been equivocal. In-depth examinations of the users' reflections during training are called for to discover how users processed their instructions. Also, a further study is needed on the nature of the explanations given to the users. Such explanations should be linked to the kind of strategic reasoning one would like to model for the user.

Cognitive load and the format of instructional aids for coherence formation

Tina Seufert, Georg August University, Germany

Roland Bruencken, Georg August University, Germany

A cognitively high demanding process in understanding complex learning mate-

– 354 –

rial presented with multimedia is to build referential connections between multiple representations like text, pictures, charts or formulas. The main question addressed in the present study is, how this process of building referential connections, called coherence formation, can be fostered by means of instructional design, and which role cognitive load plays in this context. In an experimental study 58 university students learned with a multimedia program containing six different representations on a chemical domain. Understanding the whole material the six different representations had to be integrated to a coherent mental model by building multiple referential connections. The process of coherence formation was supported by instructional helps, which were presented either visually or auditory (factor 1: modality of help). Learner's cognitive capacity was assessed by a pretest (factor

2: high vs. low cognitive capacity). Knowledge acquisition was assessed by a recall and a comprehension test. With respect to recall, the results show a significant main effect of the cognitive capacity, with respect to comprehension a significant

ATI-effect between the modality and learners' cognitive capacity could be obtained, indicating that comprehension values are decreased especially for learners with low cognitive capacity and visual instructional aids. The results are in line with cognitive load theory and show that the effectiveness of instructional aids depends on the amount of additional cognitive load which it adds to the learning situation, on learners cognitive capacity and on the demands of the learning task: for a high demanding learning task low loading (auditory) instructional help seems to be most effective, while a high loading (visual) help seems to overload especially learners with low cognitive capacity.

Cognitive load theory as a basis for the dynamic selection of learning tasks

Jeroen van Merrienboer, Open University of the Netherlands, Netherlands

Liesbeth Kester, Open University of the Netherlands, Netherlands

Van Merrienboer's four-component instructional design model describes how learning tasks fulfill the role of a backbone for an integrated course or educational program. Three requirements for this backbone are: (1) learning tasks are organized in simple-to-complex task classes, (2) learners receive support and guidance for the first learning task in a task class after which support slowly disappears (scaffolding), and (3) learning tasks within the same task class show a high variability in order to allow for transfer of learning. In a flexible educational program, it should be possible to take differences between students into account. Thus, in a process of dynamic task selection new learning tasks are selected in such a way that they are adapted to the needs of individual students. Some students are better able to acquire

– 355 –

new competencies and need therefore less practice and support than other students.

In addition, elsewhere-acquired competencies of new students should be taken into account, which makes it even more important to be able to select suitable learning tasks for students at any given point in time. In the framework sketched above, this means that for each individual student, it should be possible to select the best ìtask class to work on (i.e., select tasks at the right level of difficulty) and to select a learning task with the optimal level of guidance and support within this task class.

Furthermore, this task should be selected in such a way that variability of all tasks within the same task class is guaranteed. Electronic learning environments allow for such personalization of instruction. In this presentation, recent experimental studies on dynamic task selection on the basis of the 4C/ID-model and Cognitive

Load Theory are discussed.

E 4 24 August 2005 14:30 - 16:30 Room A010

Symposium

Metacognition

CURRENT ISSUES IN THE TRAINING OF METACOGNITION

Chair: Annemie Desoete, University of Ghent, Belgium

Organiser: Eleonora Louca, Dep. of Psychology, Cyprus College, Cyprus

Isabel Garcia-Gomez, University of Seville, Spain

Csaba Csikos, Hungary Department of Education, Hungary

Annemie Desoete, Univerisity of Ghent, Belgium

Anat Zohar, Hebrew University of Jerualem, school of Education,

Israel

Discussant: Erik De Corte, University of Leuven, Belgium

The aim of the symposium current issues in the training of metacognition with as chair A. Desoete (University of Ghent, Belgium) and as discussant E. De Corte

(University of Leuven, Belgium) is to present some examples or good practices of metacognitive trainings and to discuss several current issues regarding the usefulness of those trainings.

Despite the different emphasis researchers have given to the metacognitive constructs, teachers, educators and therapists came to believe that it is worthwhile to promote metacognitive skills of students. Hartman and Sternberg (1993) summarized the literature and presented four main approaches: promoting general aware-

– 356 –

ness by modeling by teachers, improving metacognitive knowledge, improving metacognitive skills and fostering on learning environments.

In a first presentation Louca (Cyprus College) will elaborate on the construct of metacognition and on 17 teaching strategies for the promotion of metacognition.

The next presentation by Aguilera-Jimenez, Garcia-Gomez and Mora-Roche (University of Sevilla Spain) will describe teaching patterns that promote metacognition and a maximum progress in primary students. In addition Csikos (University of Szeged Hungary) will elaborate on a metacognition-based training program in grade 4 in the fields of mathematics and reading. Desoete (University of Ghent,

Belgium) will discuss a metacognition-based training program in grade 3 in the fields of mathematics. The fifth presentation by Zohar and Peled (University of Jerusalem, Israel) will be on the effects of explicit meta -strategic teaching regarding variable control on students' strategic and meta - strategic thinking".

Teaching strategies for the promotion of metacognition

Eleonora Louca, Dep. of Psychology, Cyprus College, Cyprus

Metacognition essentially means cognition about cognition. Flavell (1981) distinguishes between metacognitive knowledge and metacognitive experience. Matsaguras (1994) explains that metacognition refers to both people's awareness and control, not only of their cognitive processes, but of their emotions and motivations as well. A number of strategies are described, which teachers can use to facilitate

Children's metacognitive development and promote the monitoring and regulation of one's own cognitive enterprises. The educational implication of the application of metacognitive strategies such as self-awareness and self -monitoring, is to develop independent learners who can control their own learning and learn how to learn for life.

Teaching Patterns that promote metacognition

Antonio Aguilera-Jimenez, University of Seville, Spain

Isabel Garcia-Gomez, University of Seville, Spain

Joaquin Mora-Roche, University of Seville, Spain

Within the context of a broader research on cognitive and metacognitive enrichment (Mora, 1995, 1997) the aim of this presentation is to describe the most effective teaching patterns for achieving maximum progress in cognitive and metacognitive abilities. Three groups of Primary students who had taken part in a cognitive

– 357 –

enrichment programme, called ìComprehending and Transforming (Mora, 1991), showed different gains from pretest to posttest. It has been established that teaching interaction patterns play a main role in relation to those differences (Aguilera

& Mora, 2003, 2004) but explanations about why these teaching patterns provide such an enrichment remain unclear. We?ll give a description of those ìActivating

Teaching Patterns and discuss the place they have in the research literature summarized by Hartman and Sternberg (1993). We conclude that ìfostering on learning enviroments is the most relevant field, although not the only one.

A metacognition-based training programm in grade 4 in the fields of mathematics and reading

Csaba Csikos, University of Szeged, Hungary

The aim of this paper was to test the hypothesis whether explicit teaching of metacognitive strategies yields better achievement in various types of mathematics and reading tasks. We restrict the use of the term metacognitive to conscious and deliberate mental processes, and we use the expression metacognitive strategy in accordance with the psychological definitions of procedural metacognitive knowledge.

The experiment involved 4th grade students from 8 schools. The schools were selected from a larger set of schools participating in a project -Differential treatment and evaluation of lower SES students. 4 schools were labeled as experimental and from each school there was one experiment class chosen. The other 4 schools were labeled as control and all of their 4th grade classes were control classes. There were two pre-tests: mathematics achievement test and reading test (with document texts), and there were four post-test: the two pre-tests re-administered, a mathematics word problem test (10 parallel tasks from Verschaffel et al. 1994), and a reading test with various types of texts. The metacognition-based training consisted of 15 mathematics and 15 reading lessons that were integrated into the regular subject content. The training aimed to develop students' achievement in both strategy-level and drill-like tasks by means of developing declarative and procedural metacognitive knowledge. Our results suggest that the training had a significant positive effect on students' achievement. These findings may point to the importance of the use of metacognitive strategies in basic skill instruction.

A metacognition-based training program in grade 3 in the fields of mathematics

Annemie Desoete, Ghent University, Belgium

This contribution is devoted to the relationship between metacognition and math-

– 358 –

ematical problem solving skills in lower-elementary-school children. Within the second and third approach of Hartmans and Sternbergs (1993) model a study is presented on 84 third graders were 42 children got a training on improving the metacognitive prediction skill and 42 got a regular math course. The results of good and poor math solvers are analyzed separately.Results show that metacognition can be trained and has some value added in the intervention of young children solving mathematical problems. It should be emphasized that metacognitive skills do not develop spontaneously with a traditional mathematics instruction in regular schools. Metacognition needs to be taught explicitly in order to develop. Presentation elaborates on the improving metacognitive knowledge-approach. Within this approach a study is presented on 84 third graders were 42 children got a 5-session during training on improving the metacognitive prediction skill and 42 got a regular math course. The results of good and poor math solvers are analyzed separately.

Results show that metacognition can be trained and has some value added in the intervention of young children solving mathematical problems. It should be emphasized that metacognitive skills did not develop spontaneously with a traditional mathematics instruction in regular schools. Metacognition needed to be taught explicitly in order to develop.

Effects of explicit meta-strategic teaching regarding variable control on students' strategic and meta-strategic thinking

Anat Zohar, Hebrew University of Jerusalem, Israel

Bracha Peled, Hebrew University of Jerusalem, Israel

The study's objective is to assess the effects of explicit teaching of meta-strategic knowledge on gains in students' strategic and meta-strategic thinking and to compare between these effects for students of low and high academic ability. Participants were 41 students. Each student participated in five weekly sessions that lasted approximately 30 minutes each, in a transfer assessment session and in a retention assessment session. Only the experimental group received explicit meta- strategic instruction. Strategic ability was measured by interviews that led students in performing the simulated "experiments" with the computerized task, asking them to draw inferences and to justify them. Meta-strategic ability was measured by asking students explicit questions about the process of their investigation and about the rule of variable control, and also by asking them to evaluate and to explain experiments of fictitious students. The findings show that explicit meta-strategic teaching affects the quality of students' strategic and meta-strategic thinking about control of variables. Teaching effects were preserved for three months after instruction, and

– 359 –

across three different tasks. Teaching effects were larger for students with low academic achievements than for students with high academic achievements.

E 5 24 August 2005 14:30 - 16:30 Room E004

Symposium

Reasoning

STATISTICAL REASONING AND SITUATED MODELLING IN THE

WORKPLACE AND IN EDUCATION

Chair: Celia Hoyles, Institute of Education, University of London, United

Kingdom

Organiser: Richard Noss, Institute of Education, University of London, United

Kingdom

Arthur Bakker, Institute of Education, University of London,

United Kingdom

Discussant: Koeno Gravemeijer, Freudenthal Institute, Utrecht University,

Netherlands

This symposium will consider the changing nature of modelling and statistical reasoning in workplaces, and the implications of this for education and work-based learning. Ubiquitous technology means that sophisticated analytical techniques are no longer the preserve of highly trained professionals, but are now available to nearly all employees. With this comes an increasing expectation for employees to make use of those techniques and to be able to develop and use models of work processes. Thus there is an emerging need for employees to have functional mathematical and statistical knowledge and a ìsituated understanding of modelling that are grounded in their workplace situations and in the technological artefacts that surround them. A conventional approach to modelling is that techniques are learnt

ìout of context and then ìapplied in context, but educational research, particularly on situated cognition, has shown the limitations of that approach. By making a crossover between researchers in workplaces and researchers in school and university education an intention of this symposium is to debate the situatedness of modelling and statistical reasoning in the workplace, and the need to re-conceptualise mathematics and statistics education at the school and university levels. The symposium will embrace two themes:

(1) the forms of modelling and statistical reasoning carried out in a variety of work-

– 360 –

place settings (Hoyles, Kent, Noss & Bakker);

(2) attempts by educational researchers to improve statistical reasoning: in the workplace (Bakker, Noss, Kent & Hoyles, using simulations and dynamic statistics software), at university level (Niglas, by focussing on how students combine everyday language with statistical language to situate formal techniques in students' everyday world) and at school level (Meletiou, using dynamic statistics software).

Enhancing the teaching and learning of early statistical reasoning in elementary schools: A study in Cyprus

Maria Meletiou-Mavrotheris, Cyprus College, Cyprus

Efi Paparistodemou, University of Cyprus, Cyprus

Despina Stylianou, City College, The City University of New York, United States

The research discussed in this article comes from an ongoing multifaceted program for the teaching and learning of early statistical reasoning in Cyprus. The program has two closely interrelated aspects - design of a strong empirical research component within which to examine the development of statistical reasoning and the teaching of statistics in early grades, and design and development of instructional materials and supporting teachers' guides informed by the study's empirical findings.

The initial stage was concerned with the design of a line of instructional materials for the development of statistical reasoning. Central to this design was the functional integration with existing core curricular ideas of dynamic statistics software, which provide students with the opportunity to model and investigate real world problems of statistics. Next, professional development seminars for the teaching of statistics with the use of technology were designed and organized. We are currently at the implementation stage of the program. One of the teachers that had attended the seminars implemented the instructional materials in her sixth-grade classroom with the support of the design team, and four more teachers will do so in the upcoming months. The paper discusses insights gained from the study regarding teacher content and pedagogical knowledge of statistics and how this might impact student learning. Preliminary analyses confirm earlier findings of the research literature which indicate that teachers have a weak knowledge base in statistics. At the start of the professional development seminars, teachers exhibited a strong deterministic mindset. They consistent focused on measures of central tendency when analyzing or interpreting data distributions, and ignored variation. However, teachers' explorations during the seminars with the Dynamic Statistics Software Fathom helped them improve their conceptions of variation and distribution. They began to appreciate the uncertainty and variability inherent in statistical data.

– 361 –

Switching between everyday and statistical languages to develop students' statistical reasoning

Katrin Niglas, Tallinn Pedagogical University, Estonia, Estonia

In 1994, an action research project was initiated at Tallinn Pedagogical University

(Estonia) with the aim to work out and implement a new course in statistics for students of social sciences and pedagogy. It was considered important to develop the students' ability to understand the practical, real life meaning of statistical concepts as well as the ability to communicate statistics in two different (Estonian) languages: in so called ìstatistical language and ìeveryday language understandable also to a layman. As a result of the first two cycles of action research project the structure of the course was proposed and the preliminary results of the project reported.

Later the structure of the basic course has not changed considerably. However, the development of the teaching style that would meet the aims described above took considerably longer time. The overall attitudes towards the course over the years of the project have been mainly positive, although many students have reported difficulties in comprehension of the concepts studied. In spring term 2004 about 200 students who had completed various data analysis courses were given semi-structured survey questionnaires focussing especially on the communication aspects of the course. The paper will give an overview of the main results which tend to support strongly the approach that assumes the active involvement of students and emphasises the development of communication skills.

Situated modelling and statistical reasoning in the workplace: An analysis of Techno-mathematical Literacies

Celia Hoyles, Institute of Education, University of London, United Kingdom

Phillip Kent, Institute of Education, University of London, United Kingdom

Richard Noss, Institute of Education, University of London, United Kingdom

Arthur Bakker, Institute of Education, University of London, United Kingdom

The Techno-mathematical Literacies in the Workplace project is investigating the needs of employees in a range of industrial and commercial workplaces to have functional mathematical and statistical knowledge that is grounded in their workplace situations and in the technological artefacts that surround them. By carrying out case studies in workplaces we have identified a set of Techno-mathematical

Literacies (TmL) which turn out to be important for understanding and improving work processes. This presentation focuses on statistical reasoning and what we call

– 362 –

situated modelling, that is how employees construct knowledge about models of processes that depend on mathematical and statistical issues as situated in the workplace. For instance, situated modelling helps to make invisible causes of problems visible by using mathematical signs (tables, graphs, charts, etc.). From a semiotic perspective this implies that people need to interpret signs and know how to respond, but their actions also depend on the activity system in which they operate.

Theoretically, we are developing a framework which combines activity theory with semiotic analysis. Activity theory provides a sophisticated ìmacro-level account of how knowledge is acquired in becoming part of a workplace community, but the

ìmicro-level analysis of knowledge at individual and group levels has been less well theorised, and we are investigating how a provisional semiotic analysis (based on the formulation of Peirce) may complement the theory. We illustrate our analysis of the TmL of situated modelling with several examples: from a food manufacturing company involved in process improvement, and from a pharmaceutical company where a data collection and analysis computer system has recently been installed as a tool for continuous process improvement. In both cases, we consider a range of ìtechno-mathematical signs which mediate individual understanding and simultaneously serve as boundary objects for communication between individuals in different activity systems within a workplace.

Developing learning opportunities for techno-mathematical literacies in the workplace

Arthur Bakker, Institute of Education, University of London, United Kingdom

Richard Noss, Institute of Education, University of London, United Kingdom

Phillip Kent, Institute of Education, University of London, United Kingdom

Celia Hoyles, Institute of Education, University of London, United Kingdom

This presentation follows from the first presentation of the ìTechno-mathematical

Literacies in the Workplace project. It discusses the development of ìlearning opportunities for Techno-mathematical Literacies ó that is, flexible resources for learning that can be incorporated within, or be presented alongside, workplace training materials. Using the methodology of design experiments, we iteratively design, prototype and evaluate learning opportunities using simulations and other software tools. The learning opportunities deal with both the macro- and micro-level of work processes. On the macro-level, we use simulations with contextual pictures and video to help employees get an understanding of the key stages of the work process and understand what the key process variables are and how they interact. On the micro-level, we use educational statistics software (TinkerPlots) to focus on the

– 363 –

statistical ideas that are key to work processes. Our research suggests that above all employees have to deal with many forms of variation and to identify stable features of processes; for example, the common and special causes that generate variations in manufactured products. Thus our learning opportunities aim at developing a situated understanding of variation, target values and distributions, all in the context of production. We will present the results of using the first prototypes of learning opportunities built around the statistical aspects involved in industrial production.

The focus will be on a pharmaceutical company in which supervisory managers are learning about how to make use of a production data system and its outputs as tools for process improvement, and communicating with process data to senior managers, engineers and production operators.

E 6 24 August 2005 14:30 - 16:30 Room E102

Symposium

Moral Development

CHEATING: AN EDUCATIONAL AND MORAL CHALLENGE FOR

SCHOOLS AND UNIVERSITIES

Chair: Kari Smith, Oranim Academic College of Education, Israel

Organiser: Nava Maslovaty, Bar Ilan University, Israel

Kari Smith, Oranim Academic College of Education, Israel

Discussant: Per Lauvaas, Hoegskolen i Oestfold, Norway

This symposium is jointly organized by SIG 1 (Assessment and Evaluation) and

SIG 13 (Moral and Democratic Education). Cheating, violation of academic rules, has become a widespread phenomenon in assessment situations among learners of all ages and in various assessment contexts and it crosses cultural boundaries.

It is a moral issue all educators, parents, teachers, and faculty, as well as students, are faced with and can choose to challenge. The symposium which includes five papers, representing four countries, USA, Sweden, Israel and Australia, and with a

Norwegian discussant, presents a truly international perspective on current research into the problematic area of academic dishonesty. The context and the focus of the studies vary, one of the two American contributions enquire why high school pupils cheat even though they know it is morally wrong, and what impact this has on their moral development. The Swedish study and the second American contribution examine the role technology plays in the increase in pupils' dishonest academic

– 364 –

behaviour in their respective contexts. A different perspective has been taken in the

Israeli study which looks into teachers discourse while discussing moral dilemmas focused on cheating in an in-service setting. The Australian paper presents a more optimistic view as it examines why some learners choose not to cheate in spite of the fact they are aware of how widely spread the phenomenon is among their peers. The discussant, Norwegian Per Lauvas, will discuss the papers in light of his rich knowledge and experience as educator and researcher in the field of education and assessment in particular. This symposium contributes new information to an extremely important, yet rather neglected area of research, the meeting point of assessment and moral education.

Moral judgment and its neutralization: Why students cheat, even though they know it is wrong

Jason M. Stephens, University of Connecticut, United States

The widespread and growing nature of academic cheating is well documented in the research literature. One of the most troubling aspects of its prevalence is that many students who report cheating believe it is morally wrong or unacceptable to do so.

Stephens (2004), for example, found that nearly half of the high school students in his study who reported engaging in behaviors they considered cheating also indicated that they believed that cheating was morally wrong. Given that adolescence is critical period in identity formation (Erikson, 1968), including the development of a moral self (Damon & Hart, 1988), this incongruity between judgment and action is especially troubling. This paper present findings from field-based studies of academic cheating among middle and high school students in the United States.

This mixed-methods study (surveys and interviews) brings together insights from several theoretical frameworks - including Turiel's (1983) domain theory, Kohlberg's (Kohlberg & Candee, 1984) theory of moral development (with its emphasis on responsibility judgments as a significant moderating variable in the relation between judgment and action) and Bandura's (1986) social cognitive theory to examine the relations between students' domain and responsibility judgments related to academic cheating and their engagement in academic cheating. By building on and extending existing theoretical and methodological approaches to understanding students' judgments related academic cheating, this study aims to offer important insights into why so many students even when they think it is wrong to do so. Such insights are critical to the development and implementation of intervention strategies aimed at ameliorating the widespread problem of academic cheating during adolescence and the troubling belief-behavior incongruity often associated with it.

– 365 –

This is clearly needed as current efforts to reduce cheating are often narrow in scope

- relying primarily on the threat of punishment - and largely ineffective.

What’s so original? The discourse on education and dishonesty in the wake of a technological revolution

Lars-Erik Nilsson, Kristianstad University, Sweden

Anders Eklaf, Kristianstad University, Sweden

Torgny Ottosson, Kristianstad University, Sweden

The purpose of this paper is to discuss research interests implied by the discourse on student cheating. Earlier research on cheating has primarily been concerned with two issues. One has been to investigate (a) the extent to which students' cheat and the other issue has been (b) and student attitudes towards and reasons for cheating attempting to explain why students cheat. Today's discourse on cheating has focused on the use of technology on example being plagiarism. Proposed ways to deal with cheating also involve technology. It has been suggested that technology provides ways to analyse texts and compare to students' style of writing and to scan databases for presumed originals. This however is not the only suggested cause and remedy for cheating and plagiarism. It is also suggested that such student behaviour is caused by teaching strategies in education and that given assignments invite cheating. It is further suggested that cheating and plagiarism can be explained by changing ethics among students, implying cheating and plagiarising to be acceptable. Cutting corners through downloading papers is just another way to get the assignments done. Another suggestion is that students are unaware of what is considered proper ethics in academia and need to be informed about rules for citing and taxonomies for paraphrasing, but also what the consequences of for breaking such ethics are. We argue that this way of looking at cheating and plagiarism disregards the real problematic questions by treating academic and educational work as stable practices that are little influenced by changes in society or new tools. It disregards difficult questions about identity and what is to be considered original work or copies as new tools influence work practices.

How students, teachers, and parents judge cheating on the internet

Althea Scott Nixon, UCLA, United States

Yasmin B. Kafai, UCLA, United States

We investigated how and why different stakeholders in a school communityóstudents, teachers, and parentsójudge the appropriateness of two uses of the compu-

– 366 –

ter and Internet and predict each other's moral judgments and reasons. The first computer use we selected involves cheating with text on the Internet by violating copyright infringement laws on plagiarism: Students copy text from the Internet at home and at school in order to complete a class assignment. The second computer use involves cheating with home and school rules for browsing the Internet:

Students' browse the Internet without permission from their teachers or parents.

Results show that there were contextual and developmental differences in participants' moral judgments for the cheating with Internet text. Looking across home and school contexts, although almost all participants judged it inappropriate for students to browse the Internet at home or at school without permission, fewer participants judged it inappropriate to copy text from the Internet at home than to do so at school. Developmental differences show that fewer children than adults judged cheating with Internet text as inappropriate, and fewer children than adults were able to predict that these differences in moral judgments existed. Many participants gave social conventional reasons and moral reasons for why it was inappropriate to do these behaviors, and many participants (especially students) gave personal reasons for why it was appropriate to do these behaviors. This pattern indicates that children and adults judge uses of the computer and Internet more harshly when they are based on moral values or on socially constructed rules than when they are considered to be of personal choice, without consequences affecting other individuals.

Participants were more able to predict each other's moral reasoning for the cheating on Internet rules. We discuss several possible explanations for these findings and implications for classroom practice.

Teachers’ modes of coping with students’ cheating on exams and assignments

Nava Maslovaty, Bar Ilan University, Israel

This paper focuses on several topics: A workshop on the teacher's moral role; an assignment aimed to develop strategies for coping with cheating in school; teachers' narratives regarding cheating on exams and assignments. Research shows that 50% to 90% of students admit to having cheated on exams and assignments. Cheating stems from the desire to obtain higher grades and to advance scholastically, and occurs because school assignments are too difficult or time consuming. Teachers' strategies for coping with cheating show a relationship between the content of the dilemma and the preferred strategy for dealing with it. A need was identified for a workshop to enhance teachers' ability to cope with issues of honesty and with other dilemmas that arise in the classroom. In such a workshop for graduate students majoring in Education, participants wrote papers describing and analyzing an actual

– 367 –

dilemma they experienced, as either students or teachers. They analyzed the possible coping strategies according to two broad theories addressing school dilemmas: character education (Lickona, 1991) and the cognitive developmental approaches

(Kohlberg, 1980; Selman, 2003). Participants were also required to reflect on their learning process. Teachers who participated in the workshop described cases of cheating on exams and assignments, from the perspective of various role-players in the school - students, teachers and the principal. The events occurred on all levels: elementary, junior high and high school. All the events were characterized by pressure to succeed which ultimately led to a willingness to cheat. In these cases, the teachers employed varied strategies to cope with cheating in various situations, from ignoring them completely, through explaining to the students that what they had done was inappropriate in a one-on-one discussion with the student or in a class discussion.

Why students maintain academic honesty and do they know how?

Mark Freeman, University of Sydney, Australia

Henrikka Clarkeburn, University of Sydney, Australia

Academic honesty has become a crucial aspect in higher education and assessment not just because it has become easier to be dishonest with the plethora of information at the touch of a key via the internet (Underwood and Szabo, 2003), but because it is having a major impact on reputations of universities and senior academic managers (The Age, 2002; Newcastle Herald, 2004). This paper outlines a study among undergraduate economics and business students to investigate student motivation to maintain academic honesty and their knowledge of honest academic practices before and after a deliberate intervention to improve it. Over 300 students completed an online learning activity on academic honesty designed both to investigate their perceptions and to assist their understanding of academic honesty.

The online activity was embedded in an elective course during academic year 2004 and the completion of the modules, but not their results, contributed to the course mark. The three-part activity entailed a pre- and post survey gathering anonymous data on student perceptions and knowledge of academic honesty - how serious they perceive various types of dishonest academic practices, how common they believe them to be, and why they may or may not choose to undertake their learning in an academically dishonest manner. Sandwiched between the pre and post surveys was a series of self-managed learning modules to assist students understand honest academic practice. Various role play scenarios were used to increase engagement. The results indicate that students have strong knowledge on technical aspects of honest

– 368 –

practices, but lack the ability or motivation to judge situations and apply them appropriately. Further, the strongest motivator for students to adhere to honest practices is fear of getting caught. These results are discussed in relation to improving compliance and developing effective methods of assisting students to understand honest academic practice.

E 7 24 August 2005 14:30 - 16:30 Room E103

Symposium

Social Interaction in Learning and Instruction

MENTORING AS AN INFORMAL LEARNING EXPERIENCE

Chair: Cheruta Wertheim, Beit Berl Academic College, Israel

Organiser: Hildegard Muller-Kohlenberg, Universitat Osnabruck, Germany

Cheruta Wertheim, Beit Berl Academic College, Israel

Lea Kozminsky, Kaye College of Education, Israel

Lena Rubinstein-Reich, Malmo University, Sweden

Barbara Fresko, Beit Berl Academic College, Israel

Discussant: Hildegard Muller-Kohlenberg, Universitat Osnabruck, Germany

Barbara Fresko, Beit Berl Academic College, Israel

The purpose of the symposium is to promote an understanding of mentoring as an informal learning activity which can contribute to the social, emotional and cognitive development of both children at-risk and their mentors. Research results from studies of four mentoring schemes in Germany, Sweden, and Israel will be presented. In all cases mentoring is carried out by young adults, generally college students. The mentored children include children of immigrants, children from broken homes, children from economically disadvantaged backgrounds, and children with learning disabilities. Two studies examine the effects of mentoring on the children.

In the first study, mentoring takes place in face-to-face weekly interactions between mentors and children, and in the second study mentoring takes place online. The third study examines the professional contribution of engaging in mentoring for the mentors who are teacher training students, whereas the fourth study explores the impact of providing guidance to mentors with respect to their own development through the mentoring activity. Common features of the different mentoring schemes will be discussed, as well as theoretical and practical implications of mentoring as an educational tool. Papers to be presented:

– 369 –

1. Informal learning compensation for the lack of basic competencies and the prevention of deviant behavior - Hildegard Muller-Kohlenberg, Universitat Osnabruck,

Germany

2. The impact of online mentoring on self-efficacy perceptions of pupils with learning disabilities - Lea Kozminsky, Kaye College of Education, Beer Sheva, Israel

3. Learning within an institutional free zone effects of mentoring on student teachers' professional development - Lena Rubinstein Reich, Malmoe University, Sweden

4. The contribution of guidance to learning through mentoring

Informal learning compensation for lacking basic competencies and prevention of deviant behavior

Hildegard Muller-Kohlenberg, Universitat Osnabruck, Germany

The enormous significance of informal learning refers to the incidental, often unconscious, life-inherent learning en passant in numerous ordinary sites of every-day life. Children, who participate in mentoring programmes, live in particular circumstances that offer scarce primary experience. Mentors, therefore, are primarily challenged with the task of offering them a broader scope of child-adequate experience.

These processes of learning may be of a various nature: They could take place incidentally and unconsciously (incidental, implicit learning), or within a reflective didactical framework incorporated into every day life (informal learning). These learning processes may also include entering into contacts with youth workers or voluntary (creative) groups (non-formal learning), or consist of a particular task within the frame of homework assistance (complementary to formal learning). In general mentors in the Balu und Du Project, which will be presented, are young persons who commit themselves voluntarily to work with children in primary school.

Most programmes are located at universities where the mentor's work is accompanied by a regular course, during which the mentor-child relation is the subject of discussion and where additional information is offered from the area of pedagogy and development psychology. What do children learn from their mentors? Excerpts from regular diaries - which the mentors send weekly by e-mail -show that basic competencies improve significantly during the course of the programme. These include the following: learning how to learn, media competencies, increased level of activity, widening the area of interests, decision making, taking responsibility, setting up moral criteria and other basic abilities that facilitate the learning process at school and supporting the entrance into adult life. Evaluation results show further that core behaviour dimensions develop in a desirable way.

– 370 –

Learning within an institutional free zone effects of mentoring on student teachers’ professional development

Lena Rubinstein-Reich, Malmo University, Sweden

Considerable changes have occurred in the teaching profession including greater emphasis on social dimensions, relational competence and an ability to understand complexity and diversity. Teacher education seems to fail to prepare for this and in spite of numerous studies of teacher education there is limited knowledge of the actual process of learning to teach. Institutions like schools and universities have their own logic that influence the relationship between children and teachers, how they relate to each other and how they communicate. The relationship is often characterised by asymmetry in power, control and discipline. The impact of an institutionalised setting will be explored in this paper in the light of results from an evaluative study of student teachers mentoring school children on a personal one-to-one basis and how that contributes to their professional development. The study is based on empirical data that focus on the benefits of the mentoring program through the eyes of the mentors/student teachers and include written narratives, follow-up interviews and questionnaires. Contrary to the institutionalised setting of teacher education the relationship between child and mentor in mentoring can be defined as taking place in an institutional ìfree-zone They relate to each other as individuals. Mentor do not primarily act within the role of a professional teacher or teacher-to-be. Meetings take place in different contexts and the relationship is not fixed to a specific place as the school or after-school centre. Being involved in a mentoring project seems to provide student teachers with special experiences and understandings, which according to themselves are difficult to attain in regular teacher education. Through the example of one child, they learn about relational competence and understand complexity.

The contribution of guidance to learning through mentoring

Barbara Fresko, Beit Berl Academic College, Israel

Cheruta Wertheim, Beit Berl Academic College, Israel

Providing guidance to mentors is necessary for successful mentoring programs involving children at-risk. However, a debate exists regarding the intensity, focus, and mode of such activity. Since mentors are non-professionals, many believe they require on-going advice from specialists in order to deal with the challenges of working with disadvantaged children. Others advocate minimal guidance and encourage mentors to utilize their own personal resources in an innovative and natural way.

– 371 –

Few studies have investigated guidance for mentors. In the present study the effects of guidance on mentoring outcomes for mentors is examined. This study was conducted in the framework of the PERACH Project in Israel which deploys thousands of students yearly to work with schoolchildren from needy social backgrounds.

Minimal guidance is provided by project coordinators who are also students. Since mentors in many teacher training colleges are often provided additional guidance by professional staff members, our study focused on student mentors who were preparing to become schoolteachers. 321 mentors from 12 colleges completed the research questionnaire. Variables included satisfaction from mentoring, contribution of mentoring to professional development, and guidance framework, format, and content. One-third of the sample did not receive guidance beyond the minimum provided by the project. As for the others, most received combined individual and group guidance from either college pedagogic specialists or psychologists, while some received only group or individual guidance. Mentors were generally satisfied with their mentoring experience and reported that mentoring contributed to their professional development as future teachers. They especially learned about children, improved their ability to cope with difficult situations, and improved communication skills. Mentors who received additional guidance reported greater professional development. Guidance content and format were found to influence both contribution and satisfaction.

E 8 24 August 2005 14:30 - 16:30 Room E111

Symposium

Teacher Education

TEACHERS' LEARNING COMMUNITIES: A SOCIOCULTURAL PER-

SPECTIVE

Chair: Michal Zellermayer, Levinsky College of Education, Israel

Organiser: Michal Zellermayer, Levinsky College of Education, Israel

Synnove Matre, South Trondelag University College, Norway

Discussant: Lily Orland-Barak, Haifa University, Israel

During the last fifteen years, teacher learning has been envisioned as a collaborative enterprise for saving schools from within (Hord, 1997; Lambert et al., 1996).

This vision resulted from the emerging image of the professional teacher as one who thinks systematically about her practice as a member of a learning commu-

– 372 –

nity (Sachs, 2003; Cochran-Smith & Lytle, 1999; Day, 1999; Lieberman & Miller,

2000; Furlong, 2000). The purpose of this symposium is to highlight the usefulness of a sociocultural approach for analyzing teacher collaborative learning. The symposium includes four studies of communities of teaching practice situated in various face to face or virtual sites for teacher collaborative learning conducted in Norway, Israel and the United States. These studies focus on learning environments for teachers in which the educational goal is to advance collective knowledge in a way that supports the growth of individual knowledge (Bielaczyc & Collins 1999, Scardamalia & Bereiter 1994). In these environments, the goal of the educational practice is to foster the teachers' pedagogical content knowledge as well as to empower their sense of belonging to an educational community which may include peers, children and teacher educators. The studies identify factors that enhance or obstruct teachers' learning within such communities. More specifically, they address the following questions: What and how do pre- or inservice teachers learn in professional communities? How do they engage with their subject matter? How is their learning supported and structured by the other participants, resources, artifacts and activities present within the group? How does learning in a community strengthen teachers' sense of agency, inter-subjectivity, accountability and ownership of the classroom learning environment? How do the larger socio-cultural systems in which teachers participate influence the sense they make of their own experience as learners and its relation to their classroom practice?

Empowering preservice teachers' ownership of their learning environment: An empirical account of an online learning community

Chrystalla Mouza, University of Delaware, United States

This study has three primary objectives. First, it reports on the development of an online learning community that included pre-service and in-service teachers, school children and teacher educators. Second, it documents the ways in which the members of the community interacted with each other, with particular emphasis on the collaborative exchanges among pre-service teachers and school children. Third, it investigates pre-service teachers' learning of content and pedagogy. The learning community included 24 pre-service teachers, 2 teacher educators, 56 school children, and 5 in-service teachers. All members were affiliated with the University of Delaware. At the time of the study, the pre-service teachers participated in a study abroad trip to Ireland in which they studied Irish literature and culture as well as adolescent development. To facilitate social interaction a dynamic computer supported collaborative learning environment was developed. The environment

– 373 –

included asynchronous communication tools that facilitated sharing of experiences among the participants. Data for this study were collected from multiple sources from January to April, 2003. These sources included transcript analysis of online interactions, classroom observations of school students, online surveys administered to pre-service teachers, and interviews conducted with pre-service and in-service teachers as well as school students. Findings indicated an increased sense of ownership of the learning environment on the part of the student teachers, who essentially shaped the direction and frequency of interactions with the students as well as the nature of the discourse. Utilizing the asynchronous communication tools, school students posed questions based on topics covered in their classroom about Irish history and culture and the pre-service teachers responded. This interaction with school students and participation in a learning community enhanced the pre-service teachers' learning of Irish literature, culture and adolescent development. In addition, interaction with the tools in the environment enabled pre-service teachers to enhance their technology proficiency.

Intersubjectivity - a presupposition for participation in a community of learners?

Vivi Nilssen, Sor-Trondelag University College, Norway

This case study focuses on Sara, a cooperating teacher, and five student teachers learning to teach math in her class of third graders. Its purpose is to show how Sara facilitates the student teachers' development of pedagogical content knowledge.

Building upon the work of Matusov (2001) and Wertsch (1984) I use the term intersubjectivity to explain how Sara establishes the practice field as an arena for learning. My work is particularly related to Matusov's notion's of "community of learners" in which three aspects of intersubjectivity construct the participants' learning: having in common, coordination and human agency. The data consist of 25 audio and videotaped supervision conferences of Sara and the student teachers and 6 interviews with Sara. In these interviews, the videotaped conferences were discussed.

I also conducted one interview with each of the student teachers. Data were collected through two periods of three weeks field experience. The data analysis shows that Sara facilitates student teachers learning through two intertwined features: she designs and allows experiences and she establishes a community of learners. The study describes Sara's efforts to design the practice field as an arena for students' learning about kids' math learning. Specifically it shows how Sara creates a space for the three aspects of intersubjectivity to develop: A shared focus of attention for the student teachers, space for respectful disagreement and engagement in a caring practical action. The importance of this finding can be highlighted in the context of

– 374 –

Edwards and Collison's (1996) claim that student teachers rarely see themselves as learners in classrooms full of pupils. They seem to already know how to teach and are eager to act as competent practitioners who need to train and get feedback from the cooperating teacher, and their mentors contribute to this transmission image of learning to teach.

Accountable talk in student teachers' on-line conferencing

Dagrun Kibsgaard Sjohelle, Sor-Trondelag University College, Norway

Computer conferencing can help students examine their ideas in a social context of different perspectives and to develop collective ways of understanding. However, the students' participation mode in this kind of communicative practices is of great importance. There are certain conditions that have to be met to make collaboration on the net a successful disciplinary tool that can foster disciplinary engagement.

The purpose of this study is to focus on how the students develop accountability to others and to disciplinary norms in a learning community.

In my work, I focus on accountability in the sense of both disciplinary and social responsiveness. The research questions were

1. What participation modes do the students use in the electronic mediated cooperative working process?

2. How do they contribute to the collective knowledge construction?

The participants are two students groups in a class of 32. My main data source consists of the students' net conversations about the topic Analysis of a 9th grade student's text in the context of the Norwegian Didactics course in which I was the instructor. I initiated the activity, but the students had been given the authority to communicate without my interfering. They had to be responsive to each other in certain ways. Data consisted of two different net conversations during a period of three weeks. The discourse analysis of these conversations shows that the two groups took two different approaches to the same task. One approached the textual analysis in a harmonic and caring manner. The other adopted a more conflict-oriented participation mode, but their ìaccountable argumentation seemed to prove a positive contribution to knowledge construction. The study indicates themes that would be of interest in the field of web-mediated collaborative learning in teacher education concerning peer tutoring and peer scaffold.

– 375 –

Developing agencies in communities of practice

Edith Tabak, Levinsky College of Education, Israel

This qualitative, interpretive study focuses on a group of cooperating teachers in one elementary school in a partnership program in Israel. It aims to investigate the factors enhancing or obstructing the development of agency in a teachers' professional learning community, which included ten cooperating teachers, ten studentteachers and the clinical supervisor. The data collected during one academic year consisted of transcriptions of twenty teacher meetings at the school, participant observations by student-teachers and cooperating teachers of each other's teaching, reflective diaries written by the student-teachers and by the clinical supervisor, seven open interviews with the principal and the cooperating teachers, transcriptions of ten conferences between student-teachers, cooperating teachers and the clinical supervisor, historical documents produced by the school. The data was analyzed through the identification of themes through a dialectical process with the activity theory literature on teachers' collaborative learning in communities (Wenger, 2004;

Engestrom,1999; Edwards, 2000), discussing how the experience of agency arises out of engagement in the social world. The validation process included (a) triangulation of the various data sources, (b) collaborative interpretation of the data, (c) comparison between the data and Engestrom's model and its application to teacher learning. (Edwards, 2004). The themes identified were: (a) Historical factors constructing teachers' conceptions of learning; (b) Teachers' interpretations of their learning experiences; (c) Teachers' understandings of their membership in the community. The main finding was that the factors obstructing the teachers' development of agency were mainly related to artifacts developed in the past that had lost their relevance for the teachers' work and, at the same time, limited their activity space.

The factors enhancing their agency were related to inquiry tools that allowed them to question the validity of the "dead" artifacts and enabled them to act according to their immediate interpretation of the situation.

– 376 –

E 9 24 August 2005 14:30 - 16:30 Room A008

Symposium

Motivation

MOTIVATION AND SELF CONTROLLED LEARNING

Chair: Manfred Hofer, University of Mannheim, Germany

Organiser: Thea Peetsma, University of Amsterdam, Netherlands

Manfred Hofer, University of Mannheim, Germany

Discussant: Marold Wosnitza, University of Koblenz-Landau, Germany

In the literature on self-controlled learning it is recognized that students' capabilities to motivate themselves and to organize their learning efficiently are limited. As a consequence, this symposium pursues the aim to identify factors that determine students' investment in learning. These determinants can be regarded as resources students more or less dispose of to facilitate their study behaviour.

More specifically, four classes of motivational factors for self-regulated learning are investigated. (1) Learning with or without knowing about an alternative action opportunity describes situations entailing different motivational consequences. The experimental study by Fries et al. asks whether students' learning is deeper when alternative leisure activities are not into reach. (2) Parents educational values are potential resources in organizing self-controlled learning. Van der Veen/Peetsma ask whether the degree parents value schooling is related to self-regulation in the lowest type of secondary school. (3) Classroom environmental variables can facilitate students to seek help from teachers and class-mates. Karabenick et al. try to identify variables of classroom climate related to students' help-seeking behaviour.

(4) Perels/Schmitz evaluate experimentally a training program that promotes students' capabilities in planning and monitoring their study-behaviour.The program includes the use of a diary and is directed also to parents. Wosnitza will discuss in which ways the papers contribute to explain motivational factors in self-regulated learning and to help students in school investment.

– 377 –

Motivational interference: The impact of attractive alternatives on learning

Stefan Fries, University of Mannheim, Germany

Franziska Dietz, University of Mannheim, Germany

Sebastian Schmid, University of Mannheim, Germany

Manfred Hofer, University of Mannheim, Germany

Students constantly face situations, in which they are engaged in a school-related task (e.g. doing their homework) while attractive alternatives such as playing computer games or meeting friends are also available to them. Such alternative opportunities for action could affect motivation and performance (cf. Lens, Lacante,

Vansteenkiste, & Herrera, in press). In order to test these influences of alternative activities, an experiment was designed, in which students learned under different conditions of presence of the alternative task. The experimental setting was designed in order to mirror the typical situation of students having to learn for school in the afternoon while other activities are also present (e.g., television). Seventy-seven students (mean age: 16.9) participated in the experiment. The psychological presence of the alternative activity was systematically manipulated in the experiment.

This was done by changing the sequence in which participants had to work on the different activities (video/text vs. text/video) and by manipulating the availability of the alternative activity during learning. Students still waiting for watching music videos were expected to report more motivational interference and to have worse learning results than those first watching the videos. Furthermore, the availability of the alternative task should be detrimental for learning. The first hypothesis was supported: Students from the first condition (video/text) reported less motivational interference and had better learning results than students from the other conditions.

Availability had no effect on learning. Implications of the study for reconstructing motivation to learn within a context of competing motivational tendencies are discussed.

The development in self-regulated learning behaviour of first-year students in the lowest type of secondary school in the Netherlands

Ineke Van der Veen, University of Amsterdam, Netherlands

Thea Peetsma, University of Amsterdam, Netherlands

For decades educators have been concerned about the decline in achievement, motivated behaviour and motivational beliefs of children after the transition from primary to secondary school. This phenomenon occurs in various countries, yet, we can expect that the decline in motivation of students in the lowest school type can

– 378 –

be more extensive, as the percentage of early school leavers is highest there. Many explanations have been given for the phenomenon of the decline in motivation and achievement after the school transition. The decline has been explained by a lack of person-environment fit, i.e. poor integration of young adolescents in the school environment. Furthermore, goal orientations, shifts in the relevance of domains concerning the future time perspective, and the degree parents value schooling may be of importance. The latter has not been studied often related to the decline in motivation and may be of importance specifically for the lowest type of secondary school, as in this type of school children from low educated parents, including ethnic minorities, are overrepresented. It has frequently been found that ethnic minority parents, more than Dutch background parents, want and expect their children to attain a high level of education. Dutch working class parents have been found to be more pessimistic, or realistic. In this study we will try to explain the development of self-regulated behaviour of first year secondary school students in the lowest type of secondary school by the extent parents value education, students' well-being, goal orientations and motivational beliefs. About 730 students filled in a questionnaire. The data will be collected in three waves, the first has taken place shortly after the students started the first year (September), the next waves will take place February and May 2005. The data will be analysed with multivariate repeated measures analysis.

Help seeking and perceived classroom context

Stuart A. Karabenick, University of Michigan, United States

Akane Zusho, Fordham University, United States

Toni Kemple, University of Michigan, United States

Whether and how learners seek help remains an ongoing focus of research on the dynamic interplay between motivation and self-regulated learning. Substantial evidence indicates that seeking help when necessary can be an adaptive learning strategy that successful learners are especially likely to employ when they reach an impasse. Investigation of the contextual characteristics that facilitate, and those likely to discourage help seeking, has centered on achievement goal structures. Recent studies of elementary, middle school and college students have established that perceptions of classroom emphasis on mastery and performance goals predict help seeking and help avoidance, respectively. The present study provides further evidence regarding how students' perceptions of their classroom achievement goal structures relate to help seeking during middle and high school years. We also examined dimensions of perceived classroom context, based on Ames' TARGET

– 379 –

framework, that are believed to affect motivation and therefore characteristics that potentially influence help seeking. Results are based on data obtained from a multigrade cross-sectional study that involved the population of middle and high school students (N = 14,000+) in mathematics classes (N = 400+) in a large, ethnically and culturally diverse metropolitan area of the United States. Structured surveys assessed students' approach and avoidance help-seeking patterns, and their perceptions of classroom achievement goals (mastery, performance approach, performance avoid), academic press, teacher support for collaboration, caring, fairness and respect, and support for help seeking and questioning. Our analyses, using HLM, focus on: (a) how approach and avoidant help-seeking patterns and perceived context differ as a function of grade level; (b) relations between class context features; and foremost (c) how perceived context predicts approach and avoidant help seeking patterns.

Improving self-regulated learning

Franziska Perels, TU Darmstadt, Germany

Bernhard Schmitz, TU Darmstadt, Germany

The aim of the study is the implementation and evaluation of a training program to improve the self-regulative competence of 5th grade students. A parent training program was also implemented in order to increase the effects of this intervention.

Therefore parents were trained how to support their children?s learning behaviour outside school. Both trainings are based on a process adaptation of Zimmerman's model of self-regulation. In this article, we focus on the student's training program.

The aim of the student's training program is to enhance the self-regulative competence by specific strategies. These should support the students in choosing and applying suitable learning strategies for homework, in observing their use and in evaluating their efficiency continously in order to adjust them. The strategies are taught by means of contents of a specific subject, in our case mathematical problem solving. The students keep a diary in order to improve Children's self-reflection and self-evaluation (monitoring). In addition, an ambition of this diary was to survey the daily learning behaviour outside school. The study is based on a 2 (parent training yes/no) x 2 (student training yes/no) design with 86 students and 63 parents assigned to the training conditions. A problem solving test and a self-regulation questionnaire is applied before and after the training sessions for the purpose of evaluating the intervention. After the pretest, the training takes place in ten 90 minute training sessions. The results show a significant improvement of self-regulative competence of the students. A combination of parent and student training leads

– 380 –

to the best effects. The process evaluation of the training referring to the diary data confirms the results of the pretest-posttest comparisons. Important implications for research and practice can be drawn from these results.

E 10 24 August 2005 14:30 - 16:30 Room E113

Symposium

Science Education

COMING TO KNOW THE UNIVERSITY CULTURE OF LEARNING IN

SCIENCE AND ENGINEERING

Chair: Cendric Linder, Uppsala University, Sweden

Organiser: Ake Ingerman, Chalmers University of Technology, Sweden

Discussant: Caroline Baillie, Faculty of Applied Science, Queen's University,

Canada

Students on the brink of studying science and engineering are at the gateway of new ways of understanding and interacting with the natural (and technological) world. However, their understanding of the idea of the subject and domain does not necessarily recognize the fundamentals and underpinnings (culture) as understood by their teachers. In this symposium, we will use different perspectives to explore how students close to the gateway (before and after starting university) understand such underpinnings of science and engineering. We will also explore possible ways of developing this understanding in these students. Tom Adawi and Cedric Linder describe the different ways of understanding heat and temperature for mature adults previously unschooled in university physics, illuminating their conflicts between everyday thought and scientific thought and its resolution.

Brandon Reed, Jenni Case, Ake Ingerman and Cedric Linder present different ways of understanding technology, a central product and process of engineering, among pre-university students. Camilla Rump and Lars Ulriksen have investigated first year physics university students' experience of the ìface of physics, and how this relates to their view of the nature of knowledge in physics. Rebecca Lippman Kung,

Anna Danielsson and Cedric Linder discuss a method for assessing students' use of metacognition in the introductory physics laboratory, and study how different types of laboratories affect their metacognition. John Airey and Cedric Linder present an analytical model drawing on multiple theoretical perspectives to better understand the links between learning and the discourse of university science for first year

– 381 –

students.

Metacognition in the student laboratory: Is increased metacognition necessarily better?

Rebecca Lippmann Kung, Uppsala University, Sweden

Anna Danielsson, Uppsala University, Sweden

Cedric Linder, Uppsala University, Sweden

In this study metacognition during the student laboratory is explored, with an aim towards quantizing the amount of metacognition used by the students and evaluating how the laboratory setup encourages the use of metacognition. Six different groups of university students were analyzed using videotapes of their behaviour during three types of introductory physics laboratories: the typical ìcookbook laboratory, the ìcookbook laboratory with added ìexplain your reasoning questions and an open-ended laboratory where students must design an experiment to answer a question. Videotaped data was transcribed and then coded with respect to the general behaviour of the group, including off-task, logistical (data taking, report writing, etc.) and sense-making behaviour (discussions about the experimental design, physics concepts, physics formulas, their data, etc.). Any verbal comments judged metacognitive were marked. Within the studied laboratories, there does not appear to be much difference in the amount of metacognition. However, there is a difference in the result of the metacognition. Students in the ìcookbook laboratory frequently evaluated their own understanding, usually in the negative sense: ìI dont get this. They then continued reading the manual or collecting data. Students in the other laboratory types were more likely to change their behaviour as a result of a metacognitive statement. For example, students might move from taking data

(logistical) to discussing how to understand a certain result (sense-making). While the students in general found it easy to evaluate their understanding, students in the typical ìcookbook laboratory were less likely to change their behaviour as a result, perhaps because they did not view such action as possible, allowed, or approved of in the situation. These results indicate that it is important to consider the outcome of metacognition, and that the laboratory setup can help encourage students to change their behaviour as a result of their metacognition.

– 382 –

What’s hot and what’s not: a phenomenographic study of lay adults’ conceptions of heat and temperature

Tom W. Adawi, Chalmers University of Technology, Sweden

Cedric Linder, Uppsala University, Sweden

This paper reports on a phenomenographic study investigating how lay adults describe heat and temperature in a selection of everyday situations involving thermal phenomena. Two groups of adults participated in the study ó some with only basic schooling in physics and some who were taking an introductory survey course in physics for the general public, but without any prior experience of physics at the university level. Data were collected through interviews (N=10) and written comments to a homework assignment dealing broadly with the topics of heat and temperature, collected from the whole group attending the course (N=60). The data were analysed using a phenomenographic approach, aiming at identifying and describing the qualitatively different ways in which heat and temperature are described. Five conceptions of heat and temperature were discerned from the data. These different ways of describing heat and temperature, and the logical relations established within and between them, constitute the main results of the study. The conceptions were also analysed in terms of internal consistency. The results indicate that heat is always described as something that is contained in objects and sometimes also described as a kind of substance. Temperature is often described as a measure of the amount of heat in an object and sometimes also described as a property of the material from which an object is made, and thus the idea of thermal equilibrium is not always appreciated. This paper examines these findings and discusses their wider implications for teaching and learning thermodynamics. In particular, the study illuminated conflict threads, not previously described in the literature, between everyday thought and scientific thought, and also patterns of the resolution.

Learners’ conceptions of technology

Brandon Reed, University of Cape Town, South Africa

Jenni Case, University of Cape Town, South Africa

Ake Ingerman, Chalmers University of Technology, Sweden

Cedric Linder, Uppsala University, Sweden

This paper reports on the outcome of an investigation into learners' conceptions of technology. The results are based on interviews with learners in their final two years of schooling in and around Cape Town, South Africa. Learners need to be technologically literate and have a firm grasp of the nature of technology to be

– 383 –

best equipped to make worthwhile contributions in this increasingly technological world. Lately, these learners might not be getting this through the direct experience of tinkering and engaging with technologies as they may have in the past. Further, it is important to illuminate questions about what they conceive technology to actually be - the focus of this paper - and not only how they gain technological knowledge. An analysis of literature relating to the description of the ìmeaning of technology shows it to be defined variously as,technology having a social aspect, technology as applied science, technology as the precursor to scientific theory, a product centred view of technology, a process centred view of technology. Each learner in the study was given a disposable camera and asked to take photographs in the days leading up to the interview of anything that they considered to be technology. Learners were interviewed about the photographs that they had taken and how these related to their conception of technology. This study was undertaken using a phenomenographic approach and resulted in a series of logically related categories of description that describe the variation in the learners' conceptions of technology at the collective level. At this stage the preliminary findings bear some useful relation to the categories above identified in the literature. The findings contribute to the process of understanding how technological knowledge may be gained and point the way to increasing learners' contribution to the increasingly technological world.

Looking for links between learning and the discursive practices of university science

John Airey, University College Kalmar, Sweden

Cedric Linder, Uppsala University, Sweden

The study presented here follows a group of Swedish physics undergraduates and their experiences of learning university physics. Students attend lectures in both

English and Swedish after which semi-structured interviews using stimulated recall are conducted focusing on the concepts taught and the discourse encountered. The student experiences are compared to the intended learning objectives as detailed by the course lecturers, and cross-referenced with the language and practices observed in the lectures. The aim of the presentation is to examine the feasibility and fruitfulness of characterising student learning as an interplay between experiential and discursive components. The paper discusses the extent to which scientific concepts may be interconnected with our discourses about them, suggesting a learning trajectory starting with discourse imitation which leads to concept experience. In this characterisation student discourse is initially a poor imitation of specialist dis-

– 384 –

course, but this gradually spirals towards something which approximates legitimate specialist discourse about the concept. Similarly, the initial student experience of the concept behind the discourse will be both tentative and fragile. As students become more familiar with the specialist discourse they experience more and more facets of the concept which were previously inaccessible to them. Although preliminary results will be available for presentation at the conference, the main purpose of the presentation is to contribute to the understanding of the links between learning and the discursive practices of university science.

Should physics be fun?-What physics curricula expect of students and students expect of physics

Camilla Rump, University of Copenhagen, Denmark

Lars Ulriksen, Roskilde University, Denmark

There is a general agreement among physicist, that physics is a particular way of thinking; a conception of the world. This view of physics could be termed ìthe true face of physics. This may, however, not at all be the face students see in the first year of studies. As it is, first year teaching is not primarily planned to show students what physics is about. Rather, the intention is to equip students with the necessary preconceptions and skills for practising physics - later on. An analysis of the first year physics major curriculum at the University of Copenhagen has lead to the concept of ìthe implied student, analogous to ìthe implied reader of a novel. The

ìimplied student is interested in the general and abstract, rather than the concrete and is a hard-working person with a strong ability to discipline one self to pursuing understanding even if the teaching calls for surface learning. This implied student differs from the actual empirical student, e.g. regarding conceptions of what physics is about. This may be part of the explanation for rather significant drop-out rates.

Initiated by an educational reform, a new initial physics course in Newtonian mechanics has been designed. It was the hope that this course would show a more ìtrue face of physics to the students, and give room for students to choose their own line of learning and pursue their own interests. Initial results indicate that this has been achieved to a certain degree, but there is still room for improvement, and furthermore that the interplay between students expectations, preconceptions of physics and the teaching activities is a rather complicated matter. The paper will discuss the causes and consequences of the results and give directions for future improvements in this course and in general.

– 385 –

E 11 24 August 2005 14:30 - 16:30 Room E009

Symposium

Writing

TEXT ANALYSIS IN WRITING RESEARCH: APPROACHES AND METH-

ODOLOGICAL QUESTIONS

Chair: Pietro Boscolo, University of Padua, Italy

Organiser: Pietro Boscolo, University of Padua, Italy

David Galbraith, University of Staffordshire, United Kingdom

Discussant: Gert Rijlaarsdam, University of Amsterdam, Netherlands

The theme of this symposium, text analysis, is central to writing research. In the

70s and 80s the cognitive approach to writing research made extensive use of text analysis to infer processes (particularly regarding planning and revising) on the basis of indexes such as cohesion and text organization. The mid 1980s saw a ìsocial turn in writing research as the field was influenced increasingly by social theories.

Social perspectives were set in opposition to cognitive, and many researchers began emphasizing the social aspects of the process over the cognitive. Attention went to collaborative aspects of writing, and recordings were made of these kinds of social interactions. In the 1990s the emphasis shifted to larger social matters, and to writing that is embedded in complex community practices. Text analyses are used to understand such practices and the larger context that surrounds them. The aim of this symposium is to highlight some methodological problems implied by the use of text indexes, and to stimulate a discussion on the role of this analysis in writing instruction. Whereas the scientific relevance of this topic needs no emphasis, its educational relevance mostly regards writing assessment. Although teachers of language skills use holistic measures for evaluating writing, we think that increasing their knowledge of text analysis may help them apply more adequate assessment tools. Nelson presents major approaches to text analysis in writing research.

Spelman Miller relates pausing activity to the topic content of the text, and presents a scheme for the identification of production units. Boscolo and De Marco compare the written syntheses of students of two communities of discourse through several indexes of text comprehension and production. Galbraith, Hallam and Torrance use content analysis to evaluate the effects of different drafting strategies on idea generation during writing.

– 386 –

Text analysis in writing research: different questions, different approaches

Nancy Nelson, Lousiana State University, United States

What is the nature of the text? What is the nature of the process? What is the nature of the social process? and What is the nature of the practices? Major developments in the field of writing research can be seen in the sorts of text analyses used to answer these questions. During the 1970s and 1980s one approach to understanding the cognitive aspects of the writing process focused on the textual product itself.

Guided by propositional models of meaning-making, researchers parsed texts into ordered propositions, and they made inferences about mental products by studying written products. Another kind of cognitive process research, conducted concurrently by different researchers, focused on component subprocesses of writing, such as planning and revising. The tapes of their writing sessions were turned into typed protocols, which became texts for analysis by the researchers. The mid 1980s saw a ìsocial turn in writing research as the field was influenced increasingly by social theories. Social perspectives were set in opposition to cognitive, and many researchers began emphasizing the social aspects of the process over the cognitive. Attention went to collaborative aspects of writing, such as peer response, and recordings were made of these kinds of social interactions. Transcripts of the exchanges became the texts researchers analyzed to gain insights. In the 1990s the emphasis shifted to larger social matters that are historical, cultural, and ideological, and to writing that is embedded in complex community practices. Text analyses are used to understand such practices and the larger context that surrounds them.

In conclusion, the point is made that questions and approaches do not have a simple unidirectional relationship. Sometimes the questions lead to particular research approaches, but it is also the case that questions can follow from the approaches available to the researcher.

Production units in writing: a discourse perspective

Kristyan Spelman Miller, University of Reading, United Kingdom

Within cognitive research on writing little attention is generally given to the nature of the text produced, and to how the language formulated relates to the processes of production. This paper argues for the need to focus attention in writing research on textual output, and in particular on aspects of the text's internal structure. My current research, presented in this paper, sets out to define units of production in writing from a discourse perspective. Working with data elicited through keystroke recording, I have developed a scheme for the identification of topic-related units of

– 387 –

production as a means of describing and measuring textual output. Complementing traditional (grammatical) characterisations, this approach focuses on the association between text structure and the processes involved in handling topic in written text. It establishes a number of units, so-called framing devices' (Spelman Miller, forthcoming), which reflect the potential discourse role of units of language in introducing, maintaining and developing topic in the discourse. These units appear to be useful in describing and explaining both planning and revision behaviour in a study of L1 and L2 writers. Significant pausing appears to coincide with the production of these topic-related units in a way which offers support for the notion of framing device as a means of interpreting the language produced on-line. Framing devices also prove useful in describing the writers' revising processes, especially where revisions occur at the point of inscription. Such revisions often coincide with critical points in the development of the discourse, which can be captured by the notion of the framing device. By illustrating the interactions between framing device, pausing and revising, this paper argues for the centrality of textual analysis, particularly addressing notions of topic and discourse development, in the discussion of written language production.

Text analysis in writing from sources: a comparison between Psychology and Architecture students' written syntheses

Pietro Boscolo, University of Padua, Italy

Barbara De Marco, University of Padua, Italy

When writing from sources readers have to take information from various sources on a same topic, and synthesize it in a new text (Spivey, 1984). From a socioconstructivist perspective, Spivey has called discourse synthesis the process writers are engaged in when they read multiple texts to produce a new and abridged one. In this task three main operations are involved: 1. selecting the information to be included in the synthesis; 2. organizing information in the text to be written; 3. connecting the information taken from various sources to obtain a cohesive text. Discourse synthesis is both comprehending and composing, as a writer uses cues from more than one text to construct meaning for the text being written (Nelson, 2001; Spivey, 1997). tudying written synthesis requires careful text analysis, since the basic synthesis operations have to be operationalized into indexes. In the present study, the written syntheses of Psychology and Architecture university students were analysed. The two faculties represent two different communities with different discourse practices, more discursive in the case of Psychology, more relying on graphic supports in the case of Architecture. Participants were 128 Archi-

– 388 –

tecture and 116 Psychology students. They were given three texts on the problem of old-aged housing, a relevant topic from both a psychological and architectural perspective. After reading the three texts, participants were asked to write a synthesis of the three texts. The written syntheses were analysed through indexes of text comprehension (informativeness, units of information, integration) and production

(cohesion and text organization). From multivariate analyses of covariance (MAN-

COVA), a principal effect of the Faculty on all the comprehension indexes and on text organization emerged.

Effects of different drafting strategies on idea generation during writing

David Galbraith, University of Staffordshire, United Kingdom

Jenny Hallam, University of Staffordshire, United Kingdom

Mark Torrance, University of Staffordshire, United Kingdom

Surveys of writers' drafting strategies suggest a contrast between two different approaches to writing. On the one hand there is a planning approach, in which writers concentrate on working out what they want to say before writing, and only start to produce full text once they have worked out what they want to say. On the other hand there is an interactive approach, in which writers work out what they want to say in the course of writing and in which content evolves over a series of drafts.

Previous research comparing the effectiveness of different drafting strategies has overwhelmingly favoured planning strategies over interactive strategies. This paper describes an experiment in which three variables were manipulated- degree of organisation required in initial draft, mode of writing in first draft, and form of rewriting employed in final draft - to create various forms of drafting strategy. The effects of these manipulations on idea generation and on the quality of the final text were assessed. Analysis of the texts produced in the initial and final drafts shows clear quantitative and qualitative differences in the amount and type of content generated in different conditions. In particular, the distinctive feature of outline planning is that it enables writers to formulate rhetorical goals for the text, rather than facilitating the generation of content. By contrast, unorganised initial drafts appear to enable writers to clarify their personal opinion of the topic. These initial differences are transformed into the final text in different ways depending on the form of rewriting employed and individual differences between writers (self-monitoring).

The effect of these differences in the way ideas are generated on the quality of the final text will be assessed, and the results will be discussed in terms of different methods of reconciling dispositional and rhetorical goals in writing.

– 389 –

E 12 24 August 2005 14:30 - 16:30 Room A107

Symposium

Computer-supported Learning Environments

TEACHING AND LEARNING IN DIFFERENT LEARNING COMPUTER

SUPPORT SCENARIOS

Chair: Wim Jochems, Open University of the Netherlands, Netherlands

Organiser: Anna Espasa, Open University of Catalonia, Spain

Elena Barbera, Open University of Catalonia, Spain

Discussant: Jan Elen, University of Leuven, Belgium

In the network society, in a range of educational situations that use ICT, there is a trend more focused in technological approaches than in educational ones due to the characteristics and qualities of the technology. From our point of view, the

ICT applied to the education field should enhance teaching and learning processes instead of being considered only a technological tool. From these assumptions, the symposium will go beyond the ICT as a medium and will be focus in knowledge construction processes, making conclusions about similarities and differences in various teaching and learning scenarios carried out with ICT support, in order to have evidences about mechanism of progressively understanding of cultural meanings. Our purpose is to analyse knowledge construction in four learning computer support settings: a multimedia self learning environment; a case of blended learning that combine face-to-face learning with on line learning; scenario carried out in a virtual environment, and teaching-learning process supported with ICT.

The presentations that conforms the symposium are:

Knowledge construction in a multimedia self-learning environment: a methodological approach (Teresa Guasch, Anna Espasa, Antoni Badia and Elena Barbera.

Open University of Catalonia).

Elicitation support for knowledge construction in distributed learning. ( Wim Jochems and Bitter Rijkema. Educational Technology Expertise Center (OTEC).

Open University of Netherlands).

Knowledge construction in a university context of blended learning (Madelen Holdmund. Centre for Educational Technology. Umea University and Anna Nordstrom,

Centre for Regional Science).

Teachers and ICT as mediators of the students' knowledge construction in a face-toface environments (Maria Jose Rochera, Rosa Colomina and Teresa Mauri. Depart-

– 390 –

ment of Development and Educational Psychology. University of Barcelona).

The proposal has an important scientific and educational relevance because of the methodological approaches that we will be discussed, and the conceptualization of the technology as a learning tool.

Elicitation support for knowledge construction in distributed learning

Wim Jochems, Open University of the Netherlands, Netherlands

Marlies Bitter-Rijpkema, Open University of the Netherlands, Netherlands

Collaboration in multi-disciplinary, multi-expertise teams is considered nowadays to be an important aspect of professional learning. However, effective knowledge exchange in computer-mediated groups appears to be problematic, especially the exchange of implicit knowledge. Tools to support knowledge elicitation are more and more used in situations where employees or students collaborate using the computer. Studies indicate that there exist differences between experts and novices regarding their methods of work and reasoning. The commonly preferred approach deals with team members as a single system with ìcommon , shared preferences.

The question is to what extent this approach is optimal. From literature potential difficulties with uniform knowledge elicitation support for teamwork can be derived.

We carried out a study to investigate whether support tools for knowledge elicitation should explicitly take into account the expertness of team members. In this study we gathered qualitative insights into user-elicitation preferences, especially in relation to a user's proficiency in the field. Subjects of this study were graduate students studying for a profession as social worker, who had to design in a distributed setting a collective intervention plan to address school absenteeism in a metropolitan city.

Based on literature we assumed that experts would follow their own line of work while novices would use the procedures and prescriptions available. With respect to their preferred work style students with an expert-like or novice-like profile didnt differ significantly. Further interpretation of the results seem to indicate that not so much the proficiency of the team members as well as the attunement with the surrounding context is critical for the effect of elicitation support.

Knowledge construction in a university context of blended learning

Madelen Holmlund, Centre for Educational Technology, Umea University,

Sweden

Anna Nordstrom, Centre for Regional Science University of Umea, Sweden

This is a case study of how knowledge is constructed in a university context when

– 391 –

students are using both virtual (online) and real (study centres) learning environments, blended learning. The study is based on a three-year study programme in pharmacy/pharmacology. The students are generally older than campus students and have a different social structure, living in small population centres with their own families. The students also generally lacked experience in university studies and especially in online studies. The learning methodology is based on a social constructivistic approach. Students combine online learning with physical meetings, bringing the socialization from the physical meetings into social structures in the virtual classroom. In order to make this approach successful, students need dedicated time and assignments for socialization. Assessments and exercises are designed in order to encourage and develop the use of different communication tools and help students to create their own knowledge using each other and interacting with each other. Due to the fact that most students have little experience in using computers and internet in university studies the need of a thorough introductory course is necessary in order to train the students in using tools and technology as well as introducing them into a constructivistic way of learning using communication and collaboration.

Teachers and ICT as mediators of the students' knowledge construction in a faceto-face environment

Maria Jose Rochera, University of Barcelona, Spain

Rosa Colomina, University of Barcelona, Spain

Teresa Mauri, University of Barcelona, Spain

The goal of this communication is to present the analysis of the educational guidance that the teacher provides in order to orient the students' learning process in a face-to-face (F2F) educational environment in which ICT are used. We depart from a socioconstructivist perspective of teaching and learning. The analysis focuses on the joint activity of teachers and students, in which both tasks with and without

ICT are carried out (Derry et al. 2000; Twining, 2002). We have analysed the uses of ICT during one instructional process following a qualitative methodology.Several sessions of a primary 6th grade group of students (12-y-olds) and their female teacher working on the topic scientific method were analysed (see note below).

Results show different types and levels of educational guidance related with the uses of ICT. These uses stand, at the same time, in relation (1) with the forms of organisation of the joint activity that teacher and students develop together, (2) with the changes in the social organisation of the class and (3) with the different contents that are dealt with thanks to ICT. Furthermore, results point out that this educational

– 392 –

guidance also depends on the instructional function of the teaching and learning activity itself. As a whole, this study contributes to an improved understanding of the processes through which the teachers might help the students learn with ICT.

We underline the importance of looking at (1) the articulation of the forms of joint activity with and without ICT, and (2) the features of the joint activity in respect of by whom, how, for what purpose and with whom ICT shall be used or not, so that both ICT and the teachers' educational guidance contribute to the students' knowledge construction. Note: The project's site can be visited at www.edebedigital.com/ proyectos/228/

Knowledge construction in a multimedia self-learning environment: a methodological approach

Teresa Guasch, Open University of Catalonia, Spain

Anna Espasa, Open University of Catalonia, Spain

Elena Barbera, Open University of Catalonia, Spain

Antoni Badia, Open University of Catalonia, Spain

The aim of this paper is to present a methodological proposal that allows for a closer look at the analysis of knowledge construction in a self-learning process in a multimedia learning environment that explains how transfer of control takes place and how students progressively understand the learning content. To be able to meet the objective set, we have based our studies on the analysis of the interactivity segments in various self-learning sessions in a multimedia environment. The results presented refer to the characterisation of the interactivity segments identified in a multimedia self-learning environment. These interactivity segments correspond to: interactivity segments for organisation and management of the activities, interactivity segments for carrying out exercises with the multimedia educational material, interactivity segments for reviewing and practice with the multimedia educational material and interactivity segments for practice with the material on paper. The interactivity segments described above have allowed us to elaborate a map of interactivity in which templates have been characterised for the various approaches to learning. A chain of interactivity segments can be seen in terms of these results, which are illustrated by the examples of three students who approached the studying of the multimedia material in two different ways: a) a sequential approach: careful following of the design of the multimedia self-learning material; b) a repetitive approach: repetition of the exercises, once completed.

– 393 –

E 13 24 August 2005 14:30 - 16:30 Room E104

Symposium

Emotion

ANALYZING EMOTIONS IN THE CLASSROOM

Chair: Michaela Glaeser-Zikuda, University of Ludwigsburg, Germany

Tina Hascher, University of Bern, Switzerland

Organiser: Michaela Glaeser-Zikuda, University of Ludwigsburg, Germany

Tina Hascher, University of Bern, Switzerland

Discussant: Sanna Jarvela, Univerity of Oulu, Finland

Andreas Krapp, University of the Bundeswehr, Munich, Germany

Students' emotions in the school context have become an important topic in the view of educators and researchers. Several empirical studies point to the key-role emotional factors play as constituting elements of the learning process, next to and in close interaction with motivational and cognitive factors. Emotions have been found to correlate with the social context, teachers' didactical competencies, students' self-regulated learning, and their academic achievement. Despite the increasing amount of empirical studies on emotions, there is a lack of adequate research methods to show whether and how emotions interact with social and contextual factors in school and in the classroom. Multiple data sources and mixed, qualitative and quantitative, as well as process-oriented methods can be helpful in approaching this aim. The purpose of this symposium is to point out the importance of emotions for learning processes in school and to present different ways of analyzing emotions in the school context. The authors will describe different emotions related to students, teachers, instructional quality, and the school context. Furthermore, the presentations will focus on different instruments applied, such as questionnaires, semi-structured interviews, stimulated recall, diary logs, video-observation and event-sampling method. Especially, the aim of the symposium is to discuss the contribution of multiple research perspectives for the analysis and understanding of the structure of emotions as well as their relation to the learning environment in school.

– 394 –

Analyzing the relation between well-being and emotions in school

Tina Hascher, University of Bern, Switzerland

Emotions are like cognitive aspects a crucial part of learning and achievement situations in school. Most studies on students' emotions have to conclude that there is an overload of negative emotions (cf. Hascher, 2004; Pekrun et al., 2002). Longitudinal and cross-sectional research could show that positive emotions already decrease during the first years of school (Helmke, 1993) but may rise again at the end of the school career (Eder, 2004). School, however, is not only a context for the development of negative emotions. Asked explicitly for their well-being in school, students' answers are far more positive. Many students at least report not to feel bad in school

(Fend & Sandmeier, 2004). How does well-being differ from the experienced emotions in school? How do students' daily emotions contribute to their well-being?

What are the sources of well-being and emotions in school? The presentation will focus on results of a quantitative cross-sectional study using a well-being questionnaire for adolescent students and on results of a longitudinal study using emotion diaries for students. 391 students, grade 7-9, participated in the quantitative study;

62 students additionally wrote daily dairies over a period of six weeks. A differentiation between the valence of emotions was made and the subjective relevance of the situations was taken into account to highlight the relation between well-being and emotions in school. The dominance of negative, intense emotions was obvious.

Students who reported more negative emotions in their school life did not differ in terms of their well-being from students who reported a dominance of positive emotions. In accordance with the sources of well-being on of the main causes for negative emotions lay in teachers' behaviour (like care for students vs. achievement pressure). Both, the quantitative and the qualitative approach, are introduced and discussed with respect to their power of explanation.

Emotions in learning situations - a mixed method approach

Michaela Glaeser-Zikuda, University of Ludwigsburg, Germany

A variety of studies demonstrate that students' emotions in classroom and school have become important topics in educational and psychological research. For a long time, a sharp distinction between emotional and cognitive dimensions of learning characterized research in this area. Nowadays, learning is considered to be an interactive and dynamic conceptualization of person-environment relationships and intra-personal functions including self-regulation of cognitive, motivational and emotional factors. Therefore, it is a specific challenge to conduct research in this

– 395 –

field. A study is presented focusing on students' emotions in varying learning situations in the classroom. The relation of students' positive and negative emotions to the learning environment in the classroom was analyzed, regarding social relations between students - students and students - teacher, teacher behavior, learning topic, perceived self-regulation and competence in learning. Based on a mixed methods design, questionnaires as well as semi-structured interviews, diary logs and videoobservation were applied. In total, 968 students from 36 classrooms of secondary school level participated in this study, 68 of them were involved in the qualitative part of the study to analyze emotions in a holistic way. The presentation will focus on students' emotions in different school lessons, regarding different learning situations and subjects (German Language, Biology, and Physics). Examples will be given to illustrate the research potential of a combination of quantitative and qualitative instruments and data to understand students' emotions in classroom.

Humor as an emotional facilitator to learning processes

Leo Guertler, University of Tuebingen, Germany

Research in education strengthens its effort to understand social aspects of learning processes. Humor becomes an important aspect of the social climate in the classroom and touches terms like quality of teaching. But humor varies according to personality and sex. Already in elementary school, girls use different kinds of humor than boys (Bonsch-Kauke, 2003). On the other hand, teacher's burden is to establish a socially warmth atmosphere - a very difficult balance to keep. But humor can act as a catalyst in school. The study reconstructs student's subjective theories on humor within the methodological framework of Groeben & Scheele (1977). It aims to help to fill a gap in literature as student's thinking is lacking there. The study used nine open questions to explore the subjective definition of humor, action sequences, visions and various scenarios as well as negative experiences. The questionnaire was given to N = 363 German students in 4 different school-types of German Realschule (m = 203, f = 114) and Gymnasium (m = 18, f = 28). The analysis (qualitative, quantitative) searches for prototypical answers for the groups sex and school type. Results replicate sex * school differences not only in word production, but also within the answer structures. Of course, humor plays an important part. Most of all, humor is necessary to create an atmosphere free from stress, pressure, and anxiety. Humor fosters the emotional and social aspects of learning and instruction.

Regarding sex differences, e.g. boys are more cool and girls are more empathically, but also more critically. On the institutional part, a lack of humor in daily life at school is predominant. But this must not be attributed solely to the teachers. It is a

– 396 –

general problem in which all participants (teachers, students, colleagues, politics, parents) play some part.

Stability and variability of emotional experiences in the classroom

Klaus-Peter Wild, University of Regensburg, Germany

The variability of emotions in short periods of time and the considerable difficulties to recall those emotional variations contribute to significant assessment problems.

Other problems of assessing emotions in the classroom are connected to the potential of negative repercussions of our assessment procedures to the flow of actions and the flow of emotional experiences in an educational setting. In the last two decades numerous methods to deal with these difficulties were developed. In our study we worked on new ways to apply the Experience Sampling Method (ESM) to classroom research in vocational education. We used a programmable pocket calculator (like a PDA) as signalling and reception device. The calculator was programmed to give a signal at random intervals. When being signalled the students enter information about their momentary situation and psychological state according to the self-report questionnaire in the device. In our study the ESM was used to assess subjective experiences related to emotions, motivation to learn, effort and attention. The sample comprised 117 students from 13 classes from a vocational school (insurance business). Each student had to respond to a beeper five times a day for a period of a week. This procedure was performed twice a year over the complete period of vocational education (3 years). This presentation will focus on two topics: First, we discuss some important methodological issues which emerge with the analysis of time dependent data coming from field studies depending partly on random time sampling procedures. Especially, questions that deal with the influence of the learning environment on the emotional ex