Harsh or Inept Parenting Avh_Vilma_040208.indd 1 08-02-06 11.40.59

Harsh or Inept Parenting Avh_Vilma_040208.indd   1 08-02-06   11.40.59
Harsh or Inept Parenting
Örebro Studies in Psychology 13
Vilmante Pakalniskiene
Harsh or Inept Parenting,
Youth Characteristics and Later Adjustment
© Vilmante Pakalniskiene, 2008
Title: Harsh or Inept Parenting, Youth Charcteristics
and Later Adjustment
Publisher: Örebro universitet 2008
www.oru.se
Editor: Heinz Merten
[email protected]
Printer: Intellecta DocuSys, V Frölunda 2/2008
issn 1651-1328
isbn 978-91-7668-587-7
Abstract
Despite most parents’ good intentions to provide a warm, supportive environment in
which the child can grow and develop socially appropriate behavior, they might
occasionally act toward their child in a negative or even harsh way. Some do this more
consistently than others. This dissertation examined the relationships between harsh or
inept parenting and children’s characteristics in predicting various adjustment problems.
The first aim of the dissertation was to examine if experienced harsh parental behavior is
associated with adjustment problems for children from different cultures in a similar way.
Study I showed that the effects of harsh parenting were very similar for children from
different countries, but the magnitude of these effects differed. The second aim was to
examine how parents and youths respond to each other over time. Studies II and III showed
that youth characteristics influenced harsh or inept parenting and, to a lesser extent,
parents’ behaviors could affect youth characteristics or behavior problems. The third aim of
this dissertation concerns the role of child or youth characteristics in the link between harsh
parenting and adjustment problems. Findings from Study II suggested that, youth
characteristics might be responsible for both harsh parenting and problematic peer
relationships, thus explaining the link between them. Studies IV and V showed that
children’s early unmanageability increased the risk of having more adjustment problems
later in life only for some children. The fourth aim was to examine how the early
characteristics of children who experience physical punishment in the context of parenting
behaviors that communicate negative emotions affect later adjustment. The findings from
Studies IV and V suggest that only for some children, those who experience certain
combinations of harsh parental behavior, is early unmanageability a risk factor for social
adjustment problems. Overall, the studies in this dissertation provide insights into the roles
of harsh or inept parenting and youth characteristics in the development of various
adjustment problems. Even though parents’ negative behaviors may affect youth social
adjustment, youth characteristics and behaviors can strongly contribute to their own
adjustment and to harsh or inept parenting.
Keywords: adolescent adjustment, harsh parenting, inept parenting, reciprocal interactions,
youth characteristics, early unmanageability
Acknowledgements
This dissertation is the fruit of several years of work and research. This memorable
journey at the Center for Developmental Research would have never happened without
people around me. Consequently, this dissertation would not have seen the light of the day
without my mentors, colleagues, friends, and family. The most important people in this
journey are my supervisors, professors Margaret Kerr and Håkan Stattin. Margaret was a
wonderful supervisor. Margaret, you were always with me when I needed your help. Your
expertise, guidance, understanding, calmness, and patience always inspired me. I could not
have asked for a better advisor and mentor. I am also very thankful to my second
supervisor, Håkan Stattin, for his time, insightful comments, support, wisdom, and new
ideas. I have learned a lot from both of you and greatly appreciate your assistance in
helping me solve many issues that I have dealt with in my professional and personal life.
You welcomed me into your research group without knowing me at all. Both of you have
opened many doors that have stimulated my professional and personal growth!
I have been fortunate to be a PhD student in Sweden. My journey to Sweden started
with professors Rita Zukauskiene from Mykolas Romeris University and Lars Bergman
from Stockholm University. Rita, thank you for building bridges between Lithuania and
Sweden! Lars, thank you for introducing me to Margaret and Håkan and helping me find
my “voice” in research! At Stockholm University, I met Daiva Daukantaite, who was my
colleague, neighbor, and friend. Daiva, I think it is fair to say that you encouraged me to
follow my dream and making me to believe that I could be a successful doctoral student in
Sweden.
Many thanks go to my colleagues at the Center for Developmental Research, for
good discussions, stimulating ideas, enjoyable coffee breaks, and many other nice
moments. I would like to specifically thank Therese Skoog, who helped me start my life in
Örebro, who brought me into her family and provided me a better understanding about the
Swedish way of life and traditions. I want to say thanks to William Burk, who was my
colleague, my neighbor, and my wonderful friend. I am grateful for your morning coffees
and evening teas, for long conversations, support, listening, help, English language lessons,
and answers to my questions, even silly ones… It seems there is too much to list. Bill,
thanks for everything! I want to express my gratitude to Nejra Beši, who constantly
reminded me there is life outside the university. I would like to thank Stefan Persson for
many things: nice food, long conversations, and his help taking care of my plants when I
have been away, rides into and out of town, and his seemingly endless knowledge about
Skåne. I would also like to personally thank Mats Larsson, who reminded me there are
many interesting and hidden things in life. Mats, thanks for letting me to discover my
harmony rings! Many thanks to other current or former colleagues that are not mentioned
by name here, but who always helped me with many things and who always made me
smile…
There are also some people from the university that made my days easier and
brighter in Örebro. I am very grateful to Monika Geisor. Without her help some paperwork
would be still be incomplete and lying on my desk. Thank you, Monika, for helping me out
all the time! I am very grateful to my friend Kjetil Duvold, who was a person with whom I
had a lot of nice conversations and could talk about life in Lithuania. Many more people
from the city of Örebro, the university and the Center for the Developmental Research
should also be acknowledged. So, I will simply say thank you to all who were there for me!
I would like to acknowledge Sheila Marshall and Henrik Andershed for their
insightful comments on a previous version of this dissertation. Your comments were of
great help, which is evident in the improved writing style and logic.
My special gratitude goes to my beloved Linas, who has been very patient all these
years. Without his love, support, understanding, and willingness to travel I would not be
where I am today. I would be remiss if I did not also acknowledge my family. I would
especially like to thank my parents. Their way of life always encouraged me to follow my
dreams and to finish what I have started, no matter how hard it could be. Many people have
asked if the topic of my research had something to do with my childhood. Considering the
loving and supportive environment in which I was raised, I guess I could say that I
experienced a lack of harsh or inept parenting!
Finally, I would like to acknowledge all the students, their parents, and school
officials who participated in (or agreed to) the collection of data, without them the studies
presented in this dissertation would not have been possible.
List of studies
This dissertation is based on the following studies, which will be refereed to in the text by
their Roman numerals.
Study I
Sebre, S., Sprugevica, I., Novotoni, A., Bonevski, D., Pakalniskiene, V.,
Popescu, D., Turchina, T., Friedrich., W., & Lewis., O. (2004). Crosscultural comparisons of child-reported emotional and physical abuse:
Rates, risk factors and psychosocial symptoms. Child Abuse and Neglect,
28, 113-127.
Study II
Pakalniskiene, V., Kerr., M., & Stattin, H. (2006). Youth characteristics as
explanations of the link between negative parenting practices and
adolescent peer relationship quality. Manuscript submitted for publication.
Study III
Kerr., M., Stattin., H., & Pakalniskiene, V. (2008). Parents react to
adolescent problem behaviors by worrying more and monitoring less. In
M. Kerr., H. Stattin & R. Engels (Eds.), What can parents do? New
insights into the role of parents in adolescent problem behavior, (pp. 91112). West Sussex: Wiley.
Study IV
Pakalniskiene, V., Kerr., M., & Stattin, H. (2007). Early temperamental
unmanageability, harsh parenting profiles, and adolescent problem
behavior: A mixture modeling approach with latent parenting classes.
Manuscript under review.
Study V
Pakalniskiene, V. (2008). Children’s temperamental unmanageability,
harsh parenting, and quality of romantic relationships in adulthood from a
longitudinal perspective. Manuscript.
Study I has been reprinted with permission from ELSEVIER.
Study III has been reprinted with permission from John Wiley & Sons Ltd.
TABLE OF CONTENTS
I Introduction.................................................................................................................................. 13
Harsh or inept parenting.................................................................................................................... 13
Harsh or inept parenting and social adjustment ................................................................................ 14
Harsh or inept parenting and the other parents’ behaviors................................................................ 17
Children’s characteristics and social adjustment............................................................................... 19
Directions of effects between negative parenting and children’s characteristics.............................. 21
Combining harsh or inept parenting practices and children’s characteristics in the prediction of later
social adjustment............................................................................................................................... 23
The aim of this dissertation ............................................................................................................... 28
II Method......................................................................................................................................... 31
Participants and procedure ................................................................................................................ 31
Sample 1 ....................................................................................................................................... 31
Sample 2 ....................................................................................................................................... 32
Sample 3 ....................................................................................................................................... 33
Measures ........................................................................................................................................... 34
III Results ......................................................................................................................................... 49
Study I ............................................................................................................................................... 49
Study II.............................................................................................................................................. 50
Study III ............................................................................................................................................ 52
Study IV ............................................................................................................................................ 54
Study V.............................................................................................................................................. 57
IV Discussion .................................................................................................................................. 59
Findings and previous research ......................................................................................................... 59
Strengths and limitations................................................................................................................... 61
What is harsh parenting? ................................................................................................................... 63
Children and youth as active agents in parent-child relationships .................................................... 65
Future directions................................................................................................................................ 67
What should parents do? ................................................................................................................... 70
Concluding remarks .......................................................................................................................... 71
V
References .................................................................................................................................. 73
2
I Introduction
Harsh or inept parenting
To understand how children develop, one must consider the environments in which
they develop. The earliest social environment for children is the family and the parent-child
relations. It is probably safe to assume that most parents want to provide a warm, safe, and
supportive environment in which their child can grow and develop socially appropriate
behavior. Everyday experience, however, suggests that parents sometimes fail at this. There
are probably few parents who can say that they have never behaved negatively or even
harshly toward their child, but some do this much more than others. What effect negative
treatment from parents has on a child’s development is, therefore, an obvious question to be
answered. Perhaps a less obvious question is why some parents do this more than others.
Although there are many possible answers, one might be found in the child him- or herself.
When a child has a temper tantrum in public, for example, most bystanders feel some
degree of irritation or anger, and they urgently want it to stop. These common emotional
reactions provide a small glimpse into the daily experience of parents with temperamentally
difficult children, and they suggest that harsh parenting might be partly a response to the
individual child. The studies in this dissertation address both questions: How negative
parenting affects development and what role children’s characteristics might play in
evoking negative parenting.
Harsh or negative parenting can mean a variety of things. In the literature, harsh or
negative parenting is often defined in terms of corporal punishment (Cohen, 1984;
Gershoff, 2002 for a review; Straus & Field, 2003; Straus & Mathur, 1995). However,
many parents do not use corporal punishment, partly because it is illegal in many countries.
Instead of corporal punishment, parents use other behaviors that may cause psychological
harm. These might include threatening the child, making the child feel guilty, or ignoring
the child. Thus, what is considered harsh parenting can include yelling, frequent negative
commands, name calling, overt expressions of anger, and physical threats and aggression
(Arnold, O’Leary, Wolf, & Acker, 1993; Deater-Deckard & Dodge, 1997). In this
dissertation, I define harsh parenting as physical punishment and verbal or nonverbal
aggression, such as anger outbursts, threats, stony silences, or rejection, thus combining
both aspects – the physical and the nonphysical.
The term “harsh parenting” has been used a lot in coercion theory (Patterson, 1982),
and in this context it is considered as a form of “inept parenting”. Inept parenting does not
13
13
have to be harsh, but it can be. Generally, it includes coercive parent-child communication,
dysfunctional disciplining practices, inconsistent control, harsh or violent physical
punishment, negative attitudes and reasoning, limited use of praise, support, or warmth, and
poor supervision and monitoring (Maccoby & Martin, 1983; Patterson, Reid, & Dishion,
1992; Reid, Patterson & Snyder, 2002; Robins & Rutter, 1990; Stoff, Breiling, & Maser,
1997). Harsh parenting and inept parenting are sometimes used synonymously. In my view,
inept parenting behaviors such as poor supervision and inconsistent discipline are not harsh
parenting practices, but are related in that they might co-occur with harsh parenting. They
might be used for the same reasons (e.g., in response to a temperamentally difficult child),
and they might have similar negative effects on children’s or adolescents’ development. I
consider harsh parenting a form of inept parenting, and I am interested in harsh parenting
specifically as well as inept parenting more generally.
In this dissertation, I focus on the role harsh or inept parenting plays in children’s
and adolescents’ development. I consider both the effects harsh and inept parenting might
have on children’s social development and how children’s behaviors or temperamental
characteristics might bring about harsh or inept parenting. I focus on the role harsh or inept
parenting plays in various aspects of adolescents’ social development in different cultures
or contexts. Some of the major issues in this area are how harsh or inept parenting is related
to social adjustment, how this link is affected by context, what role children’s own
characteristics play, and how to put all these factors together.
Harsh or inept parenting and social adjustment
Some theoretical perspectives are founded on the idea that parenting practices
determine children’s development and social adjustment. Most studies of parenting and
child adjustment problems are based on what Hartup (1978) has termed the social mold
model. This model associates family socialization processes with a mold into which the
child is placed. Problematic parenting is assumed to affect child behavior and give rise to
the development of later social adjustment problems. According to the social learning
perspective, which lies within the social mold framework, if parents solve conflicts with
their child in punitive, aggressive, or negative ways, the child may learn that this is an
appropriate way to behave in society (Bandura, 1977). Because of this, children who
experience harsh parenting or inept discipline may more often develop conduct problems,
particularly aggressive behavior, and may have more adjustment problems. Although there
may be many contributing factors in the development of adjustment problems, such as low
14
14
income, parental conflict, parental criminality, child temperament, and many others, harsh,
inconsistent, inept parenting practices and behaviors are likely to affect children’s
development and social adjustment.
Consistent with the social mold framework, or social learning theory, harsh or inept
parenting behavior is related to a number of social adjustment problems. One category of
social adjustment problems that has been linked to harsh or inept parenting comprises
externalizing problems. Various cross-sectional and longitudinal studies have suggested
that lack of parental involvement, poor acceptance, low responsiveness, lack of supervision,
harsh and inconsistent punishment, physical punishment, and insufficient rewarding of
behavior are related to higher levels, or increased risk, of the development of externalizing
behaviors during childhood and adolescence (Deater-Deckard, Dodge, Bates, & Pettit,
1996; Garstein & Fagot, 2003; Haapasalo & Tremblay, 1994; Knutson, DeGarmo, Koeppl,
& Reid, 2005; Lansford, Deater-Deckard, Dodge, Bates, & Pettit, 2004; Larzelere & Kuhn,
2005; Loeber & Dishion, 1983; Paoulucci & Violato, 2004; Pettit, Laird, Dodge, Bates, &
Criss, 2001; Rothbaum & Weisz, 1994; Straus, Sugarman, & Giles- Sims, 1997; Wakschlag
& Hans, 1999; Weiss, Dodge, Bates, & Pettit, 1992). Thus, parents’ harsh or inept behavior
has been found to be related to externalizing problems during childhood and adolescence.
Another category of social adjustment problems that has been linked to harsh or
inept parenting is internalizing problems. There is less information about how negative
behaviors in the family affect internalizing problems, but negative, harsh parenting or
corporal punishment, parental rejection and parental hostility have been associated in
several cross-sectional studies with lower self-esteem (Hertz & Gullone, 1999; Straus,
1996a), a greater probability of depression (Muris, Schmidt, Lambrichs, & Meesters, 2001;
Richter, 1994, Straus, 1996b), anxiety (Gruner, Muris, & Merckelbach, 1999; Van Brakel,
Muris, Bögel, Thomassen, 2006), and other internalizing problems and behaviors in
childhood and adolescence (e.g., Messer & Beidel, 1994). Even though there is a lack of
longitudinal evidence, parents’ harsh or inept behavior is cross-sectionally related to
various negative behaviors and internalizing problems during childhood and adolescence.
Problems in peer relationships have also been linked to harsh or inept parenting,
although this link is less well established than the link between externalizing problems and
harsh or inept parenting. By the time of adolescence, peers become more important than
parents as confidants and providers of emotional support, thus suggesting that an important
aspect of social adjustment is the ability to develop and keep relationships with other
people – starting with peer relations, and followed by romantic relations. Studies in this
area have mainly dealt with peer relationships among young children. However, a few
15
15
studies have linked negative behaviors in the family or harsh parental behaviors to peer
relationships during adolescence or early adulthood. Negative or inept parenting behaviors,
such as love withdrawal, harsh discipline, strictness, and verbal and symbolic aggression,
have been linked in cross-sectional and longitudinal studies to poor quality in children’s
and adolescents’ peer relationships, antisocial or aggressive activities in relation to peers,
antisocial activities with peers, or increases in aggressive or aversive behaviors toward
peers, peer rejection, and having trouble-making friends (e.g., Carson, & Parke, 1996;
Dekovi & Meeus, 1997; Engels, Decovic, & Meeus, 2002; Fuligni & Eccles, 1993;
Lansford, Criss, Pettit, Dodge, & Bates, 2003; McFadyen-Ketchum, Bates, Dodge, & Pettit,
1996; Paley, Conger, & Harold; 2000; Vissing, Straus, Gelles, & Harrop, 1991). Thus, it
seems that children and adolescents who experience negative parental behaviors have
various problems in relationships with their peers as well as in their behaviors.
The history of child upbringing might also have long-term consequences in
romantic or marital relationships (Andrews, Foster, Capaldi, & Hops, 2000; Capaldi &
Clark, 1998; Flouri & Buchanan, 2002; Franz, McClelland, & Weineberger, 1991; Linder
& Collins, 2005). In longitudinal studies, parent-child closeness, poor parenting practices,
such as monitoring and discipline, or aversive communication in adolescence has been
associated with the quality of relationships with a partner in midlife (Flouri & Buchanan,
2002) or with the physical aggression toward a partner in young adulthood (Andrews,
Foster, Capaldi, & Hops, 2000; Capaldi & Clark, 1998). It has also been found that adults
who have experienced harsh or inept parenting or physical punishment are not satisfied
with their relationships or do not have positive perceptions of current romantic partners
(Colman & Widom, 2004; Finkelhor, Hotaling, Lewis, & Smith, 1989; Fleming, Mullen,
Sobthorpe, & Bammer, 1999) or have trouble maintaining intimate or romantic
relationships (Colman & Widom, 2004; Felitti, 1991; Fleming et al., 1999). Thus, negative
or harsh parental behavior, especially during late childhood and adolescence, might have
long-term consequences for the child’s future relationships.
It seems safe to conclude that harsh or inept parental behavior is related to various
children’s social adjustment problems. The studies cited above, however, are mainly based
on North American samples. The question remains if the same conclusion can be drawn
from samples outside North America. There are not many studies that have explored
culture, which could be defined as all the behaviors, ways of life, and beliefs of a
population that are passed down from generation to generation (Merriam Webster online
dictionary), or ethnic group, defined as a certain population of people whose members
identify with each other, classed according to common racial, national, tribal, religious,
16
16
linguistic, or cultural background (Merriam Webster online dictionary) in the effects of
harsh parenting on adjustment problems. Exiting studies have mainly examined the effects
of physical punishment on externalizing behaviors (e.g., Deater-Deckard et al., 1996;
Dornbusch, Ritter, Leiderman, Roberts, & Fraleigh, 1987; Gunnoe & Mariner, 1997;
Lansford et al., 2004; Lansford et al., 2005; McLeod, Kruttschnitt, & Dornfield, 1994;
Rowe, Vazsonyi, & Flannery, 1994; Spieker, Larson, Lewis, Keller, & Gilchirist, 1999).
Although existing studies present mixed results, they suggest that physical punishment may
be used in some cultures or ethnic groups more than in others, for example, among African
Americans or among families in Kenya (Deater-Deckard et al., 1996; Lansford et al., 2005).
Also, physical punishment seems to increase the externalizing problems or affect the
academic achievement of European American children, but not African American children
(Deater-Deckard et al., 1996; Lansford et al., 2004). However, it is still not known if other
forms of harsh or inept parenting, such as shouting at or ignoring the child, have different
consequences for adjustment in cultures outside North America.
To summarize, adolescents who experience harsh, inept, or negative parenting are at
greater risk of a range of maladaptive behavioral outcomes. There is much variability,
however, and many with this increased risk are normally adjusted. Some children seem to
be able to accept or ignore parents’ harsh or inept behaviors such as angry outbursts, for
example, and adjust themselves. Consequently, these children might even forget that their
parents are angry at them, and behave and feel in similar ways to children who do not
experience parents’ habitual angry outbursts. However, other children may start to be more
anxious, unwilling to be open with their parents, and afraid of conflicts with their peers.
The question is why some children who experience negative parenting do not have
problems later in life while others do. The previous research, mentioned above, leaves this
question unanswered.
Harsh or inept parenting and the other parents’ behaviors
Harsh or inept parental behavior is associated with various children’s problems in
the future, but it seems to be associated with problems for some children but not for others.
What might influence whether or not children who experience harsh or inept parental
behavior will have adjustment problems later in life? Harsh or inept parenting behaviors
might be perceived differently by children, and might have different effects on their
development if they take place along with behaviors that communicate warmth, support,
and love rather than those that communicate coldness, worry, or rejection (Deater-Deckard
17
17
et al., 1996; Lansford et al., 2004). Specifically, it might be that other behaviors, in the
context of which harsh parenting occurs are more important than the harsh parenting, per
se, in predicting effects on children’s development. However, the recently suggested idea
that the effects of harsh or inept parenting could be dependent, in part, on the combination
of harsh or inept parenting and other negative or positive parenting behaviors, is not yet
very well developed in research.
Ideas about the influence of other parenting behaviors on the consequences of harsh
or inept parenting have been developed in cross-ethnic-group studies showing that spanking
by parents seems to increase children’s externalizing problems among European American
but not African American children (Deater-Deckard et al., 1996; Lansford et al., 2004). It
has been speculated that African American parents tend to use physical punishment in the
context of other behaviors that communicate love, whereas European American parents
tend to use it in the context of other behaviors that communicate rejection. Hence, what
children experience as harsh parenting might depend on what they know about their
parents’ feelings, attitudes, and intentions. It has been suggested that physical discipline is
unrelated to children’s externalizing problems, such as aggression or delinquency, after
parental behaviors like rejection or low warmth and involvement, which are associated with
physical discipline, have been taken into account (Lansford et al., 2005; Larzelere, Klein,
Schumm, & Alibrando, 1989; Rohner, Bourque, & Elordi, 1996; Simons, Johnson, &
Conger, 1994). For example, in some cross-sectional studies, physical punishment was
linked to maladjustment when it was perceived as rejection (Rohner et al., 1996), or when it
was perceived as not normative behavior in the cultural setting (Lansford et al., 2005).
Evidence from a longitudinal study suggests that spanking is associated with an increase in
behavior problems over time when mothers are not supportive; the pattern was found to
hold for European American, African American and Hispanic American children, despite
the fact that African American children were more likely to be spanked and spanked more
frequently than the others (McLoyd & Smith, 2002). Thus, the relations between physical
punishment and behavioral problems are not related to race or ethnicity, but to other
behaviors that co-occur with spanking. Other parental behaviors may change or influence
the effects of physical punishment on later problem behaviors, thus suggesting that the
effects of one parental behavior might be affected by other parental behaviors.
Consequently, adequate assessment of the effects of physical discipline on children’s
development may require taking into account other positive or negative parenting behaviors
in the context of which harsh or inept parenting occurs.
18
18
Children’s characteristics and social adjustment
Many studies in the literature on harsh or inept parenting tend to neglect the child’s
role, but children’s characteristics may be an important factor in social adjustment. This
idea is well developed in temperament research, where temperamental traits are viewed as
early-emerging individual differences that shape the course of personality development,
and both its healthy and problematic outcomes (Rutter, 1987). In this view, temperament is
an early form of personality, which becomes elaborated over time into the stable behavioral
dispositions (e.g., Caspi, 2000) that influence adjustment. In this way, early characteristics
of the child can be related to later adjustment.
It is easy to imagine that a child who is aggressive or hyperactive will have various
behavioral problems in kindergarten, at school, and even later in life. Evidence from
temperament research suggests that a child who is temperamentally prone to anger,
aggression or opposition has more behavioral and externalizing problems later in life (e.g.,
Bates, 1989; Bates, Pettit, Dodge, & Ridge, 1998; Caspi, 2000; Caspi, Henry, McGree,
Moffitt, & Silva, 1995; Caspi, Moffitt, Newman, & Silva, 1996; Eisenberg, Fabes, Shepard,
Murphy, Guthrie, Jones, et al., 1997; Guerin, Gottfried, & Thomas, 1997; Henry, Caspi,
Moffitt, & Silva, 1996; Leve, Kim, & Pears, 2005; Morris, Silk, Steinberg, Sessa,
Avenevoli, & Essex, 2002; Stoolmiller, 2001; White, Moffitt, Caspi, Bartusch, Needles, &
Stouthamer-Loeber, 1994). For example, lack of self-control in childhood, or impulsivity,
has been found to be related to externalizing problems in later childhood and adolescence
(Caspi et al., 1995; Schwartz, Snidman, & Kagan, 1996; Shaw, Owens, Giovannelli, &
Winslow, 2001). Thus, children who are temperamentally prone to anger, aggression or
opposition may have conduct problems or be more delinquent later in life. But problem
behavior is only one indicator of poor social adjustment. The question is whether or not
these children have more problems in their relationships with others.
A number of youth personality traits, or externalizing and internalizing
characteristics, have been connected with changes over time in the quality of adolescents’
and young adults’ social relationships. Characteristics that have predicted changes in
relationship quality are personality traits, such as extraversion, shyness, neuroticism, or
agreeableness (Asendorpf & Wilpers, 1998; Neyer & Asendorpf, 2001), and internalizing
problems, such as depressed mood (Nolan, Flynn, & Garber, 2003; Prinstein, Borelli,
Cheah, Simon, & Aikins, 2005; Stice, Ragan, & Randall, 2004) or self-esteem (Neyer &
Asendorpf, 2001). Thus, youth personality traits and characteristics may affect youth
relationships. There is also evidence that later romantic relationships can be affected by
early childhood characteristics or personality traits. Even children’s temperament and
19
19
behavior styles at age 3 or childhood tantrums have been linked to experiences in
relationships at age 21 (Newman, Caspi, Moffitt, & Silva, 1997) and divorce at age 40
(Caspi, Elder, & Bem, 1987). Empirical evidence also suggests that negative emotionality,
neuroticism, or aggressiveness during childhood, adolescence, or adulthood could have
long term effects on romantic or marital relationships (e.g., Bouchard, Lussier, & Sabourin,
1999; Blum & Mehrabian, 1999; Donnellan, Conger, & Bryan, 2004; Huston & Houts,
1998; Kinnunen & Pulkkinen, 2003; Watson, Hubbard, & Wiese, 2000). Taken together, a
variety of children’s and youths’ characteristics are important components of later social
adjustment, and these characteristics seem to affect peer relationships concurrently and
romantic relationships later in life.
Characteristics of the child might also determine whether the experience of harsh or
inept parenting results in later adjustment problems. Findings from a recent study suggest
that the link between physical discipline and later problem behavior is moderated by early
childhood problems and other positive or negative parenting behaviors or attitudes (Lau,
Litrownik, Newton, Black, & Everson, 2006). In this study, for both African Americans
and European Americans, physical discipline was related to subsequent child externalizing
problems when children had behavior problems at an early age. Although the authors did
not test the combination of early externalizing problems, parental discipline and warm
parental attitudes in predicting later externalizing problems, they suggested that physical
discipline might be particularly damaging when used with children who already have
various behavior problems. Thus, the results of the study suggest not only that physical
disciplines may cause social adjustment problems, but also that children’s characteristics
and parents’ attitudes and other behaviors, are important factors in this link. Thus, the
effects of negative parenting on later adjustment may be affected by the child’s earlier
behaviors, but the combined effects on adjustment of early problem behaviors, parental
discipline, and other parental behaviors remains unclear.
To summarize then, as well as other positive or negative parents’ behaviors that
could communicate love or rejection, there is another independent factor – children’s
characteristics – that may be related to later adjustment problems. Hence, negative
parenting, children’s characteristics and later social adjustment are related. Some
theoretical perspectives assume that parents’ behaviors and children’s characteristics are
related to each other because the child’s own earlier behavior shows continuity with later
problem behavior and might also evoke physical discipline or negative parenting along the
way (Belsky, 1997; Lytton, 1997). In this way of reasoning, it is obvious that harsh or inept
parenting and children’s characteristics affect each other.
20
20
Directions of effects between negative parenting and children’s
characteristics
Certain children’s characteristics may create more problems later in life than others
– not only problems with peer relationships or relationships with romantic partners, but also
in relationships with parents. Parents, like all other human beings, react to the people
around them. Given that parents spend a lot of time with their children, it is natural to
assume that parents and children will react to each other and affect each other’s behaviors.
Negative or coercive behaviors on the part of parents may be evoked by child problem
behavior, and then may strengthen child behavioral tendencies (Lytton, 1990). Theoretical
perspectives that assume bidirectional relations between parents and children allow the
development of social adjustment problems to be thought of as an ongoing and changing
process, involving interaction between children and their parents (Bell, 1968; Sameroff &
Mackenzie, 2003). For example, coercion theory, which is well-known to place an
emphasis on reciprocal influences between parents and children, assumes that certain
problem behaviors in children elicit ineffective or harsh parenting behaviors to which
children tend to react by escalating their own problem behaviors (Patterson, 1986; Patterson
et al., 1992; Snyder & Patterson, 1995). Eventually, parents back down, leaving the child
rewarded for his or her behavior. Thus, it seems that parents and children may affect each
other, starting at a very early age and extending through adolescence.
In the vast body of research on parenting and child or adolescent development, only
a small proportion of studies have adopted a bidirectional approach (Crouter & Booth,
2003). This is because, in many studies, parental and family factors have been regarded as
some of the most important environmental influences on a child’s development. In many
negative-parenting studies causality resides only with parents, and the child effect was not
considered. It was over 30 years ago that Bell (1968) proposed that child qualities influence
the harshness of parental discipline and conflicts in the parent-child relationship, but this
view has been echoed more recently by temperament researchers (Lytton, 1990; Putnam,
Sanson, & Rothbart, 2002; Rothbart & Bates, 1998). Numerous experimental and
longitudinal studies have shown that adults react negatively to various types of children’s
externalizing problems (e.g., Anderson, Lytton, & Romney, 1986; Buss, 1981; Dix, Ruble,
Grusec, & Nixon, 1986; Huh, Tristan, Wade, & Stice, 2006; Mulhern & Passman, 1981;
Passman & Blackwelder, 1981). Studies in the behavior genetic tradition have shown that
children at genetic risk of antisocial behavior are more likely to experience corporal
punishment and negative control (Jaffee, Caspi, Moffitt, Polo-Tomas, Price, & Taylor,
2004; O’Connor, Deater-Deckard, Fulker, Rutter, & Plomin, 1998). Although there is less
21
21
research on adolescents, the findings that exist suggest that antisocial or externalizing
behavior can affect parenting in negative ways (Ge, Conger, Cadoret, Neiderhiser, Yates,
Troughton, & Stewart, 1996). It appears that youth behavior evokes harsh parenting in
adoptive parents (Ge et al., 1996). Thus, children’s and adolescents’ behavior and
externalizing problems may evoke negative parenting, but the question is whether many
other children’s and youths’ characteristics may also evoke negative parental behavior.
Although obvious externalizing children’s and youth problem behaviors may affect
parents’ behaviors, youth internalizing problems may also affect parents’ behaviors. There
are only a few studies examining the relationship between youth internalizing problems and
negative, inept parental behaviors, but there is some evidence that negative parental
behaviors might be partly a reaction to youths’ internalizing problems. For example,
youths’ depression and anxiety may evoke changes in parents’ negative control (O’Connor
et al., 1998), or increase parents’ psychological control over time (Rogers, Buchanan, &
Winchell, 2003). Thus, parents react to youth depression and anxiety, but it is not clear
whether parents react to other youth internalizing problems, such as low self-esteem or
fears of various kinds.
Empirical evidence suggests that children may evoke certain parents’ behaviors, but
the question is whether parents’ behaviors have an effect on children’s later behaviors, or
whether it is possible to talk about bidirectional relations. Bidirectional effects between
parents and children have been examined mostly in empirical research on children’s and
adolescents’ engagement in antisocial or delinquent behaviors. There is good evidence for
bidirectional effects between childhood problem behaviors and negative, inept parents’
reactions (e.g., Cohen & Brook, 1998; Hastings & Rubin, 1999; Kandel & Wu, 1998;
Kochanska, 1998; Maccoby & Jacklin, 1983; Mink & Nihira, 1986; Snyder, Cramer,
Afrank, & Patterson, 2005; Stice & Barrera, 1995). Even at a very early age, the
troublesome behavior of 12-months-old boys leads to mothers backing off, which in turn
leads to greater difficulties at 18 months (Maccoby & Jacklin, 1983). Given that, in early
childhood, parents and children affect each others’ behavior, it is safe also to assume that,
during adolescence, when a lot of teenagers engage in risky behavior, parents and children
will affect each others’ behavior. There is evidence of reciprocity in negative affect or
hostile behaviors between parents and adolescents (Carlson & Parke, 1996; Conger & Ge,
1999; Eisenberg et al., 1997; Kim, Conger, Elder, & Lorenz, 2001). There are also
bidirectional relations between adolescents’ disruptive and inflexible problem-solving
behaviors and their parents’ hostile, coercive, inconsistent parenting strategies (Reuter &
Conger, 1998), and between parental conflict-negativity and adolescents’ antisocial
22
22
behavior (e.g., Neiderhiser, Reiss, Hetherington, & Plomin, 1999). Parents react to
adolescents’ behavior and adolescents react to parents’ behavior. Even though it has been
suggested that parents react negatively to youths’ externalizing problems and youth react
negatively to certain behaviors on the part of their parents, there is still the question of what
kinds of relationships can be found between negative or inept parenting practices and other
youth characteristics (such as internalizing problems).
Taken together then, the findings concerning children’s and adolescents’ problem
behaviors and harsh or negative parenting give reasons to expect that children’s and youths’
various characteristics and behaviors influence parents’ use of commands, discipline, and
punishment; certain behaviors on the part of parents can later affect children’s or youths’
behaviors. Given that parents may respond to other youth characteristics that have not been
considered in previous studies, there is also the question of how parents and youths respond
to each other over time. For example, depressed mood is often accompanied by lethargy
and lack of motivation. Thus, youths in a depressed mood may have trouble getting up for
school in the morning or attending to responsibilities such as homework or household
chores. Research suggests that parents are often unaware of youths’ depressed mood
(Mesman & Koot, 2000). Under these conditions, behaviors such as lethargy, lack of
motivation, and trouble getting out of bed can evoke angry outbursts; the longer the
duration of depressed mood, the more habitual parents’ angry outbursts may become.
Similarly, youths with internalizing problems, such as low self-esteem, which often cooccurs with depressed mood, are probably not very proactive about trying new things or
taking on responsibilities. Parents might, out of frustration or disappointment, resort to
angry outbursts or rejection, which can affect children’s self-esteem or generate sadness in
the long run, which suggests that parents and youth do respond to each other. Thus, there
may well be other child and youth characteristics that parents respond to that have not been
considered in previous studies.
Combining harsh or inept parenting practices and children’s
characteristics in the prediction of later social adjustment
Children with certain characteristics might be predisposed to develop various
adjustment problems later in life, but the ways their parents handle them can make such
development more or less likely. It is reasonable to suppose that children’s behavior evokes
the very parenting behavior that ends up contributing to the shaping of their development
(e.g., Bates et al., 1998; Leve et al., 2005; Stoolmiller, 2001). Harsh or inept parenting
might strengthen temperamental predispositions to develop aggressive conduct problems.
23
23
To test this idea, researchers have applied the concept of moderation, which suggests that
among children who are, say, temperamentally anger-prone and oppositional, courses of
development might differ substantially between those who experience harsh or inept
parenting and those who do not. Some studies have reported tests for moderation by
investigating interactions between children’s characteristics and harsh or inept parenting in
the prediction of later behavior problems. Although many researchers have proposed that
children’s adjustment is predicted by interaction between parenting and temperament (e.g.,
Lerner & Lerner, 1994), few studies have tested this interaction directly. Thus, it is
assumed that children’s characteristics and parents’ behaviors are tightly related to each
other and may both affect later adjustment problems. But the question is which children’s
characteristics can be related to harsh or inept parent behavior and also affect later
adjustment.
It is reasonable to suppose that an angry or aggressive child will experience more
negative parenting than other children and may have more problems later in life. On the
whole, children with a difficult temperament are more likely than other children to develop
behavior problems under adverse family conditions, which include unclear family rules,
low consensus between parents, parental inconsistency, maternal rejection, inconsistent
discipline, and harsh discipline (e.g., Colden, Lochman, & Wells, 1997; Lengua, Wolchik,
Sandler, & West, 2000). When difficult temperament is accompanied by harsh parental
behavior or poor mother-child relationships, problem behavior increases substantially (Leve
et al., 2005). Although child temperament and parents’ behavior, or a combination of the
two, are important determinants of later social adjustment problems, there is scant evidence
for interaction between harsh parenting and child’s temperament. For example, one early
study failed to find an interaction between under-controlled temperament before age 3 and
harsh parenting at age 3 in predicting criminal convictions by age 18 (Henry et al., 1996).
Also, it has been found that hyperactivity and mother-child interaction are equal predictors
of peer relationship problems, but that the interaction between them is not (Hinshaw,
Zupan, Simmel, Nigg, & Melnick, 1997; Keown & Woodward, 2006). The findings were
that preschool boys, age 4-5 or 6-12, with hyperactive behavior problems were less
accepted by their peers, and that hyperactivity and the quality of early mother-child
interactions both made unique contributions to the development of peer relationship
difficulties. Both these studies, however, were cross-sectional, and both considered only
preadolescent boys. Even though some results suggest that the combination of parenting
and temperament is an important aspect of the prediction of children’s adjustment, the
findings remain mixed.
24
24
Why is it that the results of the different studies do not form a clear picture? At
present, there is some empirical evidence for the importance of interactions between child
characteristics and parenting in the prediction of child problem behavior (Van Leewen,
Mervielde, Braet, & Bosmans, 2004). Studies that have looked at interactions between child
characteristics and negative parental behavior vary with regard to: (a) gender, that is, only
boys (e.g., Anderson et al., 1986; Belsky et el., 1998; Colder et al., 1997; Stoolmiller, 2001)
or both genders (e.g., Bates et el., 1998); (b) age, that is, preschool (e.g., Paterson &
Sanson, 1999; Rubin et al., 1993), school-age (e.g., Lengua et al., 2000; Stoolmiller, 2001;
Wootton, Frick, Shelton, & Silverthorn, 1997), or adolescent (e.g., Carlo, Roesch, &
Melby, 1998; or (c) design, that is, cross-sectional or longitudinal (e.g., Bates et el., 1998;
Belsky, Hsieh, & Crnic, 1998; Rubin, Burgess, Dwyer, & Hastings, 2003; Stoolmiller,
2001). These studies differ also in how parental behavior and child temperament have been
assessed. For example, the measures of temperament are either not very early (e.g.,
Stoolmiller, 2001; Leve et al., 2005) or are solely or partly retrospective judgments of early
temperament (Bates et al., 1998; Henry et al., 1996). It would be desirable to see the effect
tested using prospective measures of early temperament. One further reason why results
presented by different studies are mixed is that it is notoriously difficult to predict and find
interaction effects (for discussions of this, see McClelland & Judd, 1993; Stoolmiller,
2001). Therefore, some studies have taken another approach, that of grouping analyses, or
of adopting a person-centered approach, to test the effects of harsh or inept parenting (Bates
et al., 1998; Stoolmiller, 2001). These studies suggest that children with a difficult
temperament are in danger of unfavorable outcomes when they are exposed to negative
parenting or parental control.
The question is whether older children with certain characteristics are also in danger
of having more problems later in life when they too are exposed to harsh or inept parental
behavior. Given that temperament is thought of as the foundation of later personality (Caspi
& Silva, 1995; Eisenberg, Fabes, Guthrie, & Reiser, 2000), it follows that researchers
interested in older children or adolescents will look at the interplay between temperament
or personality and harsh or negative parenting in explaining later adjustment. It has been
suggested that ignoring personality-environment interactions and considering only main
effects can lead to spurious predictions of problem behavior (Van Leeuwen et al., 2004).
While the main effects of child temperament and parental behavior on child problem
behavior have been quite well documented in past research, there is not much evidence in
the literature about the effects of personality and harsh or inept parenting on adjustment
problems (Akse, Hale III, Engels, Raaijmakers, & Meeus, 2004; Andrews, Foster, Capaldi,
25
25
& Hops, 2000; Barber, 1992; Capaldi & Clark, 1998; Collins, Maccoby, Steinberg,
Hetherington, & Bornstein, 2000; Donnellan, Larsen-Rife, & Conger, 2005; Kim, Conger,
Elder, & Lorenz, 2001; O’Connor & Dvorak, 2001; Prinzie, Onghena, Hellinckx, Grietens,
Ghesquière, & Colpin, 2003; Van Leeuwen et al., 2004). For example, negative parental
control or coercive parental behavior has been found to be more related to externalizing
behavior for under-controllers, characterized by low scores on conscientiousness and
benevolence, than for other children. However, negative parental control influences
internalizing behavior for introverted children (Prinzie et el., 2003; Van Leeuwen et al.,
2004), thus suggesting that children with different personality traits may develop different
problems when they experience negative parenting. Thus, negative or inept parental
behavior might affect some children more than others.
Negative parenting practices and children’s characteristics can be combined in
various ways in the prediction of later social adjustment. Some researchers have tested the
idea of moderation, which suggests that among children who are, for example,
temperamentally anger-prone and oppositional, the course of development might be very
different for those who experience negative or inept parenting than for those who do not.
But other researchers have combined parents’ behaviors, child characteristics, and later
social adjustment in a different way. The researchers who have tested mediation models
assume either that parents’ behaviors mediate the link between children’s problem
behaviors and adjustment problems or that children’s behaviors mediate the link between
parents’ behavior and adjustment (e.g., Capaldi & Clark, 1998; Clark & Ladd, 2000; Engels
et al., 2002; Simons, Chao, Conger, & Elder, 2001). Researchers testing the idea of equal
predictors assume that parents’ behaviors and child’s characteristics or personality traits
equally predict adjustment problems (e.g., Donnellan, Larsen-Rife, & Conger, 2005;
Hinshaw et al., 1997; Keown & Woodward, 2006). Thus, there are several ways of
combining parents’ behavior, child characteristics and later social adjustment, each of
which has its own advantages and disadvantages. In one way or another, parents’ and
children’s behaviors are related to each other and may affect later social adjustment.
Taken together, children or adolescents with a difficult temperament or personality
traits are particularly in danger of developing various problems when they are exposed to
negative parenting or parental control. Temperament or personality studies suggest that
children’s characteristics and harsh or inept parenting are related in predicting later
adjustment problems, but these studies have some limitations and raise new questions. For
example, one question is how very early temperament, various personality traits, and
characteristics such as internalizing problems may be related to harsh or inept parenting in
26
26
predicting adjustment problems and the quality of peer or romantic relationships. One of
the limitations of temperament studies is that the measures of temperament are either not
very early or are solely or partly retrospective judgments of early temperament. Another
potential limitation is the use of single aspects of parenting; combinations of harsh
parenting behaviors and other behavior that could communicate love or rejection may be
more important than single aspects in the prediction of adjustment problems (DeaterDeckard et al., 1996; Lansford et al., 2004). A further question is how children or
adolescents with certain characteristics and who experience different combinations of
parents’ behaviors (e.g., physical punishment in the context of behaviors that communicate
either acceptance or rejection) will behave later in life. Previous studies have not answered
this question.
To summarize, previous research leaves several questions unanswered concerning
harsh or inept parenting and the consequences of experienced harsh or inept parenting. One
question is whether experienced harsh or inept parenting is associated with various
adjustment problems for children from different cultures in a similar way. From the
literature, it seems clear that cultural expectations have a lot to do with whether children
perceive parenting behaviors as harsh, and consequently, whether harsh parenting
undermines their adjustment. Most of the research, however, has been performed in North
America. Thus, studies are needed that investigate the links between harsh parenting and
adjustment in cultures outside North America. A second unanswered question concerns
how parents and youths respond to each other over time. Some research findings suggest
that parents react to adolescents’ behaviors, and adolescents react to parents’ behaviors.
However, studies examining bidirectional relations between inept parental behaviors and
adolescent behaviors have mostly concerned youth externalizing problems. Thus, studies
are needed that examine the links between inept parenting and various youth
characteristics. A third unanswered question is whether the link between harsh or inept
parenting and adjustment problems is affected by child and youth characteristics? From
some research findings, it seems that harsh or inept parenting, various adjustment problems,
and children’s or youths’ characteristics are related; consequently, it is not a simple matter
of parents affecting child or adolescent behaviors. To some degree, harsh or inept parenting
may be a response to a child’s characteristics. Thus, studies are needed that investigate the
links between harsh or inept parenting and adjustment that include children’s
characteristics. And finally, there is the question of the ways in which the earlier
characteristics of children who experience physical punishment in the context of other
behaviors that could communicate negative or positive emotions affect later behavior or
27
27
relationships. Previous research suggests that other parental behaviors, in the context of
which harsh parenting occurs, may change or influence the effects of physical punishment
on later adjustment problems. However, there are only a few studies that include other
parental behaviors when examining the impact of physical punishment on problem
behavior. Consequently, studies are needed to examine the effects of different combinations
of parenting behaviors on various adjustment problems. Thus, several questions remain
unanswered concerning harsh or inept parenting and why harsh or inept parenting is linked
to various outcomes. This dissertation focuses on these unanswered questions.
The aim of this dissertation
The main aim of this dissertation is better to understand the relationships between
harsh or inept parenting and children’s characteristics in the prediction of adjustment
problems later in life. It consists of four studies. Study I examines the child-reported
incidence of emotional and physical aggression in countries that have not been reported
upon in the existing literature; it also investigates the relationship between childexperienced harsh parenting and psychosocial symptoms during adolescence. Study II
examines whether links between harsh parenting and adolescents’ peer relationship quality
might be explained by youths’ internalizing problems and psychopathy-like personality
traits. There was an examination of whether youth characteristics may influence harsh
parenting, and also interfere with peer relationships, thus explaining the link between harsh
or inept parenting and the quality of peer relationships. Since there is evidence that youths
with characteristics like depression might perceive their friendships or friends’ behaviors
toward them differently from their friends (Daley & Hammen, 2002), an additional
investigation was made of whether the results can be verified using peers’ independent
reports of relationship quality. Study III examines how parents and youths respond to each
other over time, and also tests gender differences. Additional analyses of the mechanisms
underlying the links between youth behaviors and inept parental behaviors were
investigated. Studies IV and V examine various combinations of harsh parenting behaviors
and model their relations to early temperamental unmanageability and adolescent problem
behaviors or romantic relationships quality in adulthood. These studies examine how
distinct patterns of physical discipline and discordant relationships relate to early
unmanageable temperament and later conduct problems, norm violations, and romantic
relationship quality. They also consider how, apart from these links, early unmanageability
28
28
relates to later conduct problems and norm violations or the quality of romantic
relationships in adulthood. The following research questions were posed:
1) Is experienced harsh parental behavior associated with adjustment problems for
children from different cultures in a similar way? (Study I)
2) How do parents and youths respond to each other over time? (Study II, Study III)
3) Is the link between harsh or inept parenting and adjustment problems affected by
child and youth characteristics? (Study II, Study III, Study IV, Study V)
4) How do the early characteristics of children who experience physical punishment in
the context of other behaviors that communicate negative emotions affect later
behavior? (Study IV, Study V)
29
29
30
II Method
Participants and procedure
Sample 1
The sample for Study I included school students 10-14 years of age in 4th to 7th
grade from several Baltic and Eastern European countries, such as Latvia, Lithuania,
Macedonia, and Moldova. Children from 4th grade were mostly 10-11 years old, and those
from 7th grade mostly 13-14 years-old. Any child younger than 10 years or older than 14
years was excluded from the study. The primary purpose of the study was to evaluate
relationships between emotional and physical aggression and psychosocial symptoms in
several countries where such a type of investigation had not been conducted before. Data
collection took place in the 4th and 7th grade classrooms during the school day, and was
performed according to the same procedure by each country’s research team. Participants
included 297 children from Latvia, 300 children from Lithuania, 302 children from
Macedonia, and 246 children from Moldova. The data were collected during the spring of
1998 in Latvia, the spring of 1999 in Lithuania, and the spring of 2000 in Macedonia and
Moldova. Data collection within each country took place in two large-city schools, two
medium-city schools, and two small-city or rural schools. The cities and schools were
randomly chosen from different regions of each country, with the stipulation that the
children attending any selected school would be fluent in the major national language. Such
a stipulation was made since the questionnaires for this initial study had been translated into
the national languages of each country, namely Latvian, Lithuanian, Macedonian, and
Moldovan. Consequently, the ethnic composition of each sample was primarily Latvian,
Lithuanian, Macedonian, and Moldovan – greater than 90% in each case.
Since research ethics committees did not exist in the four countries involved in this
study, the research team consulted with epidemiological researchers from the United States
with regard to the most appropriate strategy for guaranteeing that the rights of the subjects
would be respected. Permission to conduct the study was first received from local
authorities and school boards. Parents then received information that a study was taking
place concerning adolescents’ thoughts, feelings and relationships, and that the study was
voluntary and confidential. If parents objected to their child’s participation, they were asked
to inform their research team. In fact, several parents from each location did so, and their
children were excluded from the study. Children were told that they would be asked to fill
in a questionnaire regarding their thoughts, feelings, and relationships, and were informed
that participation in the study was voluntary, completely confidential, and anonymous.
31
31
After the questionnaires had been filled-in, the children were given the contact information
of research team members if they wished to discuss any issues or questions concerning the
study.
Sample 2
This sample was used in Studies II and III. The data were from a five-year
longitudinal study that took place in one community in central Sweden. This community
has a population of about 26,000. The unemployment rate is similar to that in Sweden as a
whole (6%). The mean income is somewhat lower than the rest of the country (214,000
Swedish Crowns per year compared with 223,000 for the rest of the country). Twelve
percent of the inhabitants in the community have a foreign background. The data collection
started in the fall of 2001. The primary purpose of the longitudinal study was to understand
the joint roles of parents, peers, and individual characteristics in the development of
adolescent adjustment problems and criminality.
All students in grades 4 through 12 (roughly, ages 10 to 18) were invited to
participate in the study each year. One new cohort came into the study each year (those
entering the 4th grade) and one cohort left the study (those who graduated from high school
the year before). Every second year, parents of children participating in the study received a
questionnaire in the mail, and they participated by filling it in and returning it. Only parents
of 4th through 10th graders were asked to participate, however, because many youths in 11th
and 12th grades would have reached the legal age of independence in Sweden (18), were
living on their own, or both. We targeted all youths in the community so that when youths
named peers who were important to them, those peers were also in the study and had selfreported on their own behaviors or relationships. In this way, data on peers’ behaviors or
relationships were independent of the youths who named them and not affected by the
youths’ own perceptions and biases, which might have inflated similarity (e.g., Iannotti,
Bush, & Weinfurt, 1996).
Youths were recruited in their classrooms during school hours. They were told what
kind of questions would be included in the questionnaires and how long it would take to fill
them in. They were informed that participation was voluntary and that, if they chose not to
participate, they were free to do something else instead. They were assured that if they did
participate, their answers would not be revealed to their parents, teachers, the police, or
anyone else. Parents were informed about the study in advance, in meetings held in the
community and by mail. Some parents did not want their children to participate in the study
(1%). Parents were also told that they could withdraw their child from the study at any time
32
32
they chose. Thus, youths participated if they voluntarily chose to do so and if their parents
did not object to their participation. They filled in the questionnaires during regular school
hours in sessions administered by trained research assistants. Teachers were not present.
Youths were not paid for their participation, but for each of the classes in grades 4 through
to 6 we made a contribution to the class fund, and for each of the classes in grades 7
through to 12 we arranged a lottery with movie tickets as prizes; all those who stayed in the
room, whether they filled in a questionnaire or not, were eligible for the lottery.
Study II used data from participants who were in 7th through 9th grades (ages 13-15)
at the first of two survey waves (Time 1). Because these analyses involved close peer
relationships, the sample was limited to those who participated at both time points and had
reported having a close peer at Time 1 and Time 2, and also had reported on a relationship
with a close peer at Times 1 and 2. Study III used data from parents’ reports from Times 1
and 3 for youths who were in grades 4 through 8 at Time 1 (ages 10 through 14), so that
both youths and their parents participated at two time points.
Sample 3
The sample used for Studies IV and V was based on data from a longitudinal study
of Swedish children and their parents that was started in the mid-1950s by researchers at
the Clinic for the Study of Children’s Development and Health at the Karolinska Hospital,
Stockholm. The study was part of an international investigation organized by the Centre
International de l’Enfance in Paris. Every fourth pregnant woman who registered at the
Solna Prenatal Clinic (in a suburb of Stockholm) from April 1955 to April 1958 was invited
to participate in a long-term pediatric study (in Sweden, all pregnant women receive regular
care at prenatal clinics). Only 3% of those who were asked refused to join the study. Of the
198 mothers agreeing to participate, 6 withdrew due to abortion, and 4 due to infant death
during delivery, and a further 5 were excluded due to the premature infant’s low birth
weight (under 2,000 grams). A pilot group comprising 29 children and their mothers were
contacted either before or after birth, and were added to the study. Since 98.5% of mothers
give birth to children at hospitals in Sweden, invitation to participate to those mothers after
birth does not automatically suggest a selection effect, but a selection effect is likely due to
differences in mothers’ willingness to cooperate. In this way, 183 children from the Solna
Antenatal Clinic (103 boys and 80 girls) and 29 children from the pilot group made up the
212 children (122 boys and 90 girls) who took part in the study. During 1956-1957, there
were 52.1% male births in Solna compared with 51.6% in Sweden as a whole. A t-test,
conducted by Karlberg and colleagues (1968), did not show any significant percentage
33
33
difference. This indicates that the distribution of boys versus girls did not differ by more
than might have been expected by chance.
Comparisons on parents’ socioeconomic status, age, and marital status, as well as
on sibling order and children’s gestational age and birth weight, have shown the sample to
be representative of children in Swedish urban communities (see Karlberg, Klackenberg,
Engström, Klackenberg-Larsson, Lichtenstein, Stensson, & Svennberg, 1968; Stattin &
Klackenberg-Larsson, 1990). Extensive information about the participants has been
collected over the years by means of somatic registers, medical examinations, interviews,
inventories, ratings, objective tests, sociometric methods, and projective techniques. On
each data collection occasion, the aim of the collection was to map the participants’
somatic, psychological, and social development.
Children and their parents were examined four times with equal spacing (every
three months) during their first year, twice (every six months) during the second year, and
annually (close to their birthdays) thereafter up to the age of 18. Collections of data were
also performed at the average ages of 21, 25, and 35 years. Up to the age of 18, in order to
control for differences in chronological age, all subjects were tested as closely as possible
to their birthdays: below one year, ± 2 weeks; and, from 18 months on, ± 4 weeks. When
the participants were 25 years-old, 85% participated in the data collection. When they were
approximately 35 years-old, over 90% participated in the data collection.
Measures
Table 1 shows the measures used in each study.
34
34
35
35
Characteristics
Sexual concern (ASCQ,
Hussey & Singer, 1993)
Very important peers
Depression; Anxiety;
Anger; Dissociation;
Posttraumatic stress
(TSCC, Briere, 1995)
Internalizing problems (Nurmi,
1993; Radloff, 1977;
Rosenberg, 1979)
Adolescence
Psychopathy-like personality
traits (YPI, Andershed, Kerr,
Stattin, & Levander, 2002)
Support and trust in
relationships with peers;
Conflicts in relationships with
peers (FQQ, Parker & Asher,
1993)
Relationships with peers
Psychosocial symptoms
Social adjustment
Negative parenting practices
Emotional aggression;
Physical aggression
(CTS, Straus, 1995)
Study II
Harsh or inept
parenting
Study I
Table 1. The measures that were used in five studies
Youth’s warmth and
closedness
Negative behavior in
the family
Adolescence
Delinquency
Delinquency
Problem behaviors
Childhood
Early unmanageability
(Stattin, Janson, KlackenbergLarsson, & Magnusson, 1995)
Conduct problems (Stattin,
Janson, Klackenberg-Larsson,
& Magnusson, 1995)
Problem behaviors
Childhood
Early unmanageability
(Stattin, Janson, KlackenbergLarsson, & Magnusson, 1995)
Relationships quality
Relationships with partner
Striking; Beating (Stattin,
Janson, Klackenberg-Larsson,
& Magnusson, 1995)
Striking; Beating (Stattin,
Janson, Klackenberg-Larsson,
& Magnusson, 1995)
Monitoring efforts
(Kerr & Stattin, 2000)
Maternal rejection
Discordant relationships
(Stattin & Klackenberg, 1992)
Study V
Discordant relationships
(Stattin & Klackenberg, 1992)
Study IV
Parents’ “gut-level”
reactions
Study III
Harsh or Inept Parenting
Studies I, II, III, IV, and V featured several aspects of harsh parenting – harsh,
negative parenting practices, such as anger outbursts, coldness-rejection, or discordant
mother-child relationships, and corporal punishment, such as striking, beating, or physical
aggression – or inept parenting, i.e., parents’ behaviors that might underlie negative
parenting practices, such as worry, distrust, or monitoring.
Emotional or Psychological Aggression
For the measure of emotional or psychological aggression, or what was termed
abuse in Study I, the Conflict Tactics Scale (CTS) by Straus (1995) was used. This widely
utilized scale assesses the extent of emotional abuse or aggression a child reports on having
experienced within the past year. The stem statement for all the items was: “Please indicate
how often your parents did each of these things in the past year.” The original scale was
modified for the study, with several items added to the emotional abuse, aggression scale
(e.g., “tried to make you feel guilty”). The rating of the items was changed to a 5-point
Likert scale (never 1 to always 5). The final version included 23 items (together with
physical aggression items). Two initial questions asked the child to report on positive
parental behaviors. Emotional abuse, psychological aggression (11 items) was assessed
with items such as “insulted you,” “tried to make you feel guilty,” “made you feel like you
were a bad person,” and “sulked or refused to talk about an issue.” Alpha reliability
estimates, based on the present samples in each of the four countries, ranged from .79 to .83
for the emotional abuse, aggression scales.
Negative Parenting Practices
To measure some aspects of harsh parenting, described in Study II as negative
parenting practices (including angry outbursts and coldness-rejection), youths’ responses to
11 statements about how their parents typically responded to wrongdoing were used. There
were three response options, ranging from “never” to “most often”. There were 5 items for
angry outbursts and 6 items for coldness-rejection, and youths responded to each item for
their mother and father separately. For these analyses, the mean of angry outbursts and
coldness-rejection was used. Also, because reports for mothers’ and fathers’ behaviors were
substantially correlated (r (580) = .68, p < .01 and r (582) = .62, p < .01 for angry outbursts
and coldness-rejection, respectively), reports from both parents were combined. The stem
question for all of these items was: “What happens if you have done something your parent
really dislikes?” Youths rated angry outbursts statements were: “Becomes very angry and
36
36
has an outburst,” “Has outbursts of anger and tells you off,” “Has a hard time controlling
his or her irritation,” “Quarrels and complains loudly,” “Screams and yells at you.” The
alpha reliabilities for this scale were .87 at Time 1 and .89 at Time 2 in Study II. Youth also
rated coldness-rejection statements: “Ignores you if you try to explain,” “Doesn’t talk to
you for a long while,” “Is silent and cold towards you,” “Doesn’t listen to your opinions or
explanations,” “Makes you feel guilty for a long time,” “Avoids you.” The alpha
reliabilities for this scale were .77 at Time 1 and .83 at Time 2 in Study I. These scales were
significantly correlated with each other, both at Time 1 (r = .70, p < .001), and at Time 2 (r
= .62, p < .001).
Discordant Mother-Child Relationships
In Studies IV and V, interviewers’ ratings were used to describe discordant
relationships between mothers and children. Interviews were conducted annually. After
each interview occasion, from child ages 6 to 12 years, the interviewers made judgments of
the quality of mother-child relationships based on their experiences during the interview
and mothers’ responses to interview questions. The interviewers used a 3-point Likert scale
with scale values: (1) “good,” (2) “indifferent,” and (3) “bad.” The term “bad” was used for
cases characterized by pronounced conflict or by apparent insensibility. The term
“indifferent” was used where there was some evidence of conflict, disagreement, or
insensibility, or where relationships were changing back and forth – at times good and at
times bad (see Stattin & Klackenberg, 1992, for a description of this measure). All the
questions were posed to mothers only.
Maternal Rejection
For the measure of mother’s rejecting behavior in Study IV the (grown-up)
children’s retrospective reports given at age 25 were used. Participants were instructed to
think about how they perceived their mothers when they were 12 years of age or younger,
and to evaluate the following statements: “My mother demanded more of me than one
should from a child,” “My mother was really interested in what I did and how I felt”
(reversed), “My mother had very few rules for me,” “My mother made me feel wanted and
needed” (reversed), “My mother nagged at and quarreled with me when I behaved badly,”
“My mother made me feel that what I did was important to her” (reversed), “My mother did
not spend more time with me when it was necessary,” “My mother encouraged me to take
my own initiatives” (reversed), “My mother did not want to me to bring my friends home,”
37
37
and “My mother spoke to me in a warm and affectionate manner” (reversed). There were
four response options, ranging from “does not apply at all” to “applies perfectly.” Higher
scores indicate more rejecting behavior. The alpha reliability for this scale was .82.
Physical Aggression
For the physical aggression measure in Study I, the Conflict Tactics Scale (CTS) by
Straus (1995) was used. The stem statement for all the items was: “Please indicate how
often your parents did each of these things in the past year.” The ratings of the items were
on a 5-point Likert scale (never 1 to always 5). The final version included 23 items
(together with emotional aggression). Two initial questions asked the child to report on
positive parental behaviors. Physical abuse, aggression (10 items) was assessed by
inquiring about potential parental behaviors, ranging from less severe to more severe forms
of physical abuse, such as “threw something at you,” “slapped or spanked you,” to “beat
you up,” and “burned or scalded you.” Alpha reliability estimates, based on the present
samples in each of the four countries, ranged from .79 to .87 for the physical abuse,
aggression scales.
Striking and Beating
For the measures of harsh parenting in Studies IV and V, mothers’ reports of their
striking (milder corporal punishment) and beating (stronger corporal punishment) were
used, which were given on each assessment from 6 to 12 years. For analysis, mean scores
for striking and beating for all the age periods were created. Each mean score represented
the average level of each variable over this period. All the questions were posed to mothers
only.
Striking. To measure striking in Studies IV and V, mothers answered whether or
not (and how often) they struck their children. When children were 6 to 9 years, there were
six response options, ranging from “never” to “on many occasions every day.” Because of a
lower incidence of striking of older children, the response options were changed from age
10 to age 12; there were five response options, ranging from “never” to “daily.”
Beating. To measure beating in Studies IV and V, when the child was 6 to 9 years,
mothers answered whether or not they had given their child a real beating. For children
ages 10 to 12 years, mothers answered about beating frequency. There were five response
options, ranging from “never” to “once a day.” To keep consistency in measures over time
this measure was dichotomized.
38
38
Inept Parental Behavior
For the measures of inept parents’ behavior that might underlie negative parenting
practices, or even harsh parenting, in Study III parents’ reports of their worries, distrust,
and control were used.
Parents’ “Gut-Level” Reactions. This label for emotionally tied reactions, such as
worries and distrust, that might underlay negative parenting was used in Study III. Gutlevel reactions are the mean of two scales. Worries was a six-item scale. Parents responded
to questions such as “Are you worried that your child will not make it in school?” “Are you
worried that your child will end up in bad company?” and “Do you worry about what your
child is doing together with friends during evenings and weekends?” The alpha reliability
was .88 at Time 1 and Time 3. Trust (reversed) was a six-item scale. Parents responded to
questions such as: “Do you trust that your child does not enter into bad company?” and “Do
you trust that your child does not do anything dumb in his or her free time?” The alpha
reliability for this scale was .80 at Time 1 and .81 at Time 3. The correlation between the
two scales was .44 (p < .001).
Monitoring Efforts. This measure in Study III was composed of the items from two
scales – control and solicitation – which had been developed previously to measure parents’
active monitoring efforts (Kerr & Stattin, 2000). The scales tapped parents’ efforts to keep
track of their youth’s whereabouts and associations by requiring the youth to do things like
checking with parents before making plans to go out with friends (control), and talking to
the youth, the youth’s friends, and the friends’ parents in order to stay informed
(solicitation). Five items that assessed solicitation were: “This month, have you been in
contact with and talked to the parents of your child’s friends?” “How often do you talk to
your child’s friends when they come over to your house (ask what they do, how they think
and feel about different things)?” “During the past month, how often have you started a
conversation with your child about his or her free time?” “How often do you ask your child
to sit down and tell you what has happened during an ordinary day in school?” and “Do you
usually ask the child to tell about what happens in his or her free time (who he or she meets
in town, leisure activities, etc.)?” Five items that tapped control were: “Does your child
need to have your permission to stay out late on a weekday evening?” “Does your child
need to ask you before he/she can decide with his/her friends what they will do on a
Saturday evening?” “If your child has been out very late one night, do you require that
he/she explains what he/she did and whom he/she was with?” “Do you always require that
your child tells you where he/she has been at night, who he/she was with, and what they did
together?” and “Before your child goes out on a Saturday night, do you require him/her to
39
39
tell you where he/she is going and with whom?” The alpha reliability for the 10-item
monitoring strategies measure was .76 at Time 1 and .79 at Time 3.
Children’s and Adolescents’ Characteristics
Early unmanageability
For the measure of early unmanageability in Studies IV and V, temper tantrums
and resistance to control were used. Mothers’ ratings from 3 months to 3 years were
combined. When the child was 3-12 months, the questions posed were about anger-prone
temperament: “Does he/she often get angry?” “Does he/she often get extremely angry?”
The alpha reliability for this scale was .81. When the child was 18 months to 3 years
several age-appropriate unmanageability items were added: “Does he/she want to get
his/her own way?” “Is he/she often disobedient with you?” “Is he/she a noisy child?” “Is
he/she a destructive child?” The alpha reliability for this scale was .69.
Adolescents’ Psychopathy-Like Personality Traits
A youth self-report instrument designed to tap subclinical levels of these personality
traits in community samples of youths 12 years or older, the Youth Psychopathic Traits
Inventory (YPI) by Andershed, Kerr, Stattin, & Levander (2002), was used in Study II to
evaluate adolescents’ psychopathy-like personality traits. The ten subscales in this
instrument load on three separate factors, representing the dimensions that have been
verified in studies using clinical assessment procedures on adult forensic samples (e.g.,
Cooke & Michie, 2001). This instrument has been found to be reliable and construct valid
(Andershed, Hodgins, & Tengström, in press; Dolan, & Rennie, 2006a; Dolan, & Rennie,
2006b; Poythress, Dembo, Wareham, & Greenbaum, 2006; Skeem & Cauffman, 2003). For
Study II, a total YPI score, which was calculated as the mean of scores on the three
dimensions, was used. The Grandiose, Manipulative Traits dimension comprises 20 items,
equally divided among four subscales: Dishonest Charm, Grandiosity, Lying, and
Manipulation. Examples of the items are: “I have the ability to con people by using my
charm and my smile,” “I am better than everyone else,” “Sometimes I find myself lying
without any particular reason.” All the 20 items were averaged to create one dimension.
The alpha reliabilities were .85 at Time 1 and .85 at Time 2. The Callous, Unemotional
Traits dimension comprises 15 items from three subscales: Unemotionality,
Remorselessness, and Callousness. Some examples of the items are: “I think that crying is a
sign of weakness, even if no one sees you,” “I usually feel calm when other people are
40
40
scared,” and “I have the ability not to feel guilt and regret about things that other people
would feel guilty about.” All the 15 items were averaged. The alpha reliabilities for this
dimension were .74 at Time 1 and .79 at Time 2. The Impulsive, Irresponsible Traits
dimension includes 15 items for Impulsiveness, Thrill-Seeking, and Irresponsibility.
Examples of the items are: “I prefer to spend my money right away rather than save it,” “I
like to be where exciting things happen,” and “I have probably skipped school or work
more than most other people.” All the 15 items were averaged to create one dimension.
The alpha reliabilities were .77 at Time 1 and .77 at Time 2 in Study I. These three
dimensions were significantly and substantially correlated with each other at Time 1 and
Time 2 (rs from .54 to .78, ps < .001).
Adolescents’ Internalizing Problems
To measure youth internalizing problems in Study II, youths’ responses to
statements about self-esteem, depressed mood, and failure expectations were used. For
Study II a mean value of the three scales was used. These three scales were significantly
and substantially correlated with each other at both Time 1 and Time 2 (rs from .39 to .61,
ps < .001. To measure Self-Esteem, the Rosenberg Self-Esteem Scale (Rosenberg, 1979)
was used. Adolescents were asked to rate how well they were described by each of ten
statements. There were four response options, ranging from “do not agree at all” to “agree
totally.” The statements were: “In general, you are satisfied with yourself,” “Sometimes
you think that you are not useful for anything,” “You think that you have many good
characteristics,” “You manage to do things as well as most others do,” “You think that you
do not have a lot to be proud of,” “You feel really useless from time to time,” “You think
that you are worth a whole lot, at least as much as anyone else,” “You wish you could have
better thoughts about yourself,” “In general, it is easy for you to feel unsuccessful,” and “In
general, you see yourself as positive.” The items were reversed, when necessary, so that
high scores indicated low self-esteem. The alpha reliabilities for this scale were .89 at Time
1 and .89 at Time 2. Questions for measuring Depressed Mood in Study II were taken from
the Child Depression Scale from the Center for Epidemiological Studies (Faulstich, Carey,
Ruggiero, Enyart, & Gresham, 1986; Radloff, 1977; Roberts, Lewinsohn, & Seeley, 1991;
Schoenbach, Kaplan, Grimson, & Wagner, 1982; Weissman, Sholomskas, Pottenger,
Prusoff, & Locke, 1977). The measure includes twenty items, with three response options
ranging from “not at all” to “often.” Examples are: “Couldn’t feel happy, even if my family
or friends tried to cheer me up,” “Felt ‘down’ and unhappy,” “Felt like I wanted to cry,”
“Felt sad,” and “Thought that others didn’t like me.” The alpha reliabilities for this scale
41
41
were .89 at Time 1 and .91 at Time 2. Youths reported on their Expectations of Failure on
difficult tasks, a construct that is related to self-esteem, but not synonymous with it (Nurmi,
1993; Nurmi, Onatsu, & Haavisto, 1995; Nurmi, Salmela-Aro, & Ruotsalainen, 1994).
They evaluated four statements, with four response options, ranging from “do not agree at
all” to “agree totally.” The statements were: “I don’t have faith in my ability to cope with
hard tasks,” “I easily become uncertain when I face new tasks,” “Often I don’t even think
there is any point in trying when I face demanding tasks,” and “The feeling that it’s hard for
me to cope with things makes me not do as well in school as I could do.” The alpha
reliabilities for this scale were .74 at Time 1 and .71 at Time 2.
Negative Behavior in the Family
A measure of youths’ negative behavior in the family for Study III was formed as
the mean of three scales – defiance, disclosure (reversed), and off-task behavior. It was
designed to evaluate youth behavior at home. Defiance was a three-item scale. Parents
responded on 4-point Likert scales from “does not apply at all” to “applies exactly.” The
items were: “Often does things although we say several times that it is not allowed,” “You
often need to tell him/her several times when he/she has done something wrong to get
him/her to stop,” and “Usually it is sufficient to rebuke him/her one time to stop him/her
from doing something that he/she is not allowed to do (reversed).” The scale had an alpha
reliability of .82 at Time 1 and .83 at Time 3. Disclosure comprised five items. Parents
reported on their child’s disclosure of information about daily activities, with questions
like: “Does your child hide a lot from you about what he/she does during nights and
weekends?” “Does your child talk at home about how he or she is doing in the different
subjects at school,” and “Does your child keep a lot of secrets from you about what he or
she is doing during his or her free time?” The alpha reliability was .81 at Time 1 and .78 at
Time 3. Off-task behavior was an eight-item scale taken from a revised Strategy-Attribution
Questionnaire (Nurmi, Salmela-Aro, & Ruotsalainen, 1994). Parents responded on a 4point scale from 1 (totally disagree) to 4 (totally agree). Some examples are: “It is too easy
for him/her to think of other things, day dream or become lost in thought when he/she
should concentrate on more important tasks,” “He/she often finds other things to do when
solving a difficult problem,” and “ If a hard task comes up, he/she quickly chooses to do
something else.” The alpha reliability at both times was .90. The mean inter-correlation
between these three scales was .46, p < .001(range = .44 to .49).
42
42
Youth’s Warmth and Closedness
To measure youth characteristics or behavior in Study III youth warmth and
closedness measures were used. The warmth measure parallels what youths were asked
about their parents’ expressions of Emotional Warmth. Four items were used. Parents were
asked whether their child: “Often says or does something nice without an obvious reason,”
“Does small things to show tenderness (e.g., hugs, smiles),” “Says that he or she is proud of
us,” and “Shows that he or she likes us without a reason, almost regardless of what we do.”
Responses were given on a 4-point scale ranging from “does not apply at all” to “applies
exactly”. The alpha reliability for the scale was .77 at Time 1 and .80 at Time 3. The degree
to which the youth seemed Closed to parents’ influence was measured with five items.
Parents rated the following statements on a 4-point scale ranging from “does not apply at
all” to “applies exactly”: “Our child keeps his/her feelings to him/herself when he/she is
worried or upset, “Our child prefers to comfort him/herself,” “Our child doesn’t seem to
think about keeping track of where he/she can reach us,” “Our child does not show who
he/she really is,” and “Our child keeps his/her feelings to him/herself after we have been
apart for a week or more.” The alpha reliability for the scale was .78 at Time 1.
Youth Delinquency
For the evaluation of youth delinquency in Study III, as one of the youth
characteristics to which parents might react, self-reported delinquency was used. Youthreported delinquency was measured with 21 questions about shoplifting, being caught by
the police, vandalizing public or private property, taking money from home, creating
graffiti, breaking into a building, stealing from someone’s pocket or bag, buying or selling
stolen goods, stealing a bike, being in a physical fight in public, carrying a weapon, stealing
a car, stealing a moped or motorcycle; using marijuana or hashish, and using other drugs.
The alpha reliability at Time 1 was .92.
Social Adjustment Problems
Psychosocial Symptoms
To evaluate psychosocial symptoms among children and youth in Study I, children
and youth answered questions about various symptoms or sexual concerns that they had
experienced in everyday life situations.
Symptoms. Psychosocial symptoms were assessed using the Trauma Symptom
Checklist Children (TSCC) by Briere (1995), a 54-item 4-point scale (0 never to3 almost all
43
43
of the time). The items are grouped into six subscales: Anxiety, Depression, Anger,
Posttraumatic Stress, Dissociation, and Sexual Concerns. Mean scores were calculated for
each subscale. Alpha reliability ratings were acceptable to good for the data sets of the four
countries: Depression, .75–.82; Anxiety, .76–.81; Anger, .74–.80; Dissociation, .71–.78;
Posttraumatic stress, .74–.84; and Sexual Concerns, .69–.76.
Sexual Concerns. Adolescents completed the Sexual Concerns Questionnaire
(ASCQ) by Hussey and Singer (1993). This 31-item scale was administered to 7th grade
participants only, since it is meant specifically for adolescents. The items were rated on a 4point scale (“never” 0 to “almost all of the time” 3). The Sexual Concerns Questionnaire
rated somatic problems (5 original items, plus two items added for the purposes of this
study) because clinical experience in several of these countries had indicated markedly
elevated rates of post-traumatic somatic complaints – sexual concerns (14 items), and
relationship issues (10 items).
Adolescent Problem Behavior
Adolescent Conduct Problems. For the measures of conduct problems in Study IV,
behaviors that are included as diagnostic features of Conduct Disorder, Oppositional
Defiant Disorder, or both in the Diagnostic and Statistical Manual of Mental Disorders (4th
ed., [DSM-IV] American Psychiatric Association, 1994) were used. When the child was 15
or 16 years, behavior problems were measured by asking parents the following questions:
“Is he/she defiant when rebuked?” “Can you trust him/her not do things which he/she
should not do?” (reversed), “Does he/she break things willfully?” “Argue to have his/her
own way?” “If you (Mother) chastise him/her, does he/she seem to mind?” (reversed), “If
you (Mother) chastise him/her, will he/she do the same thing again the same day?” “If
father chastises him/her, does he/she seem to mind?” (reversed), “If father chastises
him/her, will he/she do the same thing again the same day?” “Does he/she stay out without
your approval?” “Is he/she disobedient on purpose?” “Does he/she tell fibs to get out of
trouble?” “Does he/he take things he/she knows he/she should not have?” “Does he/she get
really furious?” and “Does he/she get irritated over trifles?” There were five response
options, ranging from “never” to “always” to all the questions. Higher scores indicate more
problem behavior. The alpha reliability for this scale was .89.
Adolescent Norm Violations. In addition to mother-reported conduct problems, in
Study IV, youths’ and interviewers’ reports of behaviors that are listed as associated
features of Conduct Disorder, Oppositional Defiant Disorder, or both in the DSM-IV
(American Psychiatric Association, 1994) were used. They are truancy, alcohol drinking,
44
44
drug use, and early initiation of sexual intercourse. For the analysis, a mean score of these
four standardized items was created. The alpha reliability for the four items was .71. This
adolescent- and interviewer-rated norm violations measure was significantly correlated
with mother-rated conduct problems at 15-16 years, r = .32, p < .001.
The Truancy measure was a judgment made by those who interviewed the
adolescents and their parents when the adolescents were 16 years-old. Based on
information from both the youths and their parents, the interviewers rated truancy on a
four-point scale ranging from: (1) “no sign of being tired of school or truancy” to (4) “a lot
of truancy from school.” Alcohol Drinking at age 18 was measured by total alcohol
consumption per month (beer, wine and spirits). The questions were “How much wine do
you drink per month?”, “How much spirits do you drink per month?”, “How much beer do
you drink per month?” The response scale for wine consumption was (1) none to (9) 10
glasses per month; the response scale for spirits consumption was (1) none to (9) 18 glasses
(15cl) per month; and, the response scale for beer consumption was (1) none to (9) more
than 80 45cl cans per month. At the age of 17, youths reported on their prior Drug Use
(hash, amphetamines, LSD, opium, and other drugs). A composite measure of drug use was
formed on a 7-point scale, based on the frequency of use of any of these drugs. At age 25,
participants answered the question at what age they had their First Intercourse. There were
fifteen response options ranging from “at the age of 11” to “at the age of 25.”
Delinquency. Youth-reported delinquency in Study III was measured by 21
questions: about shoplifting, being caught by the police, vandalizing public or private
property, taking money from home, creating graffiti, breaking into a building, stealing from
someone’s pocket or bag, buying or selling stolen goods, stealing a bike, being in a
physical fight in public, carrying a weapon, stealing a car, stealing a moped or motorcycle,
using marijuana or hashish, and using other drugs. The alpha reliability at Time 3 was .93.
Relationships with Peers during Adolescence
Assessment of Peers. To evaluate relationships with peers, a definition of peers was
given to adolescents in Study II. Youths were asked about their important peers, which we
defined as follows: “Someone you talk to, hang out with, and do things with. It cannot be
your parents or another adult. It could, for example, be a friend, a sibling, or a boyfriend or
girlfriend.” It was explained, further, that these important peers could live anywhere, did
not have to be the same age as the participant, and could be either boys or girls.
Adolescents were asked to name four important peers in order of importance, and then to
rate different aspects of their relationships with their most important peer, or the one they
45
45
named first. In Study I, there was a focus on the most important peer. For 77% of
participants, the most important peer was a friend, for 11% it was a sibling, and for 7% it
was a romantic partner. Adolescents who named a sibling as their most important peer at
Time 1 (73 participants) or at Time 2 (40 participants) were excluded from the analyses to
keep parent and peer relationships independent. For most participants, the most important
peer at Time 1 was different from the most important peer at Time 2. Only 211 youths
named the same person at both times.
Relationships with Peers. Youths answered 12 questions concerning their
relationships with their first-mentioned, or most important, peer in Study II. There were
five response options, ranging from “do not agree at all” to “agree perfectly.” Questions
were taken from Parker and Asher’s (1993) Friendship Quality Questionnaire. The
questions were about conflicts and caring in relationships or the friend’s behavior in the
relationship. Principal-components analyses of the 12 variables showed two clear factors,
which we labeled “support and trust” and “conflict.” The factor loadings ranged from .64 to
.90 and the cross-loadings ranged from -.01 to -.20. Concerning support and trust, youths
rated the following items about their most important peer’s behavior: “Says I’m good at
different things,” “Says that I’m pretty smart,” “Makes me feel that I have good ideas,”
“Sticks up for me if others talk about me behind my back,” “Says ’I’m sorry‘ when he or
she has hurt my feelings or been mean,” “Would like me even if nobody else did,” “Keeps
his or her promises,” and “Doesn’t give away my secrets to others.” The alpha reliabilities
for this scale were .86 at Time 1 and .89 at Time 2. Concerning conflicts in the
relationships youths evaluated four statements about conflict: “We often get angry with
each other,” “We argue a lot,” “We often get annoyed with each other,” and “We fight a
lot.” The alpha reliabilities for this scale were .90 at Time 1 and .91 at Time 2 in Study I.
Relationship Quality with a Partner
For the measure of romantic relationships quality in Study V, participants evaluated
10 items describing the relationship with their romantic partner at age 35. The questions
were: “Does your partner talk with you about his/her problems?” (reversed), there were
four response options, ranging from “yes, always” to “never;” “How warm do you feel
towards your partner?” (reversed), there were five response options, ranging from “very
much” to “not at all;” “How do you get along with your partner?” there were five response
options, ranging from “bad” to “very good;” “How often do you really get mad at your
partner?” (reversed), there were five response options, ranging from “never” to “often;”
“How would you describe your husband/wife?” there were five response options, ranging
46
46
from “only negative features” to “only positive;” “Do you and your husband/wife have any
interests in common?” there were five response options, ranging from “no interests in
common, no possibilities for recreation together” to “totally share each others activities
always with the same pleasure;” “If you would give a picture of your relationships how
would you describe the atmosphere at home?” there were six response options, ranging
from “very disharmonic, divorce atmosphere” to “very harmonic, we share the same
attitude, open, warm home atmosphere;” “How often do you cuddle? If you would think
one month back, how often did you spontaneously kissed or hugged each other during last
month?” there were five response options, ranging from “not at all” to “daily, almost
daily;” “How is your sexual life? Are you well adapted sexually to each other? Do you
function well together?” there were six response options, ranging from “have no sexual life
or very seldom” to “very well adapted;” “Does you partner give you encouragement and
support when you have trouble at work?” (reversed), there were five response options,
ranging from “I get all the help I need” to “my partner is more of the obstacle.” For the
analyses, a mean score of all the items was created. For the latent profile analysis, all the
variables were standardized to create a common metric and created a scale from the
standardized items with higher scores indicating better relationship quality. The alpha
reliability for this scale was .81.
Background Variables
In Study I some analyses included background factors, such as a grouping variable
or a risk measure. Respondents were asked to indicate their age, sex, ethnicity, and number
of family household members. The research teams grouped the questionnaire responses as
coming from big-city, medium-city, or rural schools. The child also replied to two
questions related to potential risk factors for parents’ unemployment and alcohol abuse.
Participants were asked about their parents’ employment status with the question, “Does
your father (mother) work outside the home?” and about excessive alcohol use by a parent
with the question “Is there a person in your family who uses alcohol overly much?” The
format response for each question was “yes/no”.
47
47
48
III Results
Study I
There are not many studies that have examined cultural differences in the effects of
harsh parenting on adjustment problems, and these have mostly been concerned with the
impacts of physical punishment on externalizing behaviors (e.g., Deater-Deckard, et al.,
1996; Dornbush, et al., 1987; Gunnoe & Mariner, 1997). Existing studies suggest that
physical punishment is used in some cultures more than others. It is also suggested that
physical punishment may increase children’s externalizing problems or affect academic
achievement among some children but not others. The question still remains whether harsh
parenting, such as emotional or physical aggression, may have different effects on
adjustment problems, e.g. various psychosocial symptoms, in cultures that are not covered
by the existing literature; as noted, most of the research has been performed in North
America. To partly address this question, data were collected in some Baltic and Eastern
European countries. The goals of this study were to examine the incidence of emotional and
physical aggression, abuse in Latvia, Lithuania, Macedonia, and Moldova, and to examine
the relationship between emotional and physical aggression and the level of psychosocial
symptoms reported by children in each of these countries.
Many children in Latvia, Lithuania, Macedonia, and Moldova are exposed to
emotional or physical abuse, and aggression from their parents. 33% of the Lithuanian
children, 32% of the Moldovan children, 29% of the Latvian children, and 13% of the
Macedonian children reported experiencing emotional and physical abuse, and aggression.
Analyzing the incidence of experienced abuse and aggression by school grade revealed that
Latvian and Moldovan 7th grade children reported a higher incidence of emotional abuse
than 4th graders; however, Lithuanian 4th graders reported more physical abuse than 7th
graders. The most common form of emotional abuse or aggression reported by Lithuanian
children and Macedonian children is “yelling”. Latvian and Moldovan children reported
that their parents used the tactic of making them “feel guilty” (see Table 3, Study I). The
mean scores on the emotional and physical abuse, aggression ratings were markedly lower
for Macedonian children than children from the other countries. Thus, children from
different countries may have similar experiences, although in some countries children
experience different forms of, and more, parental aggression than others.
To determine whether experiences in the family may have associations with
adjustment problems, the relationships between reported emotional and physical aggression
and psychosocial symptoms were examined. First, on comparing the mean differences for
49
49
various symptoms between countries it was found that the symptom scores were highest for
Latvia and Lithuania, and lowest for Macedonia. A similar pattern was noted for somatic
problems. Turning to the relations between experienced abuse or aggression, and various
symptoms, the strongest relations were typically between emotional or physical abuse and
aggression and children’s anger; the highest correlation between anger and reported
physical abuse or aggression was found for Macedonia (r = .54, p < .001). Also, significant
correlations were found between various psychosomatic symptoms, such as depression,
dissociation, anxiety, sexual concerns, post-traumatic symptoms and somatic problems, and
the combined abuse mean score, with the highest correlation found for Macedonian
children (r = .69, p<.001, see Table 5, Study I). Comparing the magnitude of the
relationships between experienced abuse and psychosocial symptoms (using Fisher’s Z
transformation for independent samples) revealed that there are significant differences
between the countries. For example, the relation between Lithuanian children’s experienced
abuse and anger and the relation between Macedonian children’s experienced abuse and
anger are significantly different. Although, experienced abuse or aggression is related to
various problems in different countries, the magnitude of the relation seems to differ
between countries and cultures.
This study showed similar results to several previous studies suggesting that, in
some cultures and countries, children experience more parental aggression than in others
(see e.g., Deater-Deckard et al., 1996). Nevertheless, the current study showed that
experienced harsh parental behavior is associated with various problems for children from
different countries in a similar way. It might be that children from different countries and
cultures have similar problems if they experience harsh treatment from their parents,
although ethnic groups were not taken into account in the current study. The samples were
homogeneous, and it was not possible to examine the influence of ethnicity. Even though
this study has shown that children from different countries have similar problems, such as
depression, anger or the like, it cannot answer questions concerning the directions of
effects. To address the question of whether parents’ behaviors influence various problems
or various problems evoke parents’ behaviors requires longitudinal investigation.
Study II
Study I showed that parents’ aggression, either verbal or physical, is related to
various symptoms, such as depression, anxiety, or anger, in a number of countries. It seems
that children and adolescents in different countries may have very similar psychosocial
50
50
symptoms, although there are differences in the magnitudes of relations between parents’
aggression and various symptoms. But there remains the question of how these
psychosocial, or similar, problems and harsh or inept parenting are related over time. And
the interesting question is what are the direction of effects, e.g. whether parents’ harsh or
inept behavior causes internalizing problems, or whether the presence of these problems
evokes harsh or inept parental behavior. Study I was cross-sectional, and could not address
these questions.
Study II considered whether youth characteristics, such as internalizing problems,
and psychopathy-like personality traits might influence negative parenting practices over
time. The main purpose of Study II was to examine whether links between negative
parenting practices and adolescents’ peer relationship quality might be explained by such
internalizing problems and psychopathy-like personality traits.
To address the question of whether youth characteristics might evoke negative
parenting practices, youth characteristics were examined in relation to negative parenting
practices over time. Both internalizing problems and psychopathy-like personality traits
significantly predicted increases in perceived negative parenting practices from Time 1 to
Time 2 ( = .17, p < .01, = .09, p < .01, for internalizing problems and psychopathy-like
personality traits, respectively). Negative parenting practices predicted increases over time
in internalizing problems ( = .18, p < .001), but did not predict changes over time in
psychopathy-like traits ( = .01, p > .05). Negative parenting behavior may affect later
internalizing problems, but has little to do with personality traits. Thus, there are some
reciprocal relations between self-perceived negative parenting practices and youth
characteristics.
Given that negative parenting behavior has been linked in previous studies to
various problems in relationships with peers (e.g., Dekovi & Meeus, 1997; Lansford, et
al., 2003; Vissing, et al., 1991), the study examined whether negative parenting practices
are linked to peer relationship quality. Longitudinal results have shown that negative
parenting practices at Time 1 significantly predicted self-perceived conflict in peer
relationships at Time 2, after controlling for the Time-1 measure ( = .19, p < .01). This
suggested that perceived negative parenting practices contribute to changes over time in
perceived conflict in close peer relationships.
Then, to address the question whether the youth characteristics that might evoke
negative parenting practices may interfere with relationship quality, it was looked at
whether and how youth characteristics are related to relationships with peers. Over time,
both internalizing problems and psychopathy-like personality traits significantly predicted
51
51
increases in perceived conflict in relationships with important peers. Also, there is evidence
that internalizing problems and psychopathy-like traits might evoke negative parenting and
create conflict in peer relationships, thus providing a possible explanation for the link
between self-reports of negative parenting and peer relationship quality.
To determine whether youth characteristics explain the link between negative
parenting practices and peer relationship quality, youth characteristics were controlled for
in models designed to assess changes in peer relationship quality over time on the basis of
negative parenting practices. There were no unique links between negative parenting
practices and self-perceived relationship quality after controlling for youth characteristics,
but internalizing problems predicted increases over time in perceived support and trust in
peer relationships ( = .09, p < .05); further, psychopathy-like traits uniquely predicted
increases over time in perceived conflict in relationships ( = .10, p < .05). Thus, youth
characteristics explained the longitudinal links between negative parenting and peer
relationship quality. Also, the results suggested that psychopathy-like personality traits play
a role in evoking negative parenting and creating peer relationship problems.
Children who experienced negative parental behavior have been found in previous
studies to have problems relating to peers. Although several longitudinal studies have
suggested interplay between youth characteristics, positive or negative parenting, and
different aspects of peer relationships (Clark & Ladd, 2000; Keown & Woodward, 2006;
Simons et al., 2001), youth characteristics, such as internalizing problems psychopathy-like
personality traits, have not been considered as possible explanations for either negative
parenting practices or poor peer relations. The findings of this study suggested not only that
negative parental behavior may influence some problems and behaviors, but also that some
problems and characteristics may evoke negative parents’ behavior. Further, these same
characteristics and behaviors might be responsible for problematic peer relationships.
Study III
Study II showed not only that parents affect youth internalizing problems, but also
that youth internalizing problems and personality traits may influence parents’ behaviors,
and also children’s later adjustment problems over time. Given that it was the adolescents
who evaluated their parents’ behaviors and their own characteristics, these results might be
subject to bias. The main question addressed by Study III was how parents and youths
respond to each other over time. To test whether the results of Study II might be biased
required analysis of parents’ reports and youth self-reported information. Study II showed
52
52
that parents respond to youth internalizing problems and personality traits, but a further
issue is whether parents respond to youth delinquent behavior. Longitudinal data were used
to address this question.
First, it was asked whether parents’ reactions to delinquency and negative behavior
at home in the sample of 10 to 14 year-olds were similar to those found earlier in families
of 14 year-olds only (Kerr & Stattin, 2003). The cross-sectional results were remarkably
similar (see Figure 4.1, Study III) and the question of directions of effects over time
remained. To infer directions of effects from this age-heterogeneous sample, cross-lagged
paths between all variables used in the previous model were examined. Children’s and
parents’ behaviors were moderately stable over time, and many of the cross-paths in the
model were found to be significant (² = 3.911, df = 3, p = .270, CFI = .999, TLI = .996,
RMSEA = .013). The results (see Figure 4.3, Study III) suggested that parents react with
distrust and worry to secretive, defiant, and off-task youth behavior at home. At the same
time, in response to both negative behavior at home and youth delinquency, they seem to
make fewer efforts to track what the youth is doing away from home. It seems that when
parents were faced with adolescent problem behaviors, they reacted in ways that were
unlikely to make the situation better. Thus, similarly to what was suggested by the crosssectional findings, these longitudinal results showed that the more problem behavior youths
engage in, the more parents experienced worry and distrust, and the less they monitored.
Also, the findings suggested that gut-level reactions predicted increased negative behavior
at home and delinquency over time. Monitoring efforts were not found to be significantly
related to changes in problem behaviors. According to these results, it seems that a youth’s
problem behaviors may affect how parents act toward the youth, in terms of both emotional
reactions and monitoring efforts, but it is only the emotional reactions that have an effect on
the youth. It seems that emotional context affects adjustment problems more substantially
later in life than certain parental behaviors. Taken together, youths’ delinquency and
negative behaviors seemed to affect parents in different ways. It is natural that when a
youth is secretive at home and has engaged in illegal acts, most parents begin to worry and
distrust the youth. However, it is difficult to understand why most parents do not try to take
the situation in hand and monitor the youth in order to limit opportunities for further
delinquency. This study suggested some explanations.
Turning to why parents would decrease rather than increase their monitoring efforts
when faced with adolescent problem behaviors, such as being secretive or defiant, and not
concentrating on tasks such as school work, it was tested three possible explanations.
However, only one of these was supported by the data. This suggested that results about
53
53
defiant youth behaviors and why parents react to them by reducing their monitoring efforts
may be explained, in part at least, by general social-response patterns. When people show
that they have no interest in talking to others, the others tend to leave them alone, but when
these people are warm and open, the others tend to open up to them. It might be that parents
do a similar thing with their children. To test this idea, cluster analysis was employed, and a
four-cluster solution was found (see Table 4.4, Study III). The largest cluster was an
average cluster consisting of youths who were near average on both variables (warm-cold
and open-closed). There were two clusters similar in size – a warm-open cluster, consisting
of youths who were high on warmth and openness, and a closed cluster, consisting of
youths who were about average on warmth but highly closed. The smallest was the coldclosed cluster, comprising youths who were exceptionally low on warmth and exceptionally
closed. Parents’ monitoring efforts seemed to follow these youth clusters. Monitoring
efforts were highest for the warm-open youths and lowest for the cold-closed youths. Over
time, parents of the warm-open youths increased their monitoring relative to the rest of the
sample, whereas parents of the closed and cold-closed youths decreased their monitoring.
Thus, it appeared that parents’ monitoring efforts were very much influenced by youths’
social signals. If youths were warm and open, parents seem to feel free to keep track of
what they were doing; if not, parents seem to be hesitant in getting involved.
Taken together, youths’ delinquency and negative behaviors may affect parents’
behaviors in different ways. This study suggested that youths’ problem behaviors do affect
how parents act toward them over time, in terms of both emotional reactions and behaviors,
but it is only the emotional reactions that seem to have an effect on the youths. Why most
parents would not try to take the situation in hand and monitor their youth’s movements in
order to limit opportunities for further delinquency? The results suggest that the link
between youth behavior and inept parents’ behavior can be explained by social signals,
based on youth characteristics. It seems that youth characteristics play an important role in
explaining inept parents’ reactions and later adjustment problems.
Study IV
Study I showed that children from different cultures and countries who have
experienced harsh parenting may have very similar adjustment problems, although in Study
I the idea was not considered that other parenting behaviors might affect the consequences
of harsh parenting (Lansford, et al., 2004; Larzelere, et al., 1989; Rohner, et al., 1996;
Simons, et al., 1994). In Study IV different combinations of parenting behaviors were
54
54
investigated. Harsh parenting measures included striking and beating (measures of physical
punishment) and discordant relationships, defined as maternal insensitivity and motherchild conflict (which might communicate rejection to the child). Also, previous studies
have suggested that, as well as other parenting behaviors, child characteristics should be
taken into consideration (Lau, et al., 2006). Studies II and III suggested that the effects of
parents’ behavior on adjustment problems may be affected by various youth characteristics.
But previous studies have not examined very early characteristics of the child. Study IV
examined how distinct patterns of physical discipline and discordant relationships related to
early unmanageable temperament and later conduct problems and norm violations, and also
how – apart from these links – early unmanageability relates to later conduct problems and
norm violations. Accordingly, the hypothesis that temperamentally unmanageable children
who experience physical discipline in the context of discordant relationships will have more
behavior problems than those who experience physical punishment in the context of good
relationships was examined.
What kind of harsh parenting combinations could be observed? To address this
question, latent profiles, classes of mother’s striking, beating, and discordant relationships
at ages 6-12 were examined. The final model estimated four classes of mother’s harsh
treatment (see Figure 2, Study III). The largest normative class had low levels of all three
harsh parenting variables. The second largest class, a physical punishment class, had high
levels of mother’s striking, beating and low levels of relationship discord. The third class, a
discordant relationships class had high levels of discordant mother-child relationships, but
low levels of striking and beating. The final class, a harsh treatment class had high levels
of striking, beating, and discordant relationships. Thus, children experienced different
combinations of mother’s harsh treatment over ages 6-12. Some children experienced
striking and beating with no notable signs of discordant relationships with their mothers.
However, for some children, physical punishment occurred in the context of a conflictridden mother-child relationship.
Next, in this study, it was examined how distinct patterns of physical discipline and
discordant relationships related to early unmanageable temperament and later conduct
problems and norm violations, and also how – apart from these links – early
unmanageability related to later conduct problems and norm violations. Based on
theoretical assumptions, the latent classes were regressed on a covariate (unmanageable
temperament); distal outcomes (mother-reported conduct problems in one model and selfand interviewer-reported norm violations in the other) were added to the latent class model,
55
55
and problem behaviors (either conduct problems or adolescent norm violations) were
regressed on early unmanageability (see Figure 1, Study IV).
Does early unmanageability increase the risk of experiencing specific combinations
of harsh parenting? Children with high levels of early unmanageability were found to be
more likely to be in the physical punishment class than in the normative class (OR = 2.78, p
< .05, 95% CI – 1.49-5.21, OR = 3.05, p < .05, 95% CI – 1.32-6.96, for the models with
conduct problems and norm violations, respectively, as distal outcomes). Children with
high levels of unmanageability also showed a tendency to be in the harsh treatment class
rather than the normative class (OR = 1.78, p < .10, 90% CI – 1.02-3.10, OR = 1.36, p <
.10, 90% CI – 1.01-1.92, for models with conduct problems and norm violations,
respectively, as distal outcomes), but they were not at increased risk of being in the
discordant relationships class (OR = .85, 95% CI – .47-1.42, OR = .88, 95% CI – .50-1.57,
for models with conduct problems and norm violations, respectively). Thus, early
unmanageable temperament increased children’s risk of experiencing physical punishment
and, to a lesser extent, physical punishment in the context of discordant relationships later
in childhood.
As shown in Studies I, II, and III, harsh parenting is related to various adjustment
problems. But are different combinations of harsh parenting related to later adjustment
problems in different ways? Results from the model with conduct problems or norm
violations as outcome variable showed that children in the normative class had significantly
lower levels of conduct problems and norm violations in adolescence than children in the
other classes, and also that children who experienced harsh treatment were significantly
more likely than children in all of the other classes to have conduct problems and violate
norms in adolescence. Thus, children who experienced physical punishment in the context
of discordant mother-child relationships showed more conduct problems and norm
violations later on than those who experienced physical punishment alone; further, those
who experienced physical punishment alone had more behavior problems than those who
did not experience any harsh parenting.
It was also examined the extent to which children’s early unmanageability was
related to later conduct problems or norm violations beyond the risk associated with their
membership of the harsh parenting class. Results from the models with conduct problems
or norm violations as outcome showed that only among children in the physical punishment
class was high early unmanageability significantly related to risk of conduct problems or
norm violations. Among children in the harsh treatment class, where conduct problems or
norm violations were highest, there was no significant link between early temperament and
56
56
conduct problems, apart from the risk associated with harsh parental treatment. For the
harsh treatment group, unlike the physical punishment group, the experience seemed to
play a greater role than temperamental unmanageability in the development of problem
behavior. These group differences were consistent with the idea that physical punishment
has a different meaning in the context of good parent-child relationships than in the context
of discordant relationships.
Study V
Study IV showed that harsh parenting can occur with or without good or bad parentchild relationships, and these different combinations can affect youth problem behavior.
Children who experienced physical punishment in the context of discordant mother-child
relationships showed more conduct problems and norm violations later on than those who
experienced physical punishment alone. Study II showed that the effects of parents’
behavior on relationship quality during adolescence may be affected by various youth
characteristics. It is a question if relationship quality in adulthood might be also affected by
children’s characteristics and different combinations of harsh parenting in context of other
parenting behaviors. In Study V, it was examined how distinct patterns of physical
discipline and discordant relationships related to early unmanageable temperament and later
relationships quality, and also how – apart from these links – early unmanageability relates
to relationships quality with a romantic partner in adulthood. In Study V, the same harsh
parenting combinations (latent classes) as in Study IV were used. Study V extended Study
IV by looking at adulthood outcomes - the quality of romantic relationships at age 35, thus,
the covariates (unmanageable temperament and parents’ attitudes towards marital conflicts)
and the distal outcome (relationship quality with partner at age 35) were added to the latent
class model (Figure 1, Study V). The latent classes were regressed on early
unmanageability and parents’ conflicted marital relations, and then relationship quality with
age-35 partner was regressed on early unmanageability, marital conflicts, and the latent
classes (Log likelihood = -643.257; BIC = 1482.919, SSA BIC= 1365.696, AIC =
1360.513, entropy = .824, number or parameters = 37).
Does children’s early unmanageability increase the risk of experiencing specific
combinations of harsh parenting? Children with high levels of early unmanageability were
more likely to be in the harsh treatment class than in the normative class (OR = 1.96, p <
.05, 95% CI – 1.02-3.75). Children with high levels of unmanageability also showed a
tendency to be in the physical punishment class rather than the normative class (OR = 1.50,
57
57
90% CI – 1.04-2.18), but they were not at increased risk of being in the discordant
relationships class (OR = .84, 95% CI – .39-1.84). Thus, unmanageable temperament
increased children’s risk of experiencing physical punishment in the context of discordant
relationships and to a lesser extent physical punishment in the context of good relationships
later in childhood.
Are different combinations of harsh parenting related to relationship quality with a
partner at age 35? Results showed that children who experienced physical punishment had
significantly higher quality relationships with a partner in adulthood than children in the
other classes, and children in the normative class were significantly more likely than
children in the discordant relationships or harsh treatment class to have better relationships
in adulthood. Children who experienced discordant relationships and harsh treatment had
significantly lower quality relationships in adulthood than children in the normative and
physical punishment classes. Thus, children who experienced discordant mother-child
relationships either alone or with physical punishment had worse relationships later in life
than those who experienced physical punishment alone and those who did not experience
either type of harsh parenting.
It was also examined to what extent children’s early unmanageability was related to
later relationship quality apart from the risk associated with harsh parenting class
membership. Results showed that among children in the harsh treatment class, higher early
unmanageability was significantly related to having a poor relationship with a partner at
age 35. Notably, among children in the discordant relationships class, where quality of
relationships with a partner at age 35 was also lowest, there was no significant link between
early temperament and relationship quality apart from the risk associated with discordant
relations. For the normative and physical punishment classes, early unmanageability was
not significantly related to relationship quality and parents’ marital conflicts were not
significantly related to relationship quality. Thus, it was only for the harsh treatment group
that temperamental unmanageability seemed to play a role in the development of poor
relationships.
58
58
IV Discussion
Findings and previous research
The main purpose of this dissertation is better to understand the relations between
harsh or inept parenting and children’s characteristics in the prediction of adjustment
problems. Taken together, this dissertation provides some insights into the role of harsh or
inept parenting in the development of various socials adjustment problems. The main
lesson is that it is not a simple case of parents affecting the child’s or adolescent’s behavior.
To some degree, harsh or inept parenting can be understood as a response to the child’s
temperament or personality characteristics. Parents seem to respond emotionally to certain
characteristics or behaviors, and adopt harsh or inept parenting behaviors. This is not the
only explanation for harsh or inept parenting, but it is clear that the role of the child’s or
adolescent’s characteristics cannot be dismissed. Even though parents’ negative behaviors
may affect youth social adjustment, youth characteristics and behaviors can strongly
contribute to their own adjustment and to harsh or inept parenting. Thus, the links between
harsh or inept parenting and adolescent or adult social adjustment are likely to represent
complex, partly reciprocal processes of social influence and socialization.
The dissertation makes several contributions to knowledge in the harsh or inept
parenting arena. One major contribution is that it illuminates the active role of the child in
creating his or her relationship environment. In the developmental literature, many studies
have looked at the effects of parenting on children and youths, suggesting that the child can
not create his or her relationship environment, but youth characteristics have tended to be
neglected as possible explanations for parent-child relations. However, in recent years,
there has been a growing awareness that reciprocal processes are important for
understanding interactions between parents and children (e.g., Cohen & Brook, 1998;
Hastings & Rubin, 1999; Kandel & Wu, 1998; Kochanska, 1998; Snyder et al., 2005; Stice
& Barrera, 1995). Although reciprocal effects are now widely accepted in principle, they
are still too seldom included in research designs, and especially in harsh parenting studies.
Thus, this dissertation adds to the existing literature by making new suggestions concerning
how various youth characteristics and harsh or inept parental behaviors may affect each
other during adolescence.
Another contribution of the dissertation to knowledge about harsh parenting is to
illuminate the importance of parents’ negative emotional reactions to youths’ behaviors.
Previous studies have suggested that other parenting behaviors in which context physical
punishment takes place may determine the effects of physical punishment on a child’s
59
59
development (Deater-Deckard et al., 1996; Lansford et al., 2004; McLoyd & Smith, 2002;
Rohner et al., 1996). This dissertation suggests that parents’ negative emotional reactions
might be more important, not only more important than physical punishment, but also more
important than family management or monitoring efforts. This dissertation also suggests
that certain combinations of harsh parenting are more damaging than others. In addition,
the findings lend credibility to the idea, from the previous literature, that physical
punishment can have differential effects according to whether or not the child perceives the
parent as rejecting. Taken together, this dissertation indicates that parents’ negative
emotional reactions are more damaging than other types of parental behavior.
This dissertation sheds some new light on the role played by the cultural aspect of
harsh or inept parenting in the development of child and adolescent problem behavior.
There are not many studies that have explored cultural or ethnic group differences with
regard to the effect of harsh or inept parenting on adjustment problems. It seems that
cultural or ethnic expectations have a lot to do with whether children perceive parenting
behavior as harsh, and, consequently, whether it undermines their adjustment (DeaterDeckard et al., 1996; Dornbusch et al., 1987; Gunnoe & Mariner, 1997; Lansford et al.,
2004; Lansford et al., 2005; McLeod et al., 1994; Rowe et al., 1994; Spieker et al., 1999).
Most of the research, however, has been performed in North America. Although existing
studies suggest that physical punishment is used in some cultures more than others, and that
physical punishment increases children’s externalizing problems, or affects academic
achievement, for some but not others, this dissertation suggests that the processes and
mechanisms through which harsh parenting is related to adjustment problems are very
similar across cultures. These have not previously been considered in the harsh parenting
literature. Thus, the dissertation gives offers some new suggestions and extends previous
research concerning cultural expectations.
The dissertation also adds some new knowledge on the roles of harsh or inept
parental behaviors and children’s and youths’ characteristics in the development of various
aspects of youth social adjustment, where it focuses on several important aspects. One is
the development of socially appropriate behavior, which previous researchers have tended
to evaluate by looking specifically at problem behavior. There are studies that have
considered youth characteristics while examining relations between parenting and problem
behaviors (Colden et al., 1997; Lengua et al., 2000; Leve et al., 2005). However, these
studies have usually looked solely at temperamental characteristics and externalizing
problems, and have covered just this one aspect of social adjustment. But there are other
important aspects of social adjustment. One is the ability to develop and sustain
60
60
relationships with other people, peers or romantic partners. Several previous studies have
considered youth characteristics while examining relations between parenting and peer
relationships (Clark & Ladd, 2000; Hinshaw et al., 1997; Keown & Woodward, 2006;
Simons et al., 2001), but most have looked at children rather than adolescents, and not
assessed relationship quality from the youths’ or their peers’ points of view. There are also
previous studies that have combined youth or young adult characteristics and negative
parenting in predicting romantic relationships later in life (e.g., Capaldi & Clark, 1998;
Donnellan, Larsen-Rife, & Conger, 2005), but most have focused on adolescents and young
adults and the research questions did not lead to examining the influence of very early
temperament or harsh parenting combinations on middle adulthood romantic relationships.
This dissertation makes a unique contribution by using prospective measures of early
temperament and combinations of harsh parenting behaviors to predict social adjustment
over a longer period of time than has been done before.
Strengths and limitations
This dissertation has some limitations that should be acknowledged. One concern
regards the over-interpreting of results about the directions of effects. It is important to
distinguish between effects over time during a period of several years and the original
causes of behaviors. The dissertation deals mainly with adolescence and explained effects
over time during the adolescent period. Parents and children have histories of interactions,
however, and negative parenting practices may play a causal role in shaping people’s
characteristics earlier in life, i.e. during childhood. It is not known from this dissertation
how any of these youth characteristics developed in the first place or whether parental
behavior played a role. Harsh or inept parenting may have played a causal role in shaping
characteristics earlier in the child’s life. It is intuitively appealing to believe that parents
play a more active role in shaping the behavior of young children than adolescents, even
though in some studies young children are increasingly seen as active agents and partners in
the socialization process (e.g., Anderson et al., 1986; Dix et al., 1986; Huh et a.,, 2006;
Passman & Blackwelder, 1981). But whatever happened earlier, data from this dissertation
do reveal changes over time in adolescence, and also directional and bidirectional effects.
There is always a larger context that was not tapped in this dissertation. The
literature refers to many contextual factors, such as unemployment, social support, or
public policies, which may affect children’s later development or parent-child relations.
This dissertation does not consider the conditions that surround the family, e.g., parents’
61
61
marital relations, and parent, sibling, and peer characteristics, that may have a direct impact
on the way parents and youths act and react to each other, and on the way parents’ or
youths’ behaviors can influence later adjustment problems. Also, it may be that there are
gender differences in the relationships between parental behaviors and later adjustment that
have not been tapped. Nonetheless, the findings from this dissertation provide some new
information about processes inside the family and interactions between parents and
children.
Concerning individual studies in the dissertation, the reliance only on youths’ or
only on mothers’ reports can be seen as a limitation of certain studies, particularly because
adolescents sometimes reported on their own traits as well as parents’ behavior, and
mothers sometimes reported on their child’s characteristics and their own behavior. The
assessment of certain characteristics or parental behaviors could be driven by perceptions of
the reporter than by the child’s or mother’s own actual behavior. Even though some
scholars have argued that youths’ views of family interactions are the most accurate or
valid (e.g., Glasgow, Dornbush, Troyer, Steinberg, & Ritter, 1997), it is possible that
youths with certain traits misperceive parents’ behavior in systematic ways, and these
misperceptions explain the links between youth characteristics and negative parenting.
Some other scholars suggested that there could be several reasons why parental descriptions
of children’s behaviors or characteristics are not always valid (for discussions of this, see
Kagan, Snidman, McManis, Woodward, & Hardway, 2002; Seifer, Sameroff, Barrett, &
Krafchuk, 1994). Although it could be that the results were not affected by the fact that the
information was from either youth’ or mothers’ reports, in the studies some information
from other informants to avoid informant bias and to answer the major questions was
incorporated, as well. All in all, even though there should be awareness that some data
represent only youths’ or only mothers’ perceptions of characteristics and parenting, the
consistency of findings from all the studies taken together—those that rely on one rater and
those that use multiple raters—provides confidence that the results of the models and
mechanisms are not just due to rater bias, but do actually tap into developmental processes.
The studies in this dissertation have several strengths. One is that mainly
longitudinal data were used, which enables processes over time to be investigated. The
longitudinal data permitted examination of relations between inept or harsh parenting, very
early unmanageable temperament, and later problem behaviors or relationships in a more
elaborate manner than previously reported in the literature. An additional strength of the
dissertation is its reliance on community samples with high participation rates. Given that
previous studies suggest a cultural influence in harsh parenting effects on adjustment
62
62
problems, data from several countries that have not previously been covered by the harsh
parenting literature were used to examine the processes linking harsh parenting, adolescent
characteristics, and social adjustment problems. It seems that the processes and
mechanisms may well be very similar regardless of culture. The findings give confidence
that the results of this dissertation are not just due to culture, but that there are similarities
between cultures, although the magnitudes of the influences may differ. Even though the
studies have some limitations, they ware to some extent lessened with data from other
studies which gives confidence that the results tap into actual developmental processes.
What is harsh parenting?
In this dissertation, I define harsh parenting as parents’ physical punishment and
their verbal or nonverbal aggression, such as anger outbursts, threats, stony silences, or
rejection. Thus, both aspects of harsh parenting – physical and nonphysical – are combined.
Also, I speak of harsh parenting as a part of inept parenting. It is necessary to have a clear
definition for research purposes, but on a philosophical level, it is discussible whether a
parenting behavior can be defined objectively as harsh, irrespective of how it is perceived
by the child. There are arguments on both sides. On the one hand, historical and cultural
views on parenting may change, thereby adjusting any objective definition of harsh
parenting. What is harsh for us now might have been regarded differently in our
grandparents’ generation, or is still regarded as such in another culture. On the other hand,
there is evidence that, under certain conditions, children do not recognize the “harshness”
in what many people term harsh parenting. If parents use harsh behaviors, such as physical
punishment, but love their child and have a warm atmosphere at home, the child might even
not perceive the parental behaviors as harsh or negative. Thus, it can be discussed whether
a parenting behavior can be defined objectively as harsh, apart from how it is perceived by
the child. Although there are no clear answers to this question, my view is that harsh
parenting has to be described objectively for research purposes, but comparing studies from
different cultures or historical perspectives harsh parenting definitions has to be
acknowledged.
Harsh parenting is a much more complicated matter than is usually presented or
defined. There are several things that need to be considered. For example, if we were to ask
our grandparents what harsh parenting was or what physical punishment was for them, their
answers would be very different from ours. Many of them, or even our parents, would say
that they “spare the rode, spoil the child”. Would contemporary society agree with this?
63
63
Given that corporal punishment was encouraged by the church and society for centuries,
and was common behavior at home and in school, parents who use harsh disciplines may
not even think that they are doing something wrong; it might be that such behavior lies in
their family tradition, and that they are used to that way of thinking. Even though corporal
punishment or harsh parenting practice remains a common way of disciplining children
today, its use has significantly declined. Thus, taking into account that understanding of
harsh parenting might be dependent on timing, the question remains of how relevant it is to
compare harsh parenting studies between different decades. It should be acknowledged that
there is a possibility that results would be similar, but that would raise the further issue of
whether the similarities are due to harsh parenting per se or to other things – the
measurement of harsh parenting in particular. Therefore, when comparing various studies,
historical or generational perspectives cannot be neglected.
Cultural context, which could be described as all the behaviors, ways of life, and
beliefs of a population that are passed down from generation to generation, should also be
taken into account when describing what harsh parenting is. There is a saying in the Baltic
States countries that “one spanked child is worth ten children who have not been spanked”.
It suggests that children who experience punishment behave better, and implicitly
encourages parents to adopt harsh parenting practices. In some cultures, harsh parenting
may be more normative than in others (e.g., Deater-Deckard et al., 1996; Lansford et al.,
2005). When harsh parenting is both accepted and expected in a cultural context, parents
may feel justified in using it, and children can view it as the norm. In cultures in which a
power-assertive parenting style is believed to be in the best interests of the child, harsh
parenting may be used instrumentally more than emotionally (Grusec, Rudy, & Martini,
1997), which may predict less negative outcomes (Holden & Miller, 1999; Straus &
Mouradian, 1998). It seems that, in some cultures, spanking a child is an accepted behavior,
while in others it is taboo. Indeed, it was such cultural diversity that brought the idea of
emotional context into the literature of harsh parenting. The context in which harsh
parenting occurs opened a door into the complexity of harsh parenting. Also, it raised new
questions concerning what harsh parenting is, and also the nature of its rejecting context. A
key question concerns what kinds of parents’ behaviors might be interpreted as rejecting
from the child’s perspective? Many parents would disagree that their worries or conflicts
with the child mean that they reject the child; in particular, many parents worry during their
child’s adolescence, and this may affect their behaviors in relation to the child. Cultural and
the context in which harsh parenting occurs also raises the question of whether it is relevant
to compare harsh parenting studies between different cultures or contexts. This dissertation
64
64
has shown that that the effects that can be found between ethnic groups can also be found in
homogeneous samples, which suggests that cultural issues in relation to harsh parenting
may not really have been tapped or evaluated. Thus, new ways need to be created to
evaluate cultural or contextual differences that take into account the complexity of harsh
parenting.
Taken as a whole, harsh parenting is a complex phenomenon. It is easier in research
to measure harsh parenting according to a single definition than to take into account the
other aspects by which it is influenced. But the real world is more complex than it is in
current research. The complexity of harsh parenting suggests a need to revisit the existing
literature and results, and to be careful in interpreting the results of previous studies. For
example, many of them evaluate only one aspect of harsh parenting, namely corporal
punishment. But corporal punishment is only a part of harsh parenting, and we must be
careful in generalizing from the results of these studies; the conclusions drawn from studies
evaluating corporal punishment in a certain context may be very different from those in
which a “context-free” approach is adopted. Future research should take into account the
complexity of harsh parenting, and try to combine several ideas, in particular the
combination of harsh parental behavior and the specific context. Complexity raises new
questions and challenges for the harsh parenting literature, such as how to measure harsh
parenting, what kind of instruments should be used, and what kind of research should be
conducted. The complexity of harsh parenting cannot be neglected, although it may be
tempting and convenient not to take it into account in research.
Children and youth as active agents in parent-child relationships
Historically, parenting has typically been viewed as something that parents do to
their child. However, children and youths may affect their parents’ behaviors, just as
parents may affect their children’s behaviors. This idea has been around for some time. For
example, the idea that parents react to children’s characteristics, and children, in turn, react
to parents’ behaviors was suggested early on in coercion theory (Patterson, 1982). On the
other hand, in the literature on harsh or inept parenting, and especially in the corporal
punishment literature, many more studies have looked at the effects of harsh or inept
parenting on children and youths than at youth characteristics as possible explanations for
parenting behaviors. Many studies in the harsh parenting or corporal punishment arena
have assumed that parental behavior is a cause of various child behaviors and experiences,
but – since they are correlational – these studies cannot definitively identify parents’
65
65
behaviors as a cause. Nonetheless, for at least some child behaviors, it is conceivable that
the causal direction is the opposite of what might be expected; it may be that children are
driving the associations, e.g., in that aggressive children tend to elicit more physical
punishment from their parents (Gershoff, 2002). Although bidirectionality or reciprocal
effects are widely accepted in principle and can clearly be observed in real life, especially
at early ages, they are still too seldom included in parenting research designs.
The idea of bidirectionality, as discussed in this dissertation, might have
implications not only for the literature on harsh or inept parenting, but also for other parts
of the parenting literature. First, it may be relevant to the literature on psychological
control. Even though psychological control and harsh or inept parental behavior constructs
are rooted in distinct theoretical traditions, and can be treated as different constructs,
measures of harsh or inept parenting sometimes include items that appear in the literature as
part of the psychological control construct (e.g., Barber, 1996; Barber, Olsen, & Shagle,
1994). For the most part, psychological control has been viewed as a parental characteristic
that influences children’s self-esteem and academic achievement (e.g., Barber et al., 1994;
Bean, Bush, McKenry, & Wilson, 2003). Second, it may have an impact on the literature
concerning parenting styles, which has dominated much of the empirical research on
parenting for many years. The parenting styles literature is based on the idea that parents’
warmth and control influence and shape their children’s development (e.g., Baumrind,
1991; Darling & Steinberg, 1993; Steinberg, Lamborn, Darling, Mounts, & Dornbush,
1994). On this line of thought, in the parenting styles and psychological control literature,
children’s behavior problems are primarily due to their parents’ behaviors or the style of the
parent-child relationship. This part of the literature places a lot of pressure on parents, and
seems quite simply to blame parents alone for their children’s bad behaviors. But denying
the child effect gives rise to the risk of over-interpreting the extent of parental influence.
Acknowledging that children play an important role in parent-child relationships and, in
turn, that their behaviors may strongly contribute to their own adjustment, would violate the
work of many decades and would raise a lot of questions about the conclusions of previous
studies. We cannot say that what has been presented in the psychological control or
parenting styles literature is wrong, but it is possible to say that it presents only one side of
the coin. And there is another side.
Findings that parents react negatively to youths’ problem behaviors, and that
parents’ behaviors affect youth problem behaviors, raise new questions concerning whether
parents react only to certain youth characteristics, and also whether parents’ behaviors may
affect only certain youth characteristics. Some answers are suggested in this dissertation. It
66
66
seems that parents’ behaviors affect youth problem behaviors and internalizing problems.
However, parents’ behaviors do not seem to affect youth personality traits over time, thus
suggesting that personality traits can be very stable and might evoke parents’ behaviors, but
not the other way round. Of course, these results should be interpreted with caution, but the
idea is supported by results from temperament research, which assumes that temperamental
traits are early-emerging individual characteristics that shape the course of personality
development, which then become elaborated over time into stable behavioral dispositions
that affect parental behavior (e.g., Caspi, 2000). The temperament literature states very
clearly that early temperament affects parents’ behavior. Given that temperament becomes
elaborated over time into stable behavioral dispositions, it is natural to think that children’s
characteristics and behaviors affect their parents. This not only suggest new questions for
the future, but it also provides a further reason to return to some parts of the parenting
literature, e.g. the parenting styles literature or psychological control literature, and rethink
what it has implied in the broader context of temperament research.
Taken together, there are the ideas that parents affect children and that children
affect parents, and these ideas should be combined if we want to understand and illuminate
the two-way street, where both sides are important. We still might wander into one waystreet labyrinths of the patenting literature, but it should be acknowledged that there are
some two-way streets that we have to tread if a better map of the city called parenting is to
be obtained.
Future directions
An important issue for the future is to understand what other mechanisms might
explain the link between harsh or inept parenting and later adjustment problems, and why
these parenting practices seem to be associated with problems later in life for some children
but not for others. This dissertation provides some new information about what may
influence the link between parenting and social adjustment, and also raises some new
questions. Evaluating harsh or inept parenting usually involves combining measures of
various types of parental behaviors. Some previous studies do not distinguish between
different types of harsh or inept parenting, although this dissertation suggests that in some
countries some forms of harsh or inept parental behaviors are used more often than in
others. It might be that parental behaviors have different meanings or different outcomes
according to country. Also, different parental behaviors, such as trying to make a child feel
guilty, controlling the child, or yelling at the child, might have very similar outcomes.
67
67
Parents’ behaviors may be influenced by cultural stereotypes, or they might be influenced
by different child or youth characteristics. It seems reasonable to distinguish between
different types of harsh or inept parental behaviors, and take a closer look at how they
influence later adjustment.
Although it is commonly believed that parental behavior, especially negative
parental behavior, affects children’s development, parents react to youth characteristics and
behaviors, just as youth react to parents’ behaviors. Such bidirectional relations in
childhood and adolescence have been examined in previous studies, but there remains the
question of how these relations develop over time – starting at very early age through
adolescence. Might there be strong differences between age periods in these relationships?
Perhaps, various negative parental behaviors could be reactions to a variety of children’s or
youths’ characteristics. Might it be that a transactional model that assumes bidirectional
relations between parents and children and allows consideration of the development of
problems as an ongoing and constantly changing reciprocal process between children and
their caretaking environment, starting at very early age and through adolescence, would
reveal new information about how these relations develop. Although it is known that harsh
or inept parenting and childhood problem behavior seem to affect each other over periods
of several years, there have been no empirical tests of the bidirectional processes relating
harsh or inept parenting and children’s characteristics from infancy through adolescence.
The bidirectional nature of parent-child relations also raises new challenges for the
evaluation of bidirectionality in future studies. Existing longitudinal studies have primarily
examined bidirectional relations over certain periods of time, e.g., at intervals of one year,
two years, or several years. But, according to established theoretical perspectives, parents
react to a child’s behavior right after that behavior, not a year later, which suggests that
there should be more frequent measurement points. Another question is how frequently
parents’ and children’s behaviors should be measured: several times a day, every day, or
every month? Might it be that everyday coercive circles are different from what results over
a yearly period might suggest? Accordingly, not only bidirectional processes between harsh
or inept parenting and children’s characteristics from infancy through adolescence, but also
different time periods between parents and children’s behaviors, should be evaluated.
It has been suggested that children who experience physical punishment in a context
of behaviors that communicate negative emotions show the highest levels of problem
behaviors in adolescence, and that the experience of harsh parenting seems to play a larger
role than certain children’s characteristics in the development of problem behaviors. Why
and how do parents’ behaviors that communicate negative emotions affect later
68
68
adjustment? The mechanisms behind this link were not considered in this dissertation,
although there may be several explanations. One possibility lies in a mechanism that might
be called “context choice” (Kerr, Stattin, Biesecker, & Ferrer-Wreder, 2003). On this line
of reasoning, youths generalize feelings that arise from their interactions with parents to
other situations that are structured and controlled by adults. If they feel valued and
respected in their interactions with parents, they will gravitate toward other adult-led
structured settings. If they feel unvalued, or perhaps distrusted, by parents at home, have
bad relationships with parents, or experience harsh parenting, they will gravitate toward
situations that do not have the same negative emotions associated with them. These would
be situations where adults are not present and do not influence youths’ behaviors. There is
evidence in the literature on leisure activities that, in these kinds of situations, youths are at
risk of being drawn into delinquency by their peers (e.g., Stattin, Kerr, Mahoney, Persson,
& Magnusson, 2005). Hence, negative parent-child relationships quality would lead them to
find deviant peers, which would cause more problems (Dekovi, Wissink, & Meijer, 2004).
It is still not known, however, whether the same processes apply to harsh or inept parenting
and how they might develop over time.
Another plausible explanation of why and how parents’ behaviors that communicate
negative emotions affect later adjustment draws upon social control theory (Hirschi, 1969).
The idea is that if youths are strongly attached to their parents, they will develop the social
control that prevents them from engaging in aggressive or problem behaviors; they do not
want to do anything that would hurt or embarrass their parents. Specifically, Hirschi (1969)
suggested that when youths face opportunities to commit delinquent acts, they actually
think about their parents, and the psychological presence of their parents inhibits their
behavior. One could imagine that if parents express their distrust or use physical
punishment, youths might feel that there is little to lose in terms of disappointing parents,
and that might undermine the attachment mechanism (if, indeed, it exists). Thus, parents’
harsh disciplinary practices might initiate feelings of low self control in children or
adolescents, which may predispose children towards delinquent, antisocial, or aggressive
behaviors. The attachment aspect may be helpful in understanding why parental behavior is
linked to problems. Several studies have included attachment in the examination of parents’
behaviors and children’s adjustment problems (Gottfredson & Hirschi, 1990; Sampson &
Laub, 1994), but the question remains whether this mechanism helps to understand
relations between hash parenting and later adjustment. Thus, there are several plausible
theoretical explanations and suggestions concerning the role played by a harsh parenting
context in the development of adjustment problems.
69
69
What should parents do?
Most parents want to provide a warm, safe, and supportive environment for their
children. Parents want to make sure that children in such an environment develop socially
appropriate behaviors. However, depending on the child’s temperamental predispositions
and behaviors, shaping a child’s behavior can be either quite easy or difficult and
frustrating. What should parents do when they face parenting challenges, such as having a
child who is angry or behaves badly, who is sad and depressed, or who begins to engage in
delinquency? Many parents want to know what they should do. This dissertation has
pointed to what parents do that leads to various adjustment problems, thus suggesting what
parents should not do or how parents should not react to problems.
Parents are often told that they need to have a lot of control over their teenagers.
Some caring parents even think that if they do not control their children, they will be
blamed if their teenager develops problem behavior. And many parents believe that, if they
are not strict enough, then adolescents will be irresponsible and uncontrollable outside the
home. It seems quite natural to think that parents should control their children. However,
asking parents to monitor more is probably not a good preventive strategy. Also, prevention
or intervention efforts have to be considered in the light of how parents react emotionally to
the child. Parents’ gut-level reactions (worries and distrust) to the child’s negative behavior
at home, or physical punishment with discordant relationships, might escalate the risk of
future problem behaviors. Thus, some parents’ behaviors would generate children’s
reactions opposite to what parents expected. Parents should be aware of this, try out other
behaviors from time to time, and observe whether or not there are changes in the child’s
behavior. One lesson to be learned by parents is that they should think about what they are
told to do by other parents or society, and choose behaviors that work in their own families.
Many people think that bad parents will have bad children. It seems that this way of
thinking is influenced by the idea that came from abuse literature, which suggests that
children who experience abuse from parents will abuse other people later in life. But this
way of thinking regards children as passive, implying that they have no chance of
improving their quality of life if they experience bad parental behavior. However, there is
another way of looking at the matter, which suggests that children are active and can
influence later adjustment problems on their own, and also affect their parents’ behaviors.
What should parents do with this information? Parents should be aware that children them
selves can affect behavior problems, and also that parents reactions to certain children’s
problem behaviors can increase rather than decrease childhood problem behaviors. Parents
have the power to break these coercive circles. If parents reacted differently, it might be
70
70
that children would be affected by their reactions in a different way. There is no doubt that
most loving parents, out of frustration, can start to yell at their children, or try to make them
feel guilty, instead of trying to understand the situation and solve the problem in a
constructive way. But as many situations continue, parents’ reactions may become habitual.
In such case, the parents involved should from time to time evaluate their own behaviors,
and bear in mind that children’s and adolescents’ characteristics can influence their own
reactions. Then, they can think about how they could change that situation.
There is the well-known saying, “Spare the rode, spoil the child”. Our grandparents
tended to agree that punishing a child would keep the child out of trouble. There has been
ongoing debate in the literature on physical punishment and the consequences of physical
punishment for a long time now (see, e.g., Baumrind, 1997; Deater-Deckard & Dodge,
1997; Gershoff, 2002; Holden, 2002; Larzelere, 2000). On the one hand, under the right
cultural and family conditions, physical punishment does not necessarily produce more
aggressive or problematic behavior. This could be taken as an argument against changing
the tradition of physical punishment in countries where it is still legal. And it could even be
taken as argument for parents to use more physical punishment. On the other hand, under
less than optimal family conditions, physical punishment might contribute to the
development of behavior problems that will interfere with the goals of society and the
individual’s own enjoyment of life. This should cause concern, in the first instance among
parents, but also among people working in prevention, and then among policymakers.
Parents should be aware that their attitudes or emotions toward the child can affect their
child’s adjustment. Many parents say that they punish their child because they wish the best
for the child. But, the matter, it seems, is not about what parents wish, but about how
children perceive. Of course, children can interpret parents’ attitudes in a different way than
parents do, but parents should be aware that their children are sensitive to their parents’
attitudes and emotions, and make sure that the children perceive their attitudes in the right
way.
Concluding remarks
Many children in different countries are exposed to harsh or inept parental
behaviors, such as yelling, ignoring, trying to induce guilt, and sometimes even physical
punishment. Children who experience such types of parents’ behaviors have various
problems later in life. However, the risk is far from universal; many children who
experience harsh or inept parenting are normally adjusted. The question is why some
71
71
children who experience negative parenting do not have problems later whereas others do.
What might influence the link between harsh parenting and later adjustment problems? This
dissertation suggests that the effects of negative parenting on later adjustment are
influenced by parental behaviors that could communicate negative emotions and the child’s
earlier behaviors or characteristics. Child’s earlier behaviors or characteristics also
constitute a risk factor for experiencing certain parental behaviors. The dissertation shows
that children may experience various patterns of harsh parenting, and these patterns will
have different consequences later in life. Although one cannot simply say that parental
behaviors interfere with youths’ social adjustment, children’s or youths’ characteristics or
behaviors play important roles in both parenting and later adjustment.
72
72
V References
Akse, J., Hale III, W. Engels, R. C. M. E., Raaijmakers, Q. A. W., & Meeus, W. H. J.
(2004). Personality, perceived parental rejection and problem behavior in
adolescence. Social Psychiatry and Psychiatric Epidemiology, 39, 980-988.
Andershed, H., Hodgins, S., & Tengström, A. (in press). Convergent validity of the Youth
Psychopathic Traits Inventory (YPI): Association with the Psychopathy Checklist:
Youth Version (PCL:YV). Assessment.
Andershed, H., Kerr, M., Stattin, H., & Levander, S. (2002). Psychopathic traits in nonreferred youths: A new assessment tool. In E. Blaauw & L. Sheridan (Eds.),
Psychopaths: Current international perspectives (pp. 131- 158). The Hague:
Elsevier.
Anderson, K, E., Lytton, H., & Romney, D. M. (1986). Mothers’ interactions with normal
and conduct-disordered boys: Who affects whom? Developmental Psychology, 22,
604-609.
Andrews, J. A., Foster, S. L., Capaldi, D., & Hops, H. (2000). Adolescent and family
predictors of physical aggression, communication, and satisfaction in young adult
couples: A prospective analysis. Journal of Consulting and Clinical Psychology, 68,
195-208.
Arnold, D. S., O’Leary, S. G., Wolf, L. S., & Acker, M. M. (1993). The parenting scale: A
measure of dysfunctional parenting in discipline situations. Psychological
Assessment, 5, 137-144.
Asendorpf, J. B., & Wilpers S. (1998). Personality effects on social relationships. Journal
of Personality and Social Psychology, 74, 1531-1544.
Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice-Hall.
Barber, B. K. (1992). Family, personality, and adolescent problem behaviors. Journal of
Marriage and the Family, 54, 69-79.
Barber, B. K. (1996). Parental psychological control: Revisiting a neglected construct.
Child Development, 67, 3296-3319.
Barber, B. K., Olsen, J. E., & Shagle, S. C. (1994). Associations between parental
psychological and behavioral control and youth internalized and externalized
behaviors. Child Development, 65, 1120-1136.
Bates, J. E. (1989). Application of temperament concepts. In G. A. Kohnstamm, J. E. Bates,
& M. K. Rothbart (Eds.), Temperament in childhood (pp. 321-355). Chichester,
England: Wiley.
73
73
Bates, J. E., Pettit, G. S., Dodge, K. A., & Ridge, B. (1998). Interaction of temperamental
resistance to control and restrictive parenting in the development of externalizing
behavior. Developmental Psychology, 34, 982 – 995.
Baumrind, D. (1991). The influence of parenting style on adolescent competence and
substance use. Journal of Early Adolescence, 11, 56-95.
Baumrind, D. (1997). Necessary distinctions. Psychological Inquiry, 8, 176 – 182.
Bean, R. A., Bush, K. R., McKenry, P. C., & Wilson, S. M. (2003). The impact of parental
support, behavior control, and psychological control on the academic achievement
and self-esteem of African American and European American adolescents. Journal
of Adolescent Research, 18, 523-541.
Bell, R. (1968). A reinterpretation of the direction of effects in studies of socialization.
Psychological Review, 75, 81-95.
Belsky, J. (1997). Variation insusceptibility to environmental influences: An evolutionary
argument. Psychological Inquiry, 8, 182-186.
Belsky, J., Hsieh, K., & Crnic, K. (1998). Mothering, fathering, and infant negativity as
antecedents of boys’ externalizing problems and inhibition at age 3 years:
Differential susceptibility to rearing experience? Developmental and
Psychopathology, 10, 301-319.
Blum, J., & Mehrabian, A. (1999). Personality and temperament correlates of marital
satisfaction. Journal of Personality, 67, 93-125.
Bouchard, G., Lussier, Y., & Sabourin, S. (1999). Personality and marital adjustment:
Utility of the Five-Factor Model of Personality. Journal of Marriage and the
Family, 61, 651-660.
Briere, J. (1995). Trauma symptom checklist for children: Professional manual. Odessa,
FL: Psychological Assessment Resources.
Buss, D. M. (1981). Predicting parent-child interactions from children’s activity level.
Developmental Psychology, 17, 59-65.
Capaldi, D. M., & Clark, S. (1998). Prospective family predictors of aggression toward
female partners for young at-risk males. Developmental Psychology, 34, 1175-1188.
Carlo, G., Roesch, S. C., & Melby, J. (1998). The multiplicative relations of parenting and
temperament to prosocial and antisocial behaviors in adolescence. Journal of Early
Adolescence, 18, 266-290.
Carlson, J., & Parke, R. D. (1996). Reciprocity of parent-child negative affect and
children's social competence. Child Development, 67, 2217-2226.
74
74
Caspi, A. (2000). The child is father of the man: Personality continuities from childhood to
adulthood. Journal of Personality and Social Psychology, 78, 158 – 172.
Caspi, A., Elder, G. H., & Bem, D. J. (1987). Moving against the world: Life-course
patterns of explosive children. Developmental Psychology, 23, 308-313.
Caspi, A., Henry, B., McGee, R., Moffitt, T., & Silva, P. (1995). Temperamental origins of
child and adolescent behavior problems: From age 3 to age 15. Child Development,
66, 55 – 68.
Caspi, A., Moffitt, T. E., Newman, D. L., & Silva, P. A. (1996). Behavioral observations at
age 3 years predict adult psychiatric disorders. Archives of General Psychiatry, 53,
1033-1039.
Caspi, C., & Silva P. (1995). Temperamental qualities at age three predict personality traits
in young adulthood: Longitudinal evidence from a birth cohort. Child Development,
66, 486-498.
Clark, K. E., & Ladd, G. W. (2000). Connectedness and autonomy support in parent-child
relationships: Links to children’s socioemotional orientation and peer relationships.
Developmental Psychology, 36, 485-498.
Cohen, P., & Brook, J. S. (1998). The reciprocal influence of punishment and child
behavior disorder. In J. McCord (Ed.), Coercion and punishment in long-term
perspectives (2nd ed., pp. 154-164). Cambridge: Cambridge University Press.
Cohen, C. P. (1984). Freedom from corporal punishment: One of the human rights of
children. New York Law School Human Rights Annual, Volume II, Part 1.
Colder, C. R., Lochman, J. E., & Wells, K. C. (1997). The moderating effects of children’s
fear and activity level on relations between parenting practices and childhood
symptomology. Journal of Abnormal Child Psychology, 25, 251-263.
Collins, W. A., Maccoby, E. E., Steinberg, L., Hetherington, E. M., & Bornstein, M. H.
(2000). Contemporary research on parenting: The case for nature and nurture.
American Psychologist, 55, 218-232.
Colman, R. A., & Widom, C. S. (2004). Childhood abuse and neglect and adult intimate
relationships: a prospective study. Child Abuse and Neglect, 28, 1133-1151.
Conger, R. D., & Ge, X. (1999). Conflict and cohesion in parent-adolescent relations:
Changes in emotional expression from early to mid-adolescence. In M. J. Cox & J.
Brooks-Gunn (Eds.), Conflict and cohesion in families: Causes and consequences
(pp. 185-206). Mahwah, NJ: Erlbaum.
Cooke, D. J., & Michie, C. (2001). Refining the construct of psychopathy: Towards a
hierarchical model. Psychological Assessment, 13, 171-188.
75
75
Crouter, A. C., & Booth, A. (2003). Children’s influence on family dynamics: The
neglected side of family relationships. Mahwah, NJ: Erlbaum.
Daley, S. E., & Hammen, C. (2002). Depressive symptoms and close relationships during
the transition to adulthood: Perspectives from dysphoric women, their best friends,
and their romantic partners. Journal of Consulting and Clinical Psychology, 70,
129-141.
Darling, N., & Steinberg, L. (1993). Parenting style as context: An integrative model.
Psychological Bulletin, 113, 487-496.
Deater-Deckard, K., & Dodge, K. A. (1997). Spare the rod, spoil the authors: Emerging
themes in research on parenting and child development. Psychological Inquiry, 8,
230 – 235.
Deater-Deckard, K., Dodge, K. A., Bates, J. E., & Pettit, G. S. (1996). Physical discipline
among African American and European American mothers: Links to children’s
externalizing behaviors. Developmental Psychology, 32, 1065 – 1072.
Dekovi, M., & Meeus, W. (1997). Peer relations in adolescence: Effects of parenting and
adolescents’ self-concepts. Journal of Adolescence, 20, 163-176.
Dekovi, M., Wissink, I. B., & Meijer, A. M. (2004). The role of family and peer relations
in adolescent antisocial behavior: Comparison of four ethnic groups. Journal of
Adolescence, 27, 497-514.
Diagnostic and Statistical Manual of Mental Disorders (4th ed., [DSM-IV] American
Psychiatric Association, 1994.
Dix, T., Ruble, D. N., Grusec, J. E., & Nixon, S. (1986). Social cognition in parents:
Inferential and affective reactions to children of three age levels. Child
Development, 57, 879-894.
Dolan, N. C., & Rennie, C. E. (2006a). Reliability, validity, and factor structure of the
Swedish Youth Psychopathic Traits Inventory in a UK sample of conduct
disordered boys. Journal of Forensic Psychiatry & Psychology, 17, 217-229.
Dolan, N. C., & Rennie, C. E. (2006b). Psychopathy Checklist: Youth Version and Youth
Psychopathic Traits Inventory: A comparison study. Personality and Individual
Differences, 41, 779-789.
Donnellan, M. B., Conger, R. D., & Bryan, C. M. (2004). The Big Five and enduring
marriages. Journal of Research in Personality. 38, 481-504.
76
76
Donnellan, M. B., Larsen-Rife, D., & Conger, R. D. (2005). Personality, family history,
and competence in early adult romantic relationships. Journal of Personality and
Social Psychology. 88, 562-576.
Dornbusch, S.M., Ritter, P.L., Leiderman, P.H., Roberts, D.F., & Fraleigh, M.J. (1987).
The relation of parenting style to adolescent school performance. Journal of
Adolescent Research, 5, 143-160.
Eisenberg, N., Fabes, R. A., Guthrie, I.K., & Reiser, M. (2000). Dispositional emotionality
and regulation: Their role in predicting quality of social functioning. Journal of
Personality and Social Psychology, 78, 136-157.
Eisenberg, N., Fabes, R., Shepard, S. A., Murphy, B. C., Guthrie, I. K., Jones, S., et al.
(1997). Contemporaneus and longitudinal prediction of children’s social functioning
from regulation and emotionality. Child Development, 68, 642 – 664.
Engels, R. C. M. E., Dekovic, M., & Meeus, W. (2002). Parenting practices, social skills
and peer relationships in adolescence. Social Behavior and Personality, 30, 3-18.
Faulstich, M. E., Carey, M. P., Ruggiero, L., Enyart, P., & Gresham, F. M. (1986).
Assessment of depression in childhood and adolescence: An evaluation of the
Center for Epidemiological Studies Depression Scale for Children (CES-DC).
American Journal of Psychiatry, 143, 1024-1027.
Felitti, V. J. (1991). Long-term medical consequences of incest, rape, and molestation.
Southern Medical Journal, 84, 328-331.
Finkelhor, D., Hotaling, G. T., Lewis, I. A., & Smith, C. (1989). Sexual abuse and its
relationship to later sexual satisfaction, marital status, religion, and attitudes.
Journal of Interpersonal Violence, 4, 379-399.
Fleming, J., Mullen, P. E., Sibthorpe, B. & Bammer, G. (1999). The long-term impact of
childhood sexual abuse in Australian women. Child Abuse and Neglect, 23, 145159.
Flouri, E., & Buchanan, A. (2002). What predicts good relationships with parents in
adolescence and partner in adult life: Findings from the 1958 British birth cohort.
Journal of Family Psychology, 16, 186-198.
Flouri, E., & Buchanan, A. (2002). What predicts good relationships with parents in
adolescence and partner in adult life: Findings from the 1958 British birth cohort.
Journal of Family Psychology, 16, 186-198.
Franz, C. E., McClelland, D. C., & Weineberger, J. (1991). Childhood antecedents of
conventional social accomplishments in mid-life adults: A 35-year prospective
study. Journal of Personality and Social Psychology, 60, 586-595.
77
77
Fuligni, A. J., & Eccless J. S. (1993). Perceived parent-child relationships and early
adolescents’ orientation toward peers. Developmental Psychology, 29, 622-632.
Gartstein, M. A., & Fagot, B. I. (2003). Parental depression, parenting and family
adjustment, and child effortful control: Explaining externalizing behaviors for
preschool children. Applied Developmental Psychology, 24, 143–177.
Ge, X., Conger, R. D., Cadoret, R. J., Neiderhiser, J. M., Yates, W., Troughton, E., &
Stewart, M. A. (1996). The developmental interface between nature and nurture: A
mutual influence model of child antisocial behavior and parent behaviors.
Developmental Psychology, 32, 574-589.
Gershoff, E. T. (2002). Corporal punishment, physical abuse, and the burden of proof:
Reply to Baumrind, Larzelere, and Cowan (2002), Holden (2992), and Parke
(2002). Psychological Bulletin, 4, 602 – 611.
Glasgow, K. L., Dornbush, S. M., Troyer, L., Steinberg, L., & Ritter, P. L. (1997).
Parenting styles, adolescents’ attributions, and educational outcomes in nine
heterogeneous high schools. Child Development, 68, 507-529.
Gottfredson, M., & Hirschi, T. (1990). A general theory of crime. Stanford, CA: Stanford
University Press.
Gruner, K., Muris, P., & Merckelbach, H. (1999). The relationship between anxious rearing
behvaiors and anxiety disorders symptomatology in normal children. Journal of
Behavior Therapy and Experimental Psychiatry, 30, 27-35.
Grusec, J. E., Rudy, D., & Martini, T. (1997). Parenting cognitions and child constructs: An
overview and implications for children’s internalization of values. In J. E. Grusec &
L. Kuczynski (Eds.), Parenting and children’s internalization of values: A
handbook of contemporary theory (pp. 259-282). New York: Wiley.
Guerin, D. W., Gottfried, A. W., & Thomas, C. W. (1997). Difficult temperament and
behavior problems: A longitudinal study from 1.5 to 12 years. International Journal
of Behavioral Development, 21, 71-90.
Gunnoe, M. I., & Mariner, C. L. (1997). Toward a developmental-contextual model of the
effects of parental spanking on children's aggression. Archives of Pediatric and
Adolescent Medicine, 151, 768-775.
Haapasalo, J., & Tremblay, R. E. (1994). Physically aggressive boys from ages 6 to 12:
Family background, parenting behavior, and prediction of delinquency. Journal of
Consulting and Clinical Psychology, 62, 1044-1052.
Hartup, W. W. (1978). Perspectives on child and family interaction: Past, present, and
future. In R. M. Lerner & G. B. Spanier (Eds.), Child influences on marital and
78
78
family interaction: A life-span perspective (pp.23-46). San Francisco: Academic
Press.
Hastings, P.D., & Rubin, K.H. (1999). Predicting mothers’ beliefs about preschool-aged
children's social behavior: Evidence for maternal attitudes moderating child effects,
Child Development, 70, 722–741.
Henry, B., Caspi, A., Moffitt, T. E., & Silva, P. A. (1996). Temperamental and familial
predictors of violent and nonviolent criminal convictions: Age 3 to age 18.
Developmental Psychology, 32, 614 – 623.
Hertz, L., & Gullone, E. (1999). The relationship between self-esteem and parenting style.
Journal of Cross-Cultural Psychology, 30, 742-760.
Hinshaw, S. P., Zupan, B. A., Simmel, C., Nigg, J. T., & Melnick, S. (1997). Peer status in
boys with and without attention-deficit hyperactivity disorder: Predictions from
overt and covert antisocial behavior, social isolation, and authoritative parenting
beliefs. Child Development, 68, 880-896.
Hirschi, T. (1969). Causes of delinquency. Berkeley: University of California Press.
Holden, G. W. (2002). Perspectives on the effects of corporal punishment: Comment on
Gershoff (2002). Psychological Bulletin, 128, 590-595.
Holden, G. W., & Miller, P. C. (1997). Enduring and different: A meta-analysis of the
similarity in parents’ child rearing. Psychological Bulletin, 125, 223-254.
Huh, D., Tristan, J., Wade, E., & Stice, E. (2006). Does problem behavior elicit poor
parenting? A prospective study of adolescent girls. Journal of Adolescent Research,
21, 185-204.
Hussey, D., & Singer, M. (1993). Sexual and physical abuse: The Adolescent Sexual
Concern Questionnaire (ASCQ). In M. I. Singer, L. T. Singer, & T. M. Anglin
(Eds.), Handbook for screening adolescents at psychosocial risk (pp. 131–163).
New York: Lexington Books.
Huston, T. L., & Houts, R. M. (1998). The psychological infrastructure of courtship and
marriage. The role of personality and compatibility in romantic relationships. In T.
N. Bradbury (Ed.), The developmental course of marital dysfunction (pp.114-161).
Cambridge: Cambridge University.
Iannotti, R. J., Bush, P. J., & Weinfurt, K. P. (1996). Perceptions of friends’ use of alcohol,
cigarettes, and marijuana among urban school children: A longitudinal analysis.
Addictive Behaviors, 21, 615-632.
Jaffee, S. R., Caspi, A., Moffitt, T. E., Polo-Tomas, M., Price, T. S., & Taylor, A. (2004).
The limits of child effects: Evidence for genetically mediated child effects on
79
79
corporal punishment but not on physical maltreatment. Developmental Psychology,
40, 1047-1058.
Kagan, J., Snidman, N., McManis, M., Woodward, S., & Hardway, C. (2002). One
measure, one meaning: Multiple measures, clearer meaning. Development and
Psychopathology, 14, 463-475.
Kandel, D. B., & Wu P. (1998). Disentangling mother-child effects in the development of
antisocial behavior. In J. McCord (Ed.), Coercion and punishment in long-term
perspectives (2nd, ed., pp. 106-123). Cambridge: Cambridge University Press.
Karlberg, P., Klackenberg, G., Engström, I., Klackenberg-Larsson, I., Lichtenstein, H.,
Stensson, J., & Svennberg, I. (1968). The development of children in a Swedish
urban community: A prospective, longitudinal study – Parts I-VI. Acta Paediatrica
Scandinavica Supplement, Whole No. 187.
Keown, L. J., & Woodward, L. J. (2006). Preschool boys with pervasive hyperactivity:
Early peer functioning and mother-child relationship influences. Social
Development, 15, 23-45.
Kerr, M., & Stattin, H. (2000). What parents know, how they know it, and several forms of
adolescent adjustment: Further evidence for a reinterpretation of monitoring.
Developmental Psychology, 36, 366–380.
Kerr, M., & Stattin, H. (2003). Parenting of adolescents: Action or reaction? In A.C.
Crouter & A. Booth (Eds.) Children's influence on family dynamics: The neglected
side of family relationships. Chapter 9 (pp. 121-151) Mahwah, NJ: Lawrence
Erlbaum.
Kerr, M., Stattin, H., Biesecker, G., & Ferrer-Wreder, L. (2003). Relationships with parents
and peers in adolescence. In R. M. Lerner, M. A. Easterbrooks, J. Mistry (Eds.)
Handbook of Psychology (Volume 6: Developmental Psychology) (pp. 395-422).
Hoboken, NJ: John Wiley & Sons.
Kim, K. J., Conger, R. D., Elder, G. H., & Lorenz, F. O. (2001). Parent-adolescent
reciprocity in negative affect and its relation to early adult social development.
Developmental Psychology, 37, 775-790.
Kinnunen, U., & Pulkkinen, L. (2003). Childhood socio-emotional characteristics as
antecedents of marital stability and quality. European Psychologist, 8, 223-237.
Knutson, J. F., DeGarmo, D., Koeppl, G., & Reid, J. B. (2005) Care Neglect, supervisory
neglect and harsh parenting in the development of children’s aggression: A
replication and extension. Child Maltreatment, 10, 92-107.
80
80
Kochanska, G. (1998). Mother-child relationship, child fearfulness, and emerging
attachment: A short-term longitudinal study. Developmental Psychology, 34, 480490.
Lansford, J. E., Criss, M. M., Pettit, G. S., Dodge, K. A., & Bates, J. E. (2003). Friendship
quality, peer group affiliation, and peer antisocial behavior as moderators of the link
between negative parenting and adolescent externalizing problems. Journal of
Research on Adolescence, 13, 161-184.
Lansford, J. E., Deater-Deckard, K., Dodge, K. A., Bates, J., & Pettit, G. S. (2004). Ethnic
differnces in the link between physical discipline and later adolescent externalizing
behaviors. Journal of Child Psychology and Psychiatry, 45, 801-812.
Lansford, J. E., Chang, L., Dodge, K. A., Malone, P. S., Oburu, P., Palmerus, K., et al.
(2005). Physical discipline and children’s adjustment: Cultural normativeness as a
moderator. Child Development, 76, 1234-1246.
Larzelere, R. E. (2000). Child outcomes of nonabusive and customary physical punishment
by parents: An updated literature review. Clinical Child and Family Psychology
Review, 3, 199 – 221.
Larzelere, R. E., Klein, M., Schumm, W. R., & Alibrando, S. A., Jr. (1989). The effects of
spanking and other parenting characteristics on self-esteem and perceived fairness
of parental discipline. Psychological Reports, 64, 1140-1142.
Larzelere, R. E., & Kuhn, B. R. (2005). Comparing child outcomes of physical punishment
and alternative disciplinary tactics: A meta-analysis. Clinical Child and Family
Psychology Review, 8, 1-37.
Lau, A.S., Litrownik, A.J., Newton, R.R., Black, M.M., & Everson, M.D. (2006). Factors
affecting the link between physical discipline and child externalizing problems in
Black and White families. Journal of Community Psychology, 34, 89-103.
Lengua, L.J., Wolchik, S.A., Sandler, I.N., & West, S.G. (2000). The additive and
interactive effects of parenting and temperament in predicting adjustment problems
of children of divorce. Journal of Abnormal Child Psychology, 29, 232-244.
Lerner, J. V., & Lerner, R. M. (1994). Explorations of the goodness-of-fit model in early
adolescence. In W. B. Carey & S. C. McDevitt (Eds.), Prevention and early
intervention: Individual differences as risk factors for the mental health of children:
A festschrift for Stella Chess and Alexander Thomas (pp. 161-169). New York;
Brunner/Mazel.
81
81
Leve, L. D., Kim, H. K., & Pears, K. C. (2005). Childhood temperament and family
environment as predictors of internalizing and externalizing trajectories from age 5
to 17. Journal of Abnormal Child Psychology, 33, 505 – 520.
Linder, J. R., & Collins, W. A. (2005). Parent and peer predictors of physical aggression
and conflict management in romantic relationships in early adulthood. Journal of
Family Psychology, 19, 252-262.
Loeber, R., & Dishion, T. (1983). Early predictors of male delinquency: A reviw.
Psychological Bulletin, 94, 68-99.
Lytton, H. (1990). Child and parent effects on boys’ conduct disorder: A reinterpretation.
Developmental Psychology, 26, 683-697.
Lytton, H. (1997). Physical punishment is a problem, whether conduct disorder is
endogenous or not. Psychological Inquiry, 8, 211-214.
Maccoby, E. E., & Jacklin, C. N. (1983). The ‘person’ characteristics of children and the
family as environment. In D. Magnusson & V. L. Allen (Eds.), Human
development: An interactional perspective (pp. 75-91). New York: Academic Press.
Maccoby, E,m & Martin, J. A. (1983). Socialization in the context of the family: Parentchild interaction. In P. Mussen (Ed.), Handbook of child psychology (pp.1-101).
New York: Wiley.
McClelland, G. H., & Judd, C. M. (1993). Statistical difficulties of detecting interactions
and moderator effects. Psychological Bulletin, 114, 376 – 390.
McFayden-Ketchum, S. A., Bates, J. E., Dodge, K. A., & Pettit, G. S. (1996). Patterns of
change in early childhood aggressive-disruptive behavior: Gender differences in
predictions from early coercive and affectionate mother-child interactions. Child
Development, 67, 2417-2433.
McLeod, J. D., Kruttschnitt, C., & Dornfeld, M. (1994). Does parenting explain the effects
of structural conditions on children’s antisocial behavior? A comparison of Black
and Whites. Social Forces, 73, 575-604.
McLoyd, V. C., & Smith, J. (2002). Physical discipline and behavior problems in African
American, European American, and Hispanic children: Emotional support as a
moderator. Journal of Marriage and Family, 64, 40 – 53.
Mesman, J., & Koot, H. M. (2000). Child-reported depression and anxiety in
preadolescence: I. Associations with parent- and teacher-reported problems. Journal
of the American Academy of Child and Adolescence Psychiatry, 39, 1371-1378.
82
82
Messer, S. C., & Beidel, D. C. (1994). Psychosocial correlates of childhood anxiety
disorders. Journal of the American Academy of Child Adolescent Psychiatry, 33,
975-983.
Mink, I. T., & Nihira, K. (1986). Family life-styles and child behaviors: A study of
direction of effects. Developmental Psychology, 22, 610-616.
Morris, A. S., Silk, J. S., Steinberg, L., Sessa, F. M., Avenevoli, S., & Essex, M. J. (2002).
Temperamental vulnerability and negative parenting as interacting of child
adjustment. Journal of Marriage and the Family, 64, 461-471.
Mulhern, R. K. Jr., & Passman, R. H. (1981). Parental discipline as affected by the sex of
the parent, the sex of the child, and the child’s apparent responsiveness to discipline.
Developmental Psychology, 17, 604-613.
Muris, P., Schmidt, H., Lambrichs, R., & Meesters, C. (2001). Protective and vulnerability
factors of depression in normal adolescents. Behavior Research and Therapy, 39,
555-565.
Neiderhiser, J. M., Reiss, D., Hetherington, E. M., & Plomin, R. (1999). Relations between
parenting and adolescent adjustment over time: Genetic and environmental
contributions. Developmental Psychology, 35, 680-692.
Newman, D. L., Caspi, A., Moffitt, T. E., & Silva P. A. (1997). Antecedents of adult
interpersonal functioning: Effects of individual differences in age 3 temperament.
Developmental Psychology, 33, 206-217.
Neyer, F. J., & Asendorpf, J. B. (2001). Personality-relationship transaction in young
adulthood. Journal of Personality and Social Psychology, 81, 1190-1204.
Nolan, S. A., Flynn, C., & Garber, J. (2003). Prospective relations between rejection and
depression in young adolescents. Journal of Personality and Social Psychology, 84,
745-755.
Nurmi, J.-E. (1993). Self-handicapping and a failure-trap strategy: A cognitive approach to
problem behavior and delinquency. Psychiatria Fennica, 24, 75-85.
Nurmi, J.-E., Onatsu, T., & Haavisto, T. (1995). Underachievers’ cognitive and behavioral
strategies – Self handicapping at school. Contemporary Educational Psychology,
20, 188-200.
Nurmi, J.-E., Salmela-Aro, K., & Ruotsalainen, H. (1994). Cognitive and attributional
strategies among unemployed young adults: A case of the failure-trap strategy.
European Journal of Personality, 8, 135-148.
O’Connor, T. G., Deater-Deckard, K., Fulkel, D., Rutter, M., & Plomin, R. (1998).
Genotype-environment correlations in late childhood and early adolescence:
83
83
Antisocial behavioral problems and coercive parenting. Developmental Psychology,
34, 970-981.
O’Connor, B. P., & Dvorak, T. (2001). Conditional associations between parental behavior
and adolescent problems: A search for personality-environment interactions.
Journal of Research in Personality, 35, 1-26.
Paley, B. P., Conger, R. D., & Harold, G. T. (2000). Parents' affect, adolescent cognitive
representations, and adolescent social development. Journal of Marriage and the
Family, 62, 761-776.
Parker, J. G., & Asher S. R. (1993). Friendship and friendship quality in middle childhood:
Links with peer group acceptance and feelings of loneliness and social
dissatisfaction. Developmental Psychology, 29, 611-621.
Passman, R. H., & Blackwelder, D. E. (1981). Rewarding and punishing by mothers: The
influence of progressive changes in the quality of their sons’ apparent behavior.
Developmental Psychology, 17, 614-619.
Patterson, G. R. (1982). Coercive family process. Eugene, OR: Castalia.
Patterson, G. R. (1986). Performance models for antisocial boys. American Psychologist,
41, 432-444.
Paterson, G., & Sanson, A. (1999). The association of behavioral adjustment to
temperament, parenting and family characteristics among 5-year old children.
Social Development, 8, 293-309.
Patterson, G. R., Reid, J. B. & Dishion, T. J. (1992). Antisocial boys. Eugene, OR: Castalia.
Paoulucci, E. O., & Violato, C. (2004). A meta-analysis of the published research on the
affective, cognitive, and behavioral effects of corporal punishment. The Journal of
Psychology, 2004, 197-221.
Pettit, G. S., Laird, R. D., Dodge, K. A., Bates, J. E., & Criss, M. M. (2001). Antecedents
and behavior-problem outcomes of parental monitoring and psychological control in
early adolescence. Child Development, 72, 583–598.
Poythress, N. G., Dembo, R., Wareham, J., & Greenbaum, P. E. (2006). Construct Validity
of the Youth Psychopathic Traits Inventory (YPI) and the Antisocial Process
Screening Device (APSD) With Justice-Involved Adolescents. Criminal Justice and
Behavior, 33, 26-55.
Prinstein, M. J., Borelli, J. L., Cheah, Ch. S. L., Simon, V. A., & Aikins, J. W. (2005).
Adolescent girls’ interpersonal vulnerability to depressive symptoms: A
longitudinal examination of reassurance-seeking and peer relationships. Journal of
Abnormal Psychology, 114, 676-688.
84
84
Prinzie, P., Onghena, P., Hellinckx, W., Grietens, H, Ghesquière, P., & Colpin, H. (2003).
The additive and interactive effects of parenting and children’s personality on
externalizing behavior. European Journal of Personality, 17, 95-117.
Putnam, S. P., Sanson, A., & Rothbart, M. K. (2002). Child temperament and parenting. In
M. Bornstein (Ed.) Handbook of Parenting: Vol. 1 (2nd ed., pp. 255-278). Mahwah,
NJ: Erlbaum.
Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the
general population. Applied Psychological Measurement, 1, 385-401.
Reid, J. B., Patterson, G. R., & Snyder, J. (Eds.). (2002). Antisocial behavior in children
and adolescents: A developmental analysis and model for intervention. Washington,
DC: American Psychological Association.
Reuter, M. A., & Conger, R. D. (1998). Reciprocal influences between parenting and
adolescent problem-solving behavior. Developmental Psychology, 34, 1470-1482.
Richter, J. (1994). Parental rearing and aspects of psychopathology with special reference
in depression. In C. Perris, W. A. Arrindell, & M. Eisemann (Eds.), Parenting and
psychopathology (pp. 235-251). Chi-Chester, England: Wiley.
Roberts, R. E., Lewinsohn, P. M., & Seeley, J. R. (1991). Screening for adolescent
depression: A comparison of depression scales. Journal of the American Academy
of Child and Adolescent Psychiatry, 30, 58-66.
Robins, L. N., & Rutter, M. (Eds.). (1990). Straight and devious pathways from childhood
to adulthood. Cambridge, UK: Cambridge University Press.
Rogers, K. N., Buchanan, C. M., & Winchell, M. E. (2003). Psychological control during
early adolescence: Links to adjustment in differing parent/adolescent dyads. Journal
of Early Adolescence, 23, 349-383.
Rohner, R. P., Bourque, S. L., & Elordi, C. A. (1996). Children’s perceptions of corporal
punishment, caretaker acceptance, and psychological adjustment in a poor, biracial
Southern community. Journal of Marriage and the Family, 58, 842 – 852.
Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.
Rothbart, M. K., & Bates, J. E. (1998). Temperament. In W. Damon (Series ed.) & N.
Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional, and
personality development (5th ed., pp. 105-176). New York: Wiley.
Rothbaum, F., & Weisz, J. R. (1994). Parental caregiving and child externalizing behavior
in non-clinical samples: A meta-analysis. Psychological Bulletin, 116, 55-74.
Rowe, D.C., Vazsonyi, A. T., & Flannery, D.J. (1994). No more than skin deep: Ethnic and
racial similarity in developmental process. Psychology Review 101, 396-413.
85
85
Rubin, K. H., Burgess, K. B., Dwyer, K. M., & Hastings, P. D. (2003). Predicting
preschoolers’ externalizing behaviors from toddler temperament, conflict, and
maternal negativity. Developmental Psychology, 39, 164-176.
Rutter, M. (1987). Temperament, personality, and personality disorder. British Journal of
Psychiatry, 150, 443-458.
Sameroff, A. J., & Mackenzie, M. J. (2003). Research strategies for capturing transactional
models of development: The limits of the possible. Development and
Psychopathology, 15, 613-640.
Sampson, R. J. & Laub, J. H (1994). Urban poverty and the family context of delinquency:
A new look at structure and process in a classic study. Child Development, 65, 523540.
Schoenbach, V. J., Kaplan, B. H., Grimson, R. C., & Wagner, E. H. (1982). Use of a
symptom scale to study the prevalence of a depressive syndrome in young
adolescents. American Journal of Epidemiology, 116, 791-800.
Schwartz, C. E., Snidman, N., & Kagan, J. (1996). Early childhood temperament as a
determinant of externalizing behavior in adolescence. Development and
Psychopathology, 8, 527–537.
Seifer, R., Sameroff, A. J., Barrett, L. C., & Krafchuk, E. (1994). Infant temperament by
multiple observations and mother reports. Child Development, 65, 1478-1490.
Shaw, D. S., Owens, E. B., Giovannelli, J., & Winslow, E. B. (2001). Infant and toddler
pathways leading to early externalizing disorders. Journal of the American Academy
of Child and Adolescent Psychiatry, 40, 36-1760-1767.
Simons, R. L., Chao, W., Conger, R. D., & Elder, G. H. (2001). Quality of parenting as
mediator of effect of childhood defiance on adolescence friendship choices and
delinquency: A growth curve analysis. Journal of Marriage and Family, 63, 63-79.
Simons, R.L., Johnson, C.A., & Conger, R.D. (1994). Harsh corporal punishment versus
quality of parental involvement as an explanation of adolescent maladjustment.
Journal of Marriage & the Family, 56, 591-607.
Skeem, J. L., & Cauffman, E. (2003). Views of the Downward Extension: Comparing the
Youth Version of the Psychopathy Checklist with the Youth Psychopathic Traits
Inventory. Behavioral Sciences & the Law. 21, 737-770.
Snyder, J., Cramer, A., Afrank, J., & Patterson, G. R. (2005). The contribution of
ineffective discipline and parental hostile attribution of child misbehavior to the
development of conduct problems at home and school. Developmental Psychology,
41, 30-41.
86
86
Snyder, J., & Patterson, G. R. (1995). Individual differences in social aggression: A test of
reinforcement model of socialization in the natural environment. Behavior Therapy,
26, 371-391.
Spieker, S. J., Larson, N. C., Lewis, S. M., Keller, T. E., & Gilchirist, L. (1999).
Developmental trajectories of disruptive behavior problems in preschool children of
adolescent mothers. Child Development, 70, 443-458.
Stattin, H., Janson, H., Klackenberg-Larsson, I., & Magnusson, D. (1995). Corporal
punishment in everyday life: An intergenerational perspective. In J. McCord (Ed.),
Coercion and punishment in long-term perspectives (pp. 315-347). Cambridge:
Cambridge University Press.
Stattin, H., Kerr, M., Mahoney, J., Persson, A., & Magnusson, D. (2005). Explaining why a
leisure context is bad for some girls and not for others. In J. L. Mahoney, R. W.
Larson, & J. S. Eccles (Eds.), Organized activities as contexts of development:
Extracurricular activities, after-school and community programs. (pp. 211 – 244).
Mahwah, NJ: Erlbaum.
Stattin, H., & Klackenberg, G. (1992). Discordant relations in intact families:
Developmental tendencies over 18 years. Journal of Marriage and the Family, 54,
940-956.
Stattin, H., & Klackenberg-Larsson, I. (1990). The relationship between maternal attributes
in the early life of the child and the child’s future criminal behavior. Development
and Psychopathology, 2, 99 – 111.
Steinberg, L., Lamborn, S.D., Darling, N., Mounts, N.S., & Dornbush, S.M. (1994). Overtime changes in adjustment and competence among adolescents from authoritative,
authoritarian, indulgent, and neglectful families. Child Development, 65, 754-770.
Stice, E., & Barrera, M. Jr. (1995). A longitudinal examination of the reciprocal relations
between perceived parenting and adolescents’ substance use and externalizing
behaviors. Developmental Psychology, 31, 322–334.
Stice, E., Ragan, J., & Randall, P. (2004). Prospective relations between social support and
depression: Differential direction of effects for parent and peer support? Journal of
Abnormal Psychology, 113, 155-159.
Stoff, D. M., Breiling, J., & Maser, J. D. (1997). Handbook of antisocial behavior. New
York: Wiley.
Stoolmiller, M. (2001). Synergistic interaction of child manageability problems and parentdiscipline tactics in predicting future growth in externalizing behavior for boys.
Developmental Psychology, 37, 814 – 823.
87
87
Straus, M. A. (1995). Manual for the Conflict Tactic Scales. Durham, NH: Family Research
Laboratory, University of New Hampshire.
Straus, M. A. (1996a). Corporal punishment as a stressor among youth. Journal of
Marriage and Family, 58, 155-166.
Straus, M. A. (1996b). Spanking and the making of a violent society. Pediatrics, 98, 837842.
Straus, M. A., & Field, C. J. (2003). Psychological aggression by American parents:
National data on prevalence, chronicity, and severity. Journal of Marriage and
Family, 65, 795-808.
Straus, M. A., & Mathur, A. K. (1995, April). Corporal punishment and children's
academic achievement. Paper presented at the annual meeting of the Pacific
Sociological Society, San Francisco.
Straus M. A., & Mouradian, V. E. (1998). Impulsive corporal punishment by mothers and
antisocial behavior and impulsiveness of children. Behavioral Sciences and Law,
16, 353-374.
Straus, M. A., Sugarman, D. B., & Giles-Sims, J. (1997). Spanking by parents and
subsequent antisocial behavior of children. Archives of Pediatric and Adolescent
Medicine, 151, 761-767.
Van Brakel, A. M. L., Muris, P. Bögel, S. M., & Thomassen, C. (2006). A multifactorial
model for the etiology od anciety in non-clinical adolescents: Main and intercative
effects of behavioral inhibition, attachment, and parental rearing. Journal of Child
and Family Studies, 15, 569-579.
Van Leeuwen, K. G., Mervielde, I., Braet, C., & Bosman, G. (2004). Child personality and
parental behavior as moderators of problem behavior: Variable- and personcentered approaches. Developmental Psychology, 40, 1028-1046.
Vissing, Y. M., Straus M. A., Gelles, R. J., & Harrop, J. W. (1991). Verbal aggression by
parents and psychosocial problems of children. Child Abuse and Neglect, 15, 223–
238.
Wakschlag, L.S., & Hans, S. L. (1999). Relation of maternal responsiveness during infancy
to the development of behavior problems in high-risk youths. Developmental
Psychology, 35, 569-579.
Watson, D., Hubbard, B., & Wiese, D. (2000). General traits of personality and affectivity
as predictors of satisfaction in intimate relationships: Evidence from self- and
partner-ratings. Journal of Personality, 68, 413-449.
88
88
Weiss, B., Dodge, K. A., Bates, J. E., & Pettit, G. S. (1992). Some consequences of early
harsh discipline: Child aggression and a maladaptive social information processing
style. Child Development, 63, 1321-1335.
Weissman, M. M., Sholomskas, D., Pottenger, M., Prusoff, B. A., & Locke, B. Z. (1977).
Assessing depressive symptoms in five psychiatric populations: A validation study.
American Journal of Epidemiology, 106, 203-214.
White, J. L., Moffitt, T. E., Caspi, A., Bartusch, D. J., Needles, D. J., & StouthamerLoeber, M. (1994). Measuring impulsivity and examining its relationship to
delinquency. Journal of Abnormal Psychology, 103, 192-205.
Wootton, J. M., Frick, P. J., Shelton, K. K., & Silverthorn, P. (1997). Ineffective parenting
and childhood conduct problems: The moderating role of callous-unemotional traits.
Journal of Consulting and Clinical Psychology, 65, 301-308.
89
89
STUDY I
115
Klunserna Study
115
05-10-25, 10.10
STUDY I
Child Abuse & Neglect 28 (2004) 113–127
Cross-cultural comparisons of child-reported emotional and
physical abuse: rates, risk factors and psychosocial symptoms夽
Sandra Sebre a,∗ , Ieva Sprugevica a , Antoni Novotni b , Dimitar Bonevski b ,
Vilmante Pakalniskiene c , Daniela Popescu d , Tatiana Turchina d ,
William Friedrich e , Owen Lewis f
a
Department of Psychology, University of Latvia, Jurmalas gatve 74/76, Riga LV-1083, Latvia
b
Medical Faculty, St. Cyril and Methodius University, Skopje, Macedonia
c
Department of Clinical Psychology, Vilnius University, Vilnius, Lithuania
d
National Center for Child Abuse Prevention, Chisinau, Moldova
e
Mayo Clinic, Rochester, MN, USA
f
Columbia University College of Physicians and Surgeons, New York, NY, USA
Received 6 October 2002; received in revised form 29 May 2003; accepted 16 June 2003
Abstract
Objectives: This study was designed to assess the incidence of child emotional and physical abuse, associated risk
factors and psychosocial symptoms in a cross-cultural comparison between post-communist bloc countries.
Method: One-thousand one-hundred forty-five children ages 10–14 from Latvia (N = 297), Lithuania (N = 300),
Macedonia (N = 302), and Moldova (N = 246) participated in the study. They completed questionnaires assessing
their experience of emotional or physical abuse, and provided information about family risk-factors and psychosocial
symptoms, including PTSD-related symptoms.
Results: Incidence rates of maltreatment differed by country, as did levels of reported psychosocial symptoms.
Incidence of emotional and physical abuse differed by region, with higher levels of abuse reported in the rural
regions. In all four countries, a similar association between emotional/physical abuse and psychosocial symptoms
was found, with the uniformly largest correlation between emotional abuse and anger. When examining the combined
scores of emotional and physcial abuse, even higher correlation’s were found, particularly in relation to anger
and depression. In all four countries, parental overuse of alcohol was associated with emotional and/or physical
abuse.
夽
The authors wish to acknowledge support from the Open Society Institute, Soros Foundation, NY, and Children’s Mental
Health Alliance Foundation, NY.
∗
Corresponding author.
0145-2134/$ – see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.chiabu.2003.06.004
114
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
Conclusions: Findings show differences by country in child-reported levels of emotional and physical abuse, but
similar patterns of correlation with psychosocial symptoms and the risk factors of parental alcohol overuse and
living in a rural area.
© 2003 Elsevier Ltd. All rights reserved.
Keywords: Cross-cultural; Emotional abuse; Physical abuse; Psychosocial symptoms
Introduction
Since the break-up of the Soviet Union (1989–1991), social awareness and acknowledgment of child
maltreatment have increased enormously in the post-communist bloc countries. During the period of
Soviet influence, there was little or no official acknowledgment of child abuse. Any admission of child
abuse would have been incongruous with the official vision of the perfect socialist society (Sicher et al.,
2000). In addition, central to the Soviet ideology was a primary allegiance to the state. What happened in
the family was of less importance. As a result, systematic research and official documentation regarding
child abuse was absent.
Child abuse did exist during the period of Soviet domination. Evidence for this comes from listening
to adults speaking about their childhood abuse experiences during the Soviet period. In addition, several
contemporary studies document adults reporting instances of abuse from their childhood, when the Soviet
government was still in power (Sebre, 2000; Vanderlinden, Varga, Peuskens, & Pieters, 1995). Unfortunately, during the Soviet period there was little or no opportunity to speak about these issues. Judicial
recourse was absent, particularly for intrafamilial abuse, and few professional resources existed (Lewis
et al., 2001).
The present study examined levels of child-reported abuse in four post-Communist bloc countries. This
cross-cultural comparison takes into consideration the broader sociohistorical background, since child
abuse results from a complex interaction of social and cultural conditions wherein parental behaviors and
characteristics become manifest (Agathonos-Georgopoulous, 1992). In the present study, the attitudes
and ideologies regarding parent-child relationships during the Soviet period are of greatest concern, since
the children and adolescents of today are being parented by those who were raised according to the Soviet
childrearing beliefs of yesterday.
The sociocultural context of the Soviet period was imbued by contradictory views on parent-child
relationships. The role attributed to motherhood within the Soviet society changed according to shifting
ideological stance. These contradictions reflected the Marxist-Leninist emphasis upon the equality of the
sexes, versus other proclamations that placed greater emphasis upon the role of woman as mother (Du
Plessix Gray, 1989; Kerig, Aloyshina, & Volovich, 1993) On the one hand, Stalin proclaimed the cult
of motherhood and abolished abortions, and on the other hand directed the deportation of millions of
women and children to Siberia. Throughout the Soviet period debates centered on whether it was better
for a woman to adhere to the Marxist-Leninist ideal of sexual equality—the woman as working class
heroine driving a tractor—or to “return” to the more traditional roles of housewife and mother (Atwood,
1990). Reference has been made to the “Stalinist mother” as authoritarian, dominating and emotionally
distant towards her children, yet overly involved in her own career (Šebek, 1994).
Attitudes about parenting methods were also replete with contradictions. Some officially sanctioned
parenting journals advised parents to use disciplinary methods such as the withdrawal of affection, and
115
encouraged inducing guilt feelings (Bronfenbrenner, 1970). On the other hand, many Soviet mothers
were noted for overprotecting their children and restricting age-appropriate autonomy (Pearson, 1990).
Child rearing practices differed in relation to gender, but in a direction somewhat opposite to that of the
West (where boys have been traditionally encouraged to be the more active and autonomous). In many
parts of the Soviet Union mothers encouraged their daughters to be active and responsible, while sons
were often pampered and thereby indirectly encouraged to be passive (Kerig et al., 1993). The upbringing
of boys was also influenced by the frequent absence of a positive male role model, since often the father
was either inactive or rarely at home (Du Plessix Gray, 1989).
It is certainly important to appreciate that the Communist-bloc countries included hundreds of different ethnic groups, each of which integrated into their local child rearing attitudes and behaviors their
pre-Communist traditions. For example, the situation in the Soviet Republic of Latvia (1945–1990) was
such that women were often likely to adhere to the ideology of the strong working woman, rather than
follow those ideological dictates calling for a more “traditional” housewife and gentle mother. The tradition of strong Latvian working woman is consistent with previous historical periods when Latvian men
were inscripted in various conquering foreign armies, and the Latvian women were obliged to take care
of the farmstead. In addition, the common opinion was not only that work is primary, but that lavishing
too much attention and love upon one’s child could be harmful, that “too frequent affection will spoil the
child.” Since the breakup of the Soviet Union, Latvian women have infrequently been known to comment
upon the fact that as mothers they were “too busy building Communism,” and as grandmothers they regret
not having spent more time with their children (Sebre, 2000).
In contrast, the Macedonian family prior to World War II was typically rural and patriarchal, the father
assuming the more powerful role in the family. At the same time, he was emotionally disengaged from
the children, and their upbringing was entirely the responsibility of the mother. Multigenerational input
was typical, and the role of the grandparent was respected. During the past several decades, families are
becoming more urbanized and egalitarian, and fathers are becoming emotionally closer to their children.
However, the typical “Macedonian man” is still struggling for the most powerful position in the family.
Women are often working outside of the home, but expected to take on the role of “super woman,” which
entails both career demands and major responsibility for the care of the children and the home.
The present study gathered data nearly 10 years after the dissolution of the Soviet state and the Communist bloc. During this period each country has experienced its own difficulties of transition and development as a democratic nation state. Nevertheless, many positive changes have taken place and are
common to each of the countries in this study. There is a new interest and openness regarding parent-child
relationships—exemplified by an increase in new journals for parents, parenting literature translated from
the West, and articles written locally. Concern with the socialization processes necessary to produce a
good Soviet citizen has passed, and has been replaced by a greater awareness of the importance of a
positive parent-child relationship. These changes are occurring gradually, and it is the young parents now
beginning to raise a family who are the ones most likely to be aware that expression of love and affection
is a positive parental attitude.
Despite these improvements, child abuse continues to be a serious concern. Although at present child
maltreatment is publicly acknowledged, reported on in the mass media and sometimes considered in
court, many problems remain. Only recently have there been attempts to gain valid information on
the incidence of child maltreatment and associated symptoms in Eastern Europe. Speculation is that
child abuse is a serious problem, and Latvian mental health professionals typically believe that financial
difficulty and unemployment leads to anxiety and alcohol abuse in the parent, which then become the
STUDY I
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
116
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
biggest contributors to child abuse (Sebre, Sprugevica, Zagare, & Sluka, 1998). More precise information
will allow a better understanding of the complexities of the problem with an eye towards both intervention
and prevention.
The present study was a collaborative effort with team members from several Baltic and Eastern
European countries, together with specialists from the United States, who were sponsored by the Soros
Foundation Open Society Institute. Team members from Latvia, Lithuania, Macedonia, and Moldova
agreed to carry out this cross-cultural research project in their respective countries, with the goal of
examining the following research questions:
1. What is the child-reported incidence of emotional and physical abuse in each of the countries?
2. What is the relationship between emotional and physical abuse and demographic factors including
age, gender, and where the child is living—large city, medium city, or small city/rural area?
3. What is the level of psychosocial symptoms reported by children in each of the countries?
4. What is the relationship between child maltreatment types and psychosocial symptoms?
5. What is the relationship between parents’ employment status, overuse of alcohol and child emotional
and physical abuse?
Methods and procedures
Informed consent
Since research ethics committees do not exist in the four countries involved in this study, the research
team consulted with several epidemiological researchers from the United States as to the most appropriate
strategy to guarantee that the rights of the subjects would be respected. The principle of “passive parental
consent” was suggested. Permission to conduct the study was first received from the local school boards.
Parents then received information that a study was taking place concerning adolescents’ thoughts, feelings,
and relationships, that the study was voluntary and confidential. If parents objected to their child’s
participation, they were asked to inform school authorities. In fact, several parents from each locale did
so, and their child was excluded from the study. The remaining children were then told that they would
be asked to fill out questionnaires regarding their thoughts, feelings, and relationships. The children were
also informed that participation in the study was voluntary, completely confidential, and anonymous. We
did not request any identifying information on the forms. After the questionnaires had been completed, the
children were given an information sheet with telephone numbers of local psychological services or crisis
centers where they could call if they wished to discuss any issues or questions which they might have.
Data collection took place in the 4th and 7th grade classrooms during the school day, and was administered according to the same procedure by each country’s respective research team.
Questionnaires were completed by 1145 children ages 10–14 from several Baltic and Eastern European
countries. Participants included 297 children from Latvia, 300 children from Lithuania, 302 children from
Macedonia, and 246 children from Moldova. The data were collected during the spring of 1998 in Latvia,
the spring of 1999 in Lithuania, and the spring of 2000 in Macedonia and Moldova. Data collection
within each country took place in two large-city schools, two medium-city schools, and two small-city
or rural schools. The cities and schools were randomly chosen from different regions of each country,
with the stipulation that the children attending this school would be fluent in the major national language.
117
Table 1
Grade level, gender and mean age of respondents by country
Latvia (n = 297)
Lithuania (n = 302)
Macedonia (n = 300)
Moldova (n = 246)
n
Mean age
n
Mean age
n
Mean age
n
Mean age
4th grade
Male
Female
71
69
10.39
10.33
77
72
10.23
10.18
72
79
10.70
10.77
53
69
10.54
10.48
7th grade
Male
Female
85
72
13.19
13.15
67
84
13.21
13.19
73
78
13.19
13.14
60
64
13.76
13.69
Grade level/gender
Such a stipulation was made since the questionnaires for this initial study had been translated into the
national languages of each country—Latvian, Lithuanian, Macedonian, Moldovian. Consequently, the
ethnic composition of each sample was primarily (greater than 90%) Latvian, Lithuanian, Macedonian,
and Moldovan, respectively.
Participants
The study included 1145 primary school students 10–14 years of age from 4th to 7th grades from Latvia,
Lithuania, Macedonia, and Moldova (Table 1). Children generally begin first grade at age 7 within the
countries participating in this study, subsequently, the children from 4th grade were mostly 10–11 years
old, and those from the 7th grade were mostly 13–14 years old. Any child younger than 10 years or older
than 14 years was excluded from the study. The age range of the study participants includes both children
and adolescents, however, for purposes of brevity they will all be further referred to as children.
Measures
A questionnaire that included items concerning demographic information, reports of emotional abuse
and physical abuse, and psychosocial symptoms was compiled in English. After a final version was agreed
upon, in each country the questionnaire items were translated from English to Latvian, Lithuanian, Macedonian, and Moldovian, respectively, by two independent translators. Any discrepancies were discussed.
The consented upon version was then back-translated to English by two different independent translators,
and again any discrepancies were discussed and resolved.
Demographic information
The respondent was asked to indicate his or her age, sex, ethnicity and number of family household
members. The research teams grouped the questionnaire forms as coming from big-city, medium-city, or
rural schools. (Regrettably, the information on school region for the Moldovan sample became separated
from the data in transit to Latvia, where the questionnaire responses were entered into the computer data
set and analyzed.)
STUDY I
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
118
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
Abuse measures
Conflict Tactics Scale, Child Form R (CTS) (Straus, 1995).
This widely utilized scale assessed the extent of emotional and physical abuse the child reports as
having experienced within the past year (“Please indicate how often your parents did each of these things
in the past year”). The original scale was modified for this study, with several items added to the emotional
abuse scale (e.g., “tried to make you feel guilty”). The rating of the items was changed to a 5-point Likert
scale (never = 1 to always = 5). The final version included 23 items. Two initial items asked the child
to report on positive parental behaviors. Emotional abuse (11 items) was assessed with such items as
“insulted you,” “tried to make you feel guilty,” “made you feel like you were a bad person,” and “sulked
or refused to talk about an issue.” Physical abuse (10 items) was assessed by inquiring about potential
parental behaviors ranging from less severe to more severe forms of physical abuse, such as “threw
something at you,” “slapped or spanked you,” to “beat you up,” and “burned or scalded you.” Coefficient
alpha reliability estimates based on the present samples in each of the four countries ranged from .79 to
.87 for both the emotional and physical abuse scales.
Three separate scores were derived from the emotional and physical abuse subscales. First, was calculated a mean “emotional abuse score,” which was the average rating of all the emotional abuse items.
Second, was calculated a mean “physical abuse score,” the average rating of all of the physical abuse items.
Third, an overall rating of “abused” versus “not abused” was also derived. A coding scheme was developed by the research team, with abused/not abused ratings based on the frequency of the reported abuse,
while taking into consideration the severity of the abuse item. If the child marked “often” or “always”
regarding any of the emotional abuse items, or at least “sometimes” for the more severe emotional abuse
items, then the child was coded as “emotionally abused.” If the child marked “sometimes,” “often,” or
“always” regarding any of the physical abuse items, or at least “rarely” for the more severe physical abuse
items, then the child was coded as “physically abused.” Initially, each of the above mentioned scores was
analyzed for 4th and 7th graders separately, and then the ratings from all children within each country
were combined and examined for overall, main effects.
Psychosocial symptoms
Adolescent Sexual Concerns Questionnaire (ASCQ) (Hussey & Singer, 1993).
This 31-item scale was completed by the 7th grade participants only, since it is meant specifically for
adolescents. The items were rated on a 4-point scale (“never” = 0 to “almost all of the time” = 3). The
ASCQ rated somatic problems (5 original items plus two items added for the purposes of this study because
clinical experience in several of these countries had indicated markedly elevated rates of post-traumatic
somatic complaints), sexual concerns (14 items), and relationship issues (10 items). For purposes of the
present analysis only mean scores from the Somatic Problems scale were used for general correlations.
Trauma Symptom Checklist Children (TSCC) (Briere, 1995).
Psychosocial symptoms were assessed with the TSCC, a 54-item 4-point scale (0 = never to 3 = almost
all of the time). The items are grouped into six subscales: Anxiety, Depression, Anger, Posttraumatic
stress, Dissociation, and Sexual Concerns. Mean scores were calculated for each subscale. Cronbach
alpha reliability ratings were acceptable to good within the data sets of each of the four countries:
depression, .75–.82; anxiety, .76–.81; anger, .74–.80; dissociation, .71–.78; posttraumatic stress, .74–.84;
and sexual concerns, .69–.76.
119
Risk measure
The child also completed two questions that were related to the potential risk factors of unemployment
and alcohol abuse. Participants were asked about their parents’ employment status with the question,
“Does your father (mother) work outside the home?” and about excessive alcohol use by a parent with the
question “Is there a person in your family who uses alcohol overly much?” The format for each question
was “yes/no.”
Results
Incidence of emotional and physical abuse
The incidence of emotional and physical abuse was derived from the percentage of children within each
country who were rated as emotionally or physically abused (Table 2). Based on their responses to the CTS,
the occurrences of emotional abuse were: 29% of the Latvian children, 33% of the Lithuanian children,
13% of the Macedonian children, and 32% of the Moldovan children. Physical abuse was reported by
17% of the Latvian children, 26% of the Lithuanina children, 12% of the Macedonian children, and 30%
of the Moldovan children were categorized as physically abused.
The percentages of children who reported at least one type of abuse were: 33% of the children from
Latvia, 42% of the children from Lithuania, 18% of the children from Macedonia, and 43% of the children
from Moldova.
Table 2
Percentage of children reporting emotional and/or physical abuse by country and grade level
n
Emotional abuse (%)
Physical abuse (%)
Emotional or physical abuse (%)
Latvia
4th grade
7th grade
140
157
22.1
34.4
18.6
15.9
28.6
36.3
Overall
297
28.8
17.4
32.8
Lithuania
4th grade
7th grade
149
150
36.2
30.7
31.5
20.79
48.3
36.7
Overall
299
33.3
26.0
42.3
Macedonia
4th grade
7th grade
145
143
12.9
12.4
14.4
10.0
18.9
16.5
Overall
285
12.5
12.2
17.8
Moldova
4th grade
7th grade
119
124
21.1
43.5
25.4
35.5
33.3
54.0
Overall
243
32.1
29.7
43.1
STUDY I
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
120
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
Table 3
Percentage of children reporting “Often” or “Always” on frequently rated emotional abuse items by country
Emotional abuse scale items
Latvia
(n = 297), %
Lithuania
(n = 300), %
Macedonia
(n = 285), %
Moldova
(n = 243), %
Yelled at you
Insulted you
Criticized you
Tried to make you feel guilty
5.7
5.8
3.5
6.1
9.0
3.3
5.4
4.0
4.6
2.5
2.2
2.5
3.7
2.2
3.9
6.3
Chi-square analyses indicated several differences by grade level: Latvian 7th grade children reported
higher incidence of emotional abuse than 4th graders (p < .05); Moldovan 7th graders also reported
more emotional abuse than 4th graders (p < .001); and Lithuanian 4th graders reported more physical
abuse than 7th graders (p < .05).
The most commonly endorsed emotional abuse items from the CTS are shown in Table 3. Lithuanian
children (9%) and Macedonia children (5%) reported that the most commonly used method by their
parents for resolving conflicts is “yelling” at them. Latvian and Moldovan children (6%) reported that
their parents “often” or “always” used the tactic of making them “feel guilty.”
Differences in mean scores of abuse by country, region, and gender
An ANOVA comparing the mean scores of the emotional abuse ratings by country indicated significant
differences, F(3, 1121) = 19.89, p < .001; as did comparison of physical abuse mean scores by country,
F(3, 1108) = 10.87, p < .001. The mean scores of the emotional abuse and physical abuse ratings were
markedly lower for the Macedonian children.
A two-way (Country×Region) ANOVA indicated significant differences for emotional abuse by region,
F(2, 887) = 4.26, p < .01; and significant differences for physical abuse by region, F(2, 874) = 10.12,
p < .001. As shown in Table 4, the abuse scores were typically highest in the rural areas, somewhat less
in the middle-size cities, and least in the big-cities. Data on region were not available from Moldova. A
two-way (Country×Gender) ANOVA indicated no differences by gender for emotional or physical abuse.
Psychosocial symptom scores
Two way ANOVA (Country × Scale) of the mean scores from the TSCC subscales found significant
differences: Anxiety F(3, 1135) = 26.81, p < .001; Depression F(3, 1135) = 41.05, p < .001; Anger
F(3, 1134) = 56.89, p < .001; PTS F(3, 1135) = 37.61, p < .001; Dissociation F(3, 1134) = 49.62,
p < .001; Sexual Concerns F(3, 1135) = 12.08, p < .001. The ASCQ Somatic Problems scale also
differed across countries: F(3, 687) = 20.18, p < .001. The mean scores of the TSCC scales were
highest for Latvia and Lithuania, and the Macedonia mean scores were lowest. A similar pattern was
noted for ASCQ Somatic Problems.
Relation between psychosocial symptoms and abuse type
Pearson correlations were calculated by country to examine the relationship between reported emotional
and physical abuse and psychosocial symptom scores. The TSCC scales are all significantly correlated
1.44 (.47)
1.09 (.18)
1.48 (.57)
1.12 (.28)
Note. Standard deviations in parentheses.
Emotional abuse
Pysical abuse
1.52 (.60)
1.19 (.43)
1.50 (.44)
1.11 (.19)
Big-city
(n = 100)
Rural
(n = 106)
Big-city
(n = 81)
Medium-city
(n = 112)
Lithuania
Latvia
1.51 (.48)
1.15 (.21)
Medium-city
(n = 100)
1.61 (.55)
1.19 (.31)
Rural
(n = 100)
1.17 (.26)
1.03 (.07)
Big-city
(n = 92)
Macedonia
1.25 (.47)
1.05 (.12)
Medium-city
(n = 84)
1.32 (.47)
1.14 (.34)
Rural
(n = 99)
121
STUDY I
Table 4
Mean scores of child-reported emotional and physical abuse by country and region
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
122
Table 5
Correlations of child-reported trauma symptoms with combined mean emotional and physical abuse scores
Combined emotional
and physical abuse
Depression
Anger
Dissociation
PTSD
Anxiety
Sexual
concerns
Somatic
problems
Latvia
Lithuania
Macedonia
Moldova
.47∗∗
.43∗∗
.56∗∗
.52∗∗
.54∗∗
.46∗∗
.69∗∗
.53∗∗
.39∗∗
.47∗∗
.56∗∗
.47∗∗
.39∗∗
.38∗∗
.49∗∗
.48∗∗
.31∗∗
.41∗∗
.47∗∗
.47∗∗
.30∗∗
.33∗∗
.47∗∗
.44∗∗
.33∗∗
.27∗∗
.19∗
.28∗∗
∗
p < .05.
p < .01.
∗∗
with the emotional and physical abuse scales. The largest correlations were typically between abuse and
TSCC Anger scores, with the highest correlation is between TSCC Anger and reports of physical abuse
in Macedonia, r(275) = .54, p < .001.
Because of literature reporting that cumulative abuse is more problematic than single types of abuse
(Garbarino, 1999), we calculated the correlation between total abuse and symptom level from the TSCC
and the ASCQ for the entire sample (Table 5). The mean score of the summed emotional abuse and
physical abuse mean scores served as the combined abuse score. Again, significant correlations were
found between the psychosomatic symptoms and the combined abuse mean score, with the highest
correlation found between combined abuse scores and TSCC Anger, r(297) = .69, p < .001 for the
Macedonian children.
Risk factors and reported abuse
The reports of father not working outside of the home were: Latvia, 9%; Lithuania, 17%; Macedonia,
7%; Moldova, 24%. Many of the fathers not working outside the home were unemployed, but others may
have been self-employed as craftsmen or involved in subsistence farming. Mothers not working outside
of the home were reported as follows: Latvia, 19%; Lithuania, 26%; Macedonia, 30%; Moldova, 23%.
The children reported the following rates of parental overuse of alcohol: Latvia, 12%; Lithuania, 13%;
Macedonia, 4%; Moldova, 4%. Analysis of alcohol overuse by region indicates that there is higher rate of
alcohol overuse in the rural areas in comparison to the big cities: in Latvia 22% versus 4%; in Lithuania
17% versus 8%; in Macedonia 6% versus 3% (Moldova data are not registered by region).
We calculated a series of correlations between the potential risk factors of parental employment status,
alcohol abuse, and family size with the abuse ratings. When fathers were not working outside the home,
Table 6
Correlations of child-reported parental overuse of alcohol and emotional/physical abuse
Overuse of alcohol
Emotional abuse
Physical abuse
Latvia
Lithuania
Macedonia
Moldova
.28∗∗
.23∗∗
.35∗∗
.14∗
.35∗∗
.29∗∗
.32∗∗
.06
∗
p < .05.
p < .01.
∗∗
123
abuse ratings were significantly higher for the Latvian sample: emotional abuse, r(297) = .150, p < .05,
and physical abuse, r(297) = .179, p < .01. Abuse ratings were not correlated with mother’s employment
status nor with family size for any country.
Parental overuse of alcohol was significantly correlated with emotional and physical abuse for Latvia,
Lithuania, and Macedonia (Table 6). Abuse of alcohol and emotional abuse were found to be correlated
for Moldova, but at a lower rate, r(237) = .14, p < .05. Father not working outside the home and overuse
of alcohol were found to be correlated for the Lithuanian sample, r(300) = .132, p < .05.
Discussion
This is the first study of its kind which has examined both the incidence and correlates of emotional and
physical abuse of children, that is, risk factors and psychosocial symptoms, in a large sample of 4th and
7th graders from four post-Communist bloc countries. Studies in which children are questioned directly
are generally less frequent than those in which adults are questioned retrospectively about abuse trauma:
in addition, this study involves a cross-cultural comparison of child-reported data from Latvia, Lithuania,
Macedonia, and Moldova. Results show that children reporting emotional and/or physical abuse during
the past year, ranged from 18% of the children from Macedonia, 33% of the children from Latvia,
42% of the children from Lithuania, and 43% of the children from Moldova. Significant differences in
the incidence of emotional and physical abuse were noted across countries. Children from Macedonia
reported the lowest rates of both types of maltreatment. Children from Lithuania reported the highest rate
of emotional abuse, and children from Moldova reported the highest rate of physical abuse.
These between-country differences can be examined from several perspectives. First, it may be that
the overall parental behavior in the various countries is genuinely different. Second, maltreatment may
be conceptualized differently in these countries, and what “falls outside the range of acceptability”
is variously defined (Korbin, 1991). For example, children may consider as “normal” certain parental
behaviors which mental health professionals view as abusive. These children do not pay special attention
to such “normal” parent behaviors and therefore do not report them. Third, sociocultural traditions and
prohibitions may influence the child’s willingness to report freely about experience within the family.
The Macedonian researchers do not believe that Macedonian parents are less abusive, but that Macedonian children are raised with a very strong sociocultural taboo against speaking negatively of one’s family.
These sociocultural constraints imply that in order to be a good child one must speak positively of one’s
parents. These sociocultural prohibitions against speaking badly of one’s family become internalized
and create a constraint on the self-reporting of the Macedonian children. In addition, the sociocultural
traditions in Macedonia concerning what “falls outside the range of acceptability” may encourage Macedonian children to perceive certain abusive parental behaviors as positive. For example, the Macedonian
sociocultural belief system includes presuppositions that strong discipline is a good thing, “spankings
come from the heaven,” and discipline is necessary to help children grow into good persons (Bonevski
& Novotni, 2000).
Practical implications and planning future programs for child abuse treatment and prevention might
begin with the specific abusive parental behaviors that children have reported and rated in terms of
frequency of occurrence. By taking note of the specific abusive parental behaviors in each country, one
can make parents more aware and encourage change. For example, the Latvian children rated as most
common the emotional abuse item “tried to make me feel guilty” and the Lithuanian children most
STUDY I
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
124
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
frequently marked the item “yelled at me.” These reported parental behaviors seem to fit the cultural
stereotype of Lithuanians as more outwardly expressive than Latvians (Lieven, 1993)—hence the more
frequent yelling of Lithuanian parents fits this model of active, outward expression. The “making one
feel guilty” fits a more passive-aggressive, subtle method of disciplining. The opinion of the Latvian
team is that guilt inducement is a serious problem within parent-child relationships in Latvia. Although
this discipline technique is widely used, many Latvian parents are not yet aware that such a method is
emotionally abusive and has specific consequences.
The reports of physical abuse by the Lithuanian (26%) and Moldovan (30%) children are higher than
those of the children from Latvia (17%) and Macedonia (12%). A comparison of the two neighboring
Baltic countries, Latvia and Lithuania, again seems to fit the cultural stereotype of Lithuanians as more
outwardly expressive than Latvians, and therefore perhaps more likely to “slap or spank” their child.
Comparison of reports between 4th and 7th graders showed higher incidence of emotional abuse by 7th
graders in two countries, Latvia and Moldova. The higher reporting of emotional abuse by 7th graders is
likely to be the result of a bi-directional effect. The 7th graders (13–14 year olds) are developmentally more
eager to assert their claims to autonomy, and such assertions frequently can result in conflict situations
between parent and adolescent, possible provocations by the adolescent, and emotionally abusive response
from the parents, especially if the parent is so inclined. In addition, early adolescence is often a time of
increased emotional sensitivity, and even relatively benign comments by parents can at times be interpreted
by the adolescent as offending. There is need for further research to examine more closely these differences
of abusive parental behaviors in relation to the child’s developmental level, especially since the elevation
in 7th graders’ reporting of emotional abuse was evident only in Latvia and Moldova, not in Lithuania
and Macedonia.
Findings from this study also indicate that the levels of emotional and physical abuse in all four countries
are higher in the rural areas than in the big or medium cities. The most likely explanation for this finding
is that there is a greater amount of financial and psychological stress in the rural areas, due specifically
to the breakup of the Soviet-style kolhozes or collective farms. During the Soviet period the kolhozes
provided all of the local residents with employment and a small but regular income. After the dissolution
of the kolhozes, many residents in the rural areas have been faced with dire financial difficulties and
psychological problems due to the loss of one’s professional standing as a skilled collective farm worker.
Although only in Latvia did we find a significant relationship between father’s employment status (not
working outside the home) and higher rates of abuse, it seems likely that in all four countries the difficult
socioeconomic situation in the rural areas is a serious factor affecting the psychological climate within
the family.
The results also show that in three of the countries there is higher rate of alcohol overuse in the rural
areas in comparison to the big cities (Modova data were not registered by region). In all four countries
correlations were found between parental oversure of alcohol and emotional and/or physical abuse. This
relationship is similar to that found in Western countries (Chaffin, Kelleher, & Hollenberg, 1996). The
fact that alcohol overuse and emotional and physical child abuse are related has implications for the need
for coordinated services between alcohol rehabilitation programs and child abuse treatment programs.
Of note is that the rate of parental alcohol overuse reported by the Latvian children (12%) is similar to
the rate of “about 10%” cited by specialists at the major alcohol and narcotics rehabilitation center of
Latvia (Caunitis, 2002, personal communication), but lower than the rate of 15–19% of adults reporting
problems related to alcohol overuse cited in a recent sociological study in Latvia (Koroleva & Rungule,
2002).
125
There are a number of limitations to this study. The participants of this study are not completely
representative of all children in each respective country. This study took place in the classroom, and
only those children present on the day of the distribution of the questionnaires participated in the study.
Certainly, in each of these countries there are considerable numbers of children not attending school
regularly, children who often are living in more dysfunctional family situations where rates of emotional or
physical abuse are most likely to be higher. Secondly, the fact that these questionnaires were administered
in a group format may have affected the openness of the children. In addition, alcohol overuse was assessed
with only one question, and in future studies it would be advisable to pose several questions concerning
parental overuse of alcohol in order to strengthen the reliability of this factor. Future studies should also
aim to obtain reports regarding the abuse experienced by children not attending school, and consider
possibilities for comparing the reports of multiple informants.
For purposes of planning future treatment and prevention programs this cross-cultural comparison
allows us to consider more specifically which abusive behaviors are more prevalent and correlate most
strongly with symptoms, in order to target these aspects. Identification of similar risk factors, such as
parental alcohol abuse and the greater risk of abuse in the rural areas, provides the basis for joint discussion
of prevention programs between team members from different countries, specific targeting of at-risk
families for preventive measures, and joint efforts between abuse prevention and alcohol rehabilitation
programs.
In summary, in each country of this cross-cultural comparison there were considerable percentages
of children who reported emotional and physical abuse, and these abuse ratings correlated significantly
with all of the assessed psychosocial symptoms, thereby adding construct validity to these measures as
applied cross-culturally. Parental alcohol overuse and living in a rural area were identified as risk factors
these countries.
References
Agathonos-Georgopoulous, H. (1992). Cross-cultural perspectives in child abuse and neglect. Child Abuse Review, 1, 80–
88.
Atwood, L. (1990). The new Soviet man and woman: Sex-role socialisation in the USSR. Bloomington, IN: Indiana University
Press.
Bonevski, D., & Novotni, A. (2000). Difference in patterns of intrafamilial abusive behavior, and trauma symptoms in children between urban and rural families in Macedonia. Paper presented at the Fifth International Conference on Family Violence, Advocacy, Assessment, Intervention, Research, Prevention and Policy, September, 2000, San Diego, CA,
USA.
Briere, J. (1995). Trauma Symptom Checklist for Children: Professional manual. Odessa, FL: Psychological Assessment Resources.
Bronfenbrenner, U. (1970). Two worlds of childhood: US and USSR. New York: Russel Sage Foundation.
Chaffin, M., Kelleher, K., & Hollenberg, J. (1996). Onset of physical abuse and neglect: Psychiatric, substance abuse, and social
risk factors from prospective community data. Child Abuse & Neglect, 20(3), 191–203.
Du Plessix Gray, F. (1989). Soviet women: Walking the tightrope. New York: Doubleday Press.
Garbarino, J. (1999). Lost boys. New York: Free Press.
Hussey, D., & Singer, M. (1993). Sexual and physical abuse: The Adolescent Sexual Concern Questionnaire (ASCQ). In M. I.
Singer, L. T. Singer, & T. M. Anglin (Eds.), Handbook for screening adolescents at psychosocial risk (pp. 131–163). New
York: Lexington Books.
Kerig, P. K., Alyoshina, Y. Y., & Volovich, A. S. (1993). Gender-role socialization in contemporary Russia. Psychology of Women
Quarterly, 17, 389–408.
STUDY I
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
126
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
Korbin, J. (1991). Cross-cultural perspective and research directions for the 21st century. Child Abuse & Neglect, 15, 67–77.
Koroleva, R., & Rungule, R. (2002). Rı̂gas pilsçtas iedzı̂votâju alkoholisko dzçrienu lietoanas paradumi un attieksme pret
alkohola izplatı̂bas ierobe ojumiem (“Alcoholic Beverage Use and Attitudes Towards Alcohol Use Restrictions for City of
Riga Inhabitants”). Rı̂ga: LU Filozofijas un Socioloı̀ijas Institûts.
Lewis, O., Sargent, J., Friedrich, W., Chaffin, M., Cunningham, N., & Sicher Cantor, P. (2001). The impact of social change on
child mental health in Eastern Europe. Child and Adolescent Psychiatric Clinics of North America, 10(4), 815–824.
Lieven, A. (1993). The Baltic Revolution: Estonia, Latvia, Lithuania and the path to independence. New Haven: Yale University
Press.
Pearson, L. (1990). Children of Glasnost: Growing up Soviet. Toronto: Lester & Orpen Dennys Limited.
Sebre, S. (2000). Dissociation, hostility, resentment and intolerance in relation to familial and political abuse. Journal of Baltic
Psychology, 1(1), 27–45.
Sebre, S., Sprugevica, I., Zagare, I., & Sluka, D. (1998). Risk factors and symptoms associated with child abuse in Latvia. Paper
presented at the International Congress of the International Association for Child and Adolescent Psychiatry and Allied
Professions, Stockholm, Sweden.
Sicher, P., Lewis, O., Sargent, J., Chaffin, M., Friedrich, W. N., Cunningham, N., Thomas, R., Thomas, P., & Villani, S. V.
(2000). Developing child abuse prevention, identification, and treatment systems in Eastern Europe. Journal of the American
Academy of Child & Adolescent Psychiatry, 39(5), 660–667.
Straus, M. A. (1995). Manual for the Conflict Tactics Scales. Durham, NH: Family Research Laboratory, University of New
Hampshire.
Šebek, M. (1994). The true self and the false self: The clinical and social perspective. Journal of the British Association of
Psychotherapy, 26, 22–39.
Vanderlinden, J., Varga, K., Peuskens, J., & Pieters, G. (1995). Dissociative symptoms in a population sample of Hungary.
Dissociation, 8(4), 205–208.
Résumé
Objectif: Cette étude a voulu mesurer l’incidence des mauvais traitements émotionnels et physiques,
les facteurs qui y sont associés et les symptômes psychosociaux dans le contexte d’une comparaison de
diverses cultures dans certains pays anciennement du bloc communiste.
Méthode: Mille cent quarante-cinq enfants âgés de 10-14 ans, venant de la Lettonie (N = 297), de la
Lituanie (N = 300), de la Macédoine (N = 302) et de la Moldovie (N = 246) ont participé à l’étude. Ils
ont complété un questionnaire cherchant à évaluer les mauvais traitements physiques ou émotionnels et à
fournir des renseignements sur leurs symptômes psychosociaux, y compris les symptômes du syndrome
du stress post traumatique, et les facteurs de risque présents dans leurs familles.
Résultats: Les taux de maltraitance varient d’un pays à l’autre, tout comme les niveaux des symptômes
psychosociaux. Les taux de mauvais traitements émotionnels et physiques varient d’une région à l’autre,
c.-à-d. qu’il existe un taux plus élevé dans les régions rurales. Dans les quatre pays on retrouve une
relation semblable entre les mauvais traitements et les symptômes; la relation la plus définitive est celle
entre les mauvais traitements émotionnels et la colère. Lorsqu’on combine les scores des deux types de
mauvais traitements, la corrélation est encore plus prononcée, surtout en ce qui a trait à la colère et la
dépression. Dans les quatre pays, on retrouve un lien entre la consommation excessive d’alcool et les
deux types de mauvais traitements.
Conclusions: On constate des différences entre les pays en ce qui a trait à l’incidence des mauvais
traitements, cependant on note des ressemblances par rapport à la corrélation entre la maltraitance et
les symptômes psychosociaux, la consommation excessive d’alcool et le fait de vivre dans un milieu
rural.
127
Resumen
Objetivos: Este estudio se diseñó para evaluar la incidencia del maltrato emocional y fı́sico, factores
de riesgo asociados, y sı́ntomas psicosociales en una comparación entre culturas de paı́ses del bloque
post-comunista.
Método: Un total de mil ciento cuarenta y cinco niños de edades comprendidas entre 10 y 14 años formaron parte del estudio. Estos niños procedı́an de Letonia (N = 297), Lituania (N = 200), Macedonia
(N = 302), y Moldavia (N = 246). Completaron cuestionarios que evaluaban su experiencia de maltrato emocional o fı́sico, y recogı́an información sobre los factores de riesgo familiares y los sı́ntomas
psicosociales, incluidos sı́ntomas relacionados con el PTSD.
Resultados: Las tasas de incidencia de maltrato variaban de un paı́s a otro, ası́ como la notificación de
sı́ntomas psicosociales. La incidencia de maltrato emocional y fı́sico varió en cada región, con niveles más
altos de notificación de maltrato en regiones rurales. En los cuatro paı́ses, se dio una asociación similar
entre el maltrato fı́sico/emocional y los sı́ntomas psicosociales, con una correlación mayor y uniforme
entre maltrato emocional e ira. Cuando se examinaron las puntuaciones combinadas de maltrato emocional
y fı́sico, se encontraron correlaciones mayores, particularmente en relación a la ira y a la depresión. En
los cuatro paı́ses, el abuso paterno de alcohol estuvo asociado al maltrato emocional y/o fı́sico.
Conclusiones: Los hallazgos muestran diferencias entre paı́ses en niveles de notificación de maltrato
emocional y fı́sico, pero patrones similares de correlación con sı́ntomas psicosociales y el riesgo de
factores de abuso paterno de alcohol en un área rural.
STUDY I
S. Sebre et al. / Child Abuse & Neglect 28 (2004) 113–127
STUDY II
Klunserna Study
123
05-10-25, 10.12
Vilmante Pakalniskiene, Margaret Kerr, & Håkan Stattin
Örebro University
STUDY II
Youth Characteristics as Explanations of the Link Between
Negative Parenting Practices and Adolescent Peer
Relationship Quality
Abstract
Negative parenting has been linked to problems in peer relationships for children and
adolescents. Youth characteristics have been linked to both negative parenting practices and
peer relationships, but they have not been examined as possible explanations for the link
between negative parenting practices and peer relationship quality. In this study, we
examined whether youths’ characteristics such as internalizing problems or psychopathylike personality traits might play a role in parents’ negative practices and the quality of
youths’ peer relationships, thus explaining the link between them. Participants were 663
adolescents, aged 13-15. Individuals and peers reported independently on the quality of
their relationships. In the initial analyses, negative parenting practices were related to
poorer peer relationships. However, cross-lagged longitudinal models suggested that youth
characteristics contributed to both negative parenting practices and conflict in peer
relationships. After controlling for youth characteristics, negative parenting practices were
still related to peer relationships cross-sectionally but not longitudinally, and even in the
cross-sectional models, psychopathy-like traits were the strongest predictor. From peers’
reports, psychopathy-like traits consistently predicted relationship quality. These findings
suggest that one cannot simply say that parental behavior interferes with youths’ abilities to
develop close, supportive peer relationships; youths’ characteristics play important roles in
both parenting and peer relationships.
Keywords: adolescents, negative parenting practices, youth characteristics, peer
relationships
Youth characteristics as explanations of the link between
negative parenting practices and adolescent peer
relationship quality
STUDY II
Close relationships are important aspects of life. Parents provide the first
experiences in close relationships, but by adolescence, peers become more important than
family as confidants and providers of emotional support (Bagwell, Newcomb, & Bukowski,
1998; Parker, Rubin, Price, & DeRosier, 1995). Both attachment theory and social learning
theory suggest that youths might carry expectations or behaviors over from their
relationships with their parents that would affect their peer relationships. This suggests that
if children have experienced negative parenting behavior at home, they might have trouble
establishing and maintaining supportive peer relationships.
Although the relations tend to be modest, a number of studies link various
indicators of negative parenting behavior with youths’ problems in peer relationships.
Much of this research deals with young children, but some deals with adolescents, and the
results are similar. Negative parenting behaviors such as love withdrawal, harsh discipline,
strictness, and verbal and symbolic aggression have been linked with poor quality in
children’s and adolescents’ peer relationships, extreme peer orientation, antisocial activities
with peers, and trouble making friends (Dekovic & Meeus, 1997; Engels, Decovic, &
Meeus, 2002; Fuligni & Eccles, 1993; Lasford, Criss, Pettit, Dodge, & Bates, 2003;
Vissing, Straus, Gelles, & Harrop, 1991). Thus, negative parental behavior or practices in
the literature are linked to problematic peer relationships for children and adolescents.
Consistent with attachment and social learning theories, the results have been interpreted as
showing that the experience of negative parenting practices produces expectations or
shapes behaviors that interfere with youths’ abilities to form or maintain supportive peer
relationships.
Few studies have considered characteristics of the youths, themselves, in the link
between negative parenting practices and poor peer relationships, but there are both
theoretical and empirical reasons for doing so. Patterson’s (1982) coercion theory, which is
based on social learning theory, suggests that children who are disruptive or aggressive can
elicit harsh, ineffective parenting behaviors and also tend to be rejected by peers because of
their behavior. Consistent with this, numerous experimental and longitudinal studies have
shown that adults react negatively to various types of children’s externalizing problems
(e.g., Anderson, Lytton, & Romney, 1986; Buss, 1981; Dix, Ruble, Grusec, & Nixon, 1986;
Mulhern & Passman, 1981; Passman & Blackwelder, 1981). There is less research on
adolescents, but the findings are similar. Externalizing behavior, antisociality, and
undesirable characteristics such as lying and manipulation can affect parenting negatively
(Ge, Conger, Cadoret, Neiderhiser, Yates, Troughton, & Stewart, 1996; Kerr & Stattin,
2003). For internalizing problems, the findings are mixed (c.f., O’Connor, Deater-Deckard,
Fulkel, Rutter, & Plomin, 1998; Rogers, Buchanan, & Winchell, 2003), but there is some
evidence that internalizing problems might evoke negative parental behavior. Concerning
peer relationships, longitudinal research has connected personality traits such as
extraversion, shyness, neuroticism, and agreeableness (Asendorpf & Wilpers, 1998; Neyer
& Asendorpf, 2001) and internalizing problems such as depressed mood (Nolan, Flynn, &
Garber, 2003; Prinstein, Borelli, Cheah, Simon, & Aikins, 2005; Stice, Ragan, & Randall,
2004) and self-esteem (Neyer & Asendorpf, 2001) with changes over time in the quality of
adolescents’ and young adults’ social relationships. Thus, there is evidence that
externalizing and internalizing behaviors can influence both negative parenting practices
and peer relationship quality. To our knowledge, however, no study has looked at whether
youth characteristics might explain the link between negative parenting practices and poor
peer relationship quality.
There have been a few studies that have included all three of these aspects—youth
characteristics, parenting or parent-child relationships, and peer relationships (Clark &
Ladd, 2000; Hinshaw, Zupan, Simmel, Nigg, & Melnick, 1997; Keown & Woodward,
2006; Lansford, Criss, Pettit, Dodge, & Bates, 2003; Simons, Chao, Conger, & Elder,
2001). They have focused mainly on children rather than adolescents, however, and they
did not examine the directions of effects that we are suggesting. Two of these studies
looked at child characteristics and mother-child interactions or mothers’ beliefs about
parenting as equal predictors of peer relationship problems (Hinshaw et al., 1997; Keown &
Woodward, 2006). They were both cross-sectional studies, however, and both used only
preadolescent boys. Another study tested a mediation model and positive instead of
problematic characteristics in children (Clark & Ladd, 2000). It was also cross-sectional,
and negative parenting practices were not part of the study. Two other studies used
longitudinal data to examine mediation or moderation models with parenting, youth
characteristics, and peer relationships (Lansford et al., 2003; Simons et al., 2001). Again,
peer relationship quality was not the outcome of interest in either of these studies and
alternative directions of effects were not tested. These studies do, however, suggest that it is
important to consider youth characteristics when examining relationships between
parenting and peer relationships.
In this study, we ask whether links between negative parenting practices and
adolescents’ peer relationship quality might be explained by youths’ internalizing problems
Method
Participants
Data are from two waves of a longitudinal study that is taking place in a community
of about 26,000. The present analyses involve participants who were in 7th through 9th
grades (ages 13-15) at the first of the two waves (Time 1). The second wave (Time 2) took
place one year later. One thousand forty-six youths (or 94% of the target sample in these
grades) participated at Time 1. Because these analyses involved close peer relationships, we
STUDY II
and psychopathy-like personality traits. As measures of negative parenting practices, we
use angry outbursts and coldness-rejection, both of which have been used previously (Isley,
O’Neil, Clatfelter, & Parke, 1999; Kim, Conger, Elder, & Lorenz, 2001; Rueter & Conger,
1998; Straus & Field, 2003). First, we look at whether negative parenting practices are
linked to peer relationship quality, defined as conflict and support and trust in close
relationships. Then, we examine whether youth characteristics might influence negative
parenting practices and also create conflict or interfere with support and trust in peer
relationships, thus explaining the links between negative parenting practices and peer
relationship quality. We consider two categories of youth characteristics: internalizing
problems and psychopathy-like personality traits. As mentioned above, internalizing
problems have been linked in separate literatures to peer relationship problems and some
forms of harsh parental treatment. Thus, they offer a possible explanation for the link
between negative parenting practices and poor peer relationships. The psychopathy-like
personality profile involves three dimensions, two of which (grandiose-manipulative and
callous-unemotional) directly involve unsupportive behavior in relationships with others
and the other of which (irresponsible-impulsive) is similar to externalizing problems, which
have also been connected in research and coercion theory with negative parenting practices
and poor peer relationships. Thus, like internalizing problems, these characteristics offer a
possible explanation for the link between parents’ harsh behavior and poor peer
relationships. We examine how youth characteristics are related to negative parenting
practices and peer relationship quality cross-sectionally and longitudinally, and for the
longitudinal analyses we examine both directions of effects. We then control for youth
characteristics when examining the link between negative parenting practices and peer
relationship quality. In each of the analyses involving peer relationships, we try to verify
the findings with youths’ reports of peer relationship quality by using peers’ independent
reports.
limited the sample to those who participated at both time points and had reported on a
relationship with a close peer at Times 1 and 2 (n = 797). To determine whether those who
participated at both times differed from those who did not, we compared participants with
complete peer relationship data and participants with only Time-1 data (171), participants
with only Time-2 data (40), and participants who did not answer questions about peer
relationships at either time point (38) on all of the other measures used in analyses. These
groups did not differ significantly on any of the variables examined. Furthermore, as we
will describe below, youths were free to name a sibling as their closest peer, and in order to
keep parents’ behavior and close peer relationships independent in these analyses we
eliminated those who named a sibling at either of the two times (n = 134). Thus, these
analyses include 663 adolescents—346 boys (52.2%) and 317 girls (47.8%). At Time 1,
29% of parents were divorced or separated. Sixty-seven percent of participants lived with
both mother and father; 13% lived with mother only; 11% with father only; 10% lived
mother and stepfather; 6% with father and stepmother; 2% lived with other relatives or with
someone else. The demographic characteristics at Time 2 were very similar.
Youths were recruited in their classrooms during school hours. They were given a
description of the study and informed that participation was voluntary. Parents were
informed about the study ahead of time in meetings held in the community and by mail.
They could send in a prepaid postcard if they did not want their youth to participate (1%
did so). Adolescents filled out the questionnaires during regular school hours in sessions
administered by trained research assistants. Teachers were not present. Youths were not
paid for their participation.
Measures
Negative Parenting Practices
To measure the two aspects of negative parenting practices – angry outbursts and
coldness-rejection – we used youths’ responses to 11 relevant statements about how their
parents typically responded to wrongdoing. There were three response options, ranging
from “never” to “most often.” There were 5 items for angry outbursts and 6 items for
coldness-rejection, and youths answered each item for their mothers and fathers separately.
For these analyses we used the mean of angry outbursts and coldness-rejection. Also,
because reports for mothers’ and fathers’ behaviors were substantially correlated (r (580) =
.68, p < .01 and r (582) = .62, p < .01 for angry outbursts and coldness-rejection,
respectively), we averaged them if we had reports from both parents. The stem question for
all of these items was: “What happens if you have done something your parent really
From parents’ points of view, angry outbursts and coldness-rejection could be
responses to undesirable youth behavior. Thus, to examine criterion-related validity of the
youth-reported negative parenting measure, we correlated it with parents’ reports of youth
behaviors that might elicit negative parenting. Youth-reported negative parenting correlated
significantly with parents’ reports of the youths’ insensitivity to punishment, r (539) = .30,
p = .001 (e.g., “Although we parents rebuke him or her for a specific behavior, he or she
continues with it”); callous, unemotional traits, r (542) = .25, p = .001 (e.g., “Seldom
expresses remorse when he or she has done something that we parents consider wrong”);
hyperactivity-impulsivity-attention problems, r (542) = .24, p = .001 (e.g., “Often interrupts
others or intrudes on their conversation or activities”); and off-task behavior, r (539) = .20,
p = .001 (e.g., “He or she begins doing many things, but has a hard time finishing them”).
Peers
Assessment of Peers
We asked youths to name three important peers in order of importance to them. We
defined an important peer as, “someone you talk to, hang out with, and do things with. It
cannot be your parents or another adult. It could, for example, be a friend, a sibling, or a
boyfriend or girlfriend.” We explained, further, that this important peer could live
anywhere, did not have to be the same age as the participant, and could be either a boy or a
girl. Here, we focus on the first-mentioned, or most important, peer. For 77% of
participants, the most important peer was a friend, for 11% it was a sibling and for 7% it
was a romantic partner. As mentioned above, adolescents who named a sibling as their
most important peer at Time 1 (73 participants) or at Time 2 (40 participants) were
excluded from the analyses. For most participants, the most important peer at Time 1 was
different from the most important peer at Time 2. Two hundred eleven youths named the
same person at both times.
Relationships with Peers
STUDY II
dislike?” Youths rated the statements below. For angry outbursts, some examples are:
“Becomes very angry and has an outburst,” “Has a hard time controlling his or her
irritation,” “Screams and yells at you.” The alpha reliabilities for this scale were .87 at
Time 1 and .89 at Time 2. For coldness-rejection, examples are: “Is silent and cold towards
you,” “Makes you feel guilty for a long time,” “Avoids you.” The alpha reliabilities for this
scale were .77 at Time 1 and .83 at Time 2. These scales were significantly correlated with
each other at Time 1 (r = .70, p < .001) and Time 2 (r = .62, p < .001).
Youths answered 12 questions about their relationships with their first-mentioned,
or most important, peers. There were five response options, ranging from “do not agree at
all” to “agree perfectly.” Questions were taken from Parker and Asher’s (1993) Friendship
Quality Questionnaire. The questions were about conflicts and caring in relationships or the
friend’s behavior in the relationship. Principal-components analyses of the 12 variables
showed two clear factors, which we labeled “support and trust” and “conflict.” The factor
loadings ranged from .64 to .90 and the cross-loadings ranged from -.01 to -.20.
Support and trust. Some examples of items are: “Says I’m good at different
things,” “makes me feel that I have good ideas,” and “doesn’t tell my secrets to others.”
The alpha reliabilities for this scale were .86 at Time 1 and .89 at Time 2.
Conflict. Some examples of items are: “We often get angry with each other,” “We
argue a lot,” and “We often get annoyed with each other.” The alpha reliabilities for this
scale were .90 at Time 1 and .91 at Time 2.
Adolescents’ Characteristics
Psychopathy-Like Personality Traits
We used a youth self-report instrument designed to capture subclinical levels of
these personality traits in community samples of youths 12 years and older, the Youth
Psychopathic traits Inventory (YPI: Andershed et al., 2002). Reliability and construct
validity of this instrument have been reported elsewhere (Andershed, Hodgins, &
Tengström, in press; Dolan, & Rennie, 2006a; Dolan, & Rennie, 2006b; Poythress, Dembo,
Wareham, & Greenbaum, 2006; Skeem & Cauffman, 2003). For this study we used a total
YPI score, which was calculated as a mean value of the scores for three dimensions. The
grandiose, manipulative dimension comprises 20 items, equally divided among four
subscales: Dishonest Charm, Grandiosity, Lying, and Manipulation. Examples of the
questions are: “I have the ability to con people by using my charm and my smile,” “I am
better than everyone else,” “Sometimes I find myself lying without any particular reason.”
The alpha reliabilities were .85 at Time 1 and .85 at Time 2. The callous, unemotional
dimension comprises 15 items from three subscales: Unemotionality, Remorselessness, and
Callousness. Some examples of items are: “I think that crying is a sign of weakness, even if
no one sees you,” “I usually feel calm when other people are scared,” and “I have the
ability not to feel guilt and regret about things that other people would feel guilty about.”
The alpha reliabilities for this dimension were .74 at Time 1 and .79 at Time 2. The
impulsive, irresponsible dimension includes 15 items for Impulsiveness, Thrill-Seeking,
and Irresponsibility. Examples of the questions are: “I prefer to spend my money right
away rather than save it,” “I like to be where exciting things happen,” and “I have probably
skipped school or work more than most other people.” The alpha reliabilities were .77 at
Time 1 and .77 at Time 2. These three dimensions were significantly and substantially
correlated with each other at Time 1 and Time 2 (rs from .54 to .78, ps < .001).
We used youths’ responses to relevant statements about self-esteem, depressed
mood, and failure expectations. For this study we used a mean value of three scales, which
were significantly and substantially correlated with each other at Time 1 and Time 2 (rs
from .39 to .61, ps < .001). To measure self-esteem, we used the Rosenberg Self-Esteem
Scale (Rosenberg, 1979), consisting of ten statements rated on a 4-point scale from “do not
agree at all” to “agree totally.” The alpha reliabilities for this scale were .89 at Time 1 and
.89 at Time 2. The measure of depressed mood was the Child Depression Scale from the
Center for Epidemiological Studies (Faulstich, Carey, Ruggiero, Enyart, & Gresham, 1986;
Roberts, Lewinsohn, & Seeley, 1991), consisting of twenty questions rated on a 3-point
scale from “not at all” to “often.” The alpha reliabilities for this scale were .89 at Time 1
and .91 at Time 2. Youths also reported on their expectations of failure on difficult tasks
(Nurmi, 1993; Nurmi, Onatsu, & Haavisto, 1995). They evaluated four statements using a
4-point scale from “do not agree at all” to “agree totally.” The statements included: “I don’t
have faith in my ability to cope with hard tasks” and “Often I don’t even think there is any
point in trying when I face demanding tasks,” The alpha reliabilities for this scale were .74
at Time 1 and .71 at Time 2.
Analyses
Correlational and multiple regression strategies were used with cross-sectional and
longitudinal data. For the longitudinal analyses involving peer relationship quality, we
formed a dummy variable to distinguish youths who named the same person as their most
important peer at both times (n = 211) from youths who named different peers at the two
time points (n = 442). In all of these analyses, we included interaction terms to test whether
having the same peer at both times made a difference in the results. Differences are
presented separately for the two groups. Also, because internalizing problems and
psychopathy-like traits might be comorbid, we included interactions of these two in the
models. None of these interactions was significant, so they were removed.
STUDY II
Adolescents’ Internalizing Problems
Results
Links Between Negative Parenting Practices and Peer Relationships
The intercorrelations between all variables in the study within time points are shown
in Table 1, with Time 1 above the diagonal and Time 2 below. As shown in the first row
and the first column, at both time points negative parenting practices were significantly but
modestly associated with youth-reported support and trust. These cross-sectional links,
presented in Table 1, are similar in magnitude to those reported in previous studies. At
Time 1, negative parenting was also related to peers’ independent reports of conflict in the
relationship. This link was not found at Time 2, however, and no links were found for
support and trust at either time.
The longitudinal analyses appear in the first row of Table 2. In these analyses,
negative parenting practices at Time 1 significantly predicted self-perceived conflict in peer
relationships at Time 2, controlling for the Time-1 measure. This suggests that perceived
negative parenting practices contribute to changes over time in perceived conflict in close
peer relationships. This was not substantiated with peers’ independent reports of support
and conflict, however. Overall, then, there is evidence that perceived negative parenting is
linked to perceived conflict in close peer relationships concurrently and to increased
perceived conflict in close peer relationships over time. From the peers’ points of view,
there is some support for this cross-sectionally but not longitudinally.
The Role of Youth Characteristics
To examine whether youth characteristics might evoke negative parenting practices
and also interfere with relationship quality, we looked first at whether and how youth
characteristics are related to negative parenting practices. As shown in the upper part of
Table 1, both youth characteristics were significantly correlated with self-perceived
negative parenting practices at Times 1 and 2. The longitudinal results with youth
characteristics as predictors of negative parenting and peer relationship quality appear in
the second and third rows of Table 2. As shown in the first column, both internalizing
problems and psychopathy-like personality traits significantly predicted increases in
perceived negative parenting practices from Time 1 to Time 2, and as shown in the third
column, both youth characteristics predicted increases in perceived conflict in relationships
with important peers from Time 1 to Time 2. For conflict, this was significant for youths
who named different close peers at the two times, thus suggesting that at Time 2 they
perceived more conflict with their closest peer than they had with an earlier close peer.
Again, however, this was not found when peers’ independent ratings of conflict in the
STUDY II
relationship were used longitudinally. Thus, there is evidence that internalizing problems
and psychopathy-like traits might evoke negative parenting and create conflict in peer
relationships from the youths’ points of view, thus providing a possible explanation for the
link between self-reports of negative parenting and peer relationship quality. In the next set
of analyses, we test this explanation.
To determine whether youth characteristics explain the apparent link between
negative parenting practices and peer relationship quality, we controlled for the two youth
characteristics in models predicting peer relationship quality from negative parenting
practices. The results appear in Table 3. As shown in the upper part of the table, after
controlling for youth characteristics, negative parenting practices still remains a significant
predictor of self-perceived support and trust and conflict in relationships at Time 1 and of
conflict at Time 2. The beta coefficients, however, are only about half of what they were
before. But negative parenting practices were not related to peers’ independent reports of
either support and trust or conflicts at either time. Both youth characteristics also predicted
perceived support and trust, and psychopathy-like traits predicted perceived conflict.
Psychopathy-like traits were linked to lower support and trust at both Times 1 and 2, and
conflicts at Time 1, according to peers’ independent reports. Cross-sectionally, then,
controlling for youth characteristics did not completely eliminate the links between
negative parenting and self-perceived peer relationship quality, but it did eliminate the
weaker link between negative parenting and peer-reported relationship conflict. Youth
characteristics, particularly psychopathy-like traits, showed the most consistent links to
relationship quality from both individuals’ and peers’ perspectives.
As shown in the lower part of the table, longitudinally there were no unique links
between negative parenting practices and self-perceived relationship quality after
controlling for youth characteristics, but internalizing problems predict increases over time
in perceived support and trust in peer relationships and psychopathy-like traits uniquely
predict increases over time in perceived conflict in relationships. For the longitudinal
analyses predicting changes in peers’ views of relationship quality, we focused only on
youths who were selected by the same peer at both time points. This insured that
differences over time represented changes in a peers’ perception of the relationship rather
than differences between one peer and another. In these analyses, there was no significant
link between negative parenting and peer relationship quality, but psychopathy-like
personality traits significantly predicted increases in peers’ perceptions of conflicts. Thus,
when youths’ reports of peer relationships are used, youth characteristics explain the
longitudinal but not the cross-sectional links between negative parenting and peer
relationship quality. When peers’ reports of relationship quality are used youth
characteristics explain both the longitudinal and the cross-sectional links between selfperceived negative parenting and peer-reported relationship quality.
The results that are supported by both youths’ and peers’ perceptions of relationship
quality suggest that psychopathy-like traits play a role in evoking negative parenting and
creating peer relationship problems. One might ask, however, whether there are reciprocal
relations in which negative parenting and experiences in peer relationships contribute to
these traits. To test this idea, in a final set of analyses we predicted changes in youth
characteristics (Time-2 measures controlling for Time-1 measures) from negative parenting
and the two measures of peer relationship quality. These longitudinal results are reported in
Table 4. As shown in the table, negative parenting practices predicted increases over time
in internalizing problems, but did not predict changes over time in psychopathy-like traits.
In these analyses, there was no significant link between peer-rated relationship quality and
youth characteristics, but self-rated support and trust in peer relationships significantly
predicted decreases in youth internalizing problems and self-perceived conflicts
significantly predicted increases in youth psychopaty-like traits. Thus, there are some
reciprocal relations between self-perceived negative parenting practices, experiences in
peer relationships, and youth characteristics.
Discussion
Children who experience negative parental behavior have been found to have
problems relating to peers (Dekovic & Meeus, 1997; Engels, Decovic, & Meeus, 2002;
Fuligni & Eccles, 1993; Isley, O’Neil, Clatfelter, & Parke, 1999; Schwartz, Dodge, Pettit,
Bates, & The Conduct Prevention Research Group, 2000; Vissing, Straus, Gelles, &
Harrop, 1991). Although several studies have suggested an interplay between youth
characteristics, positive or negative parenting, and different aspects of peer relationships
(Clark & Ladd, 2000; Keown & Woodward, 2006; Simons, Chao, Conger, & Elder, 2001),
until now youth characteristics have not been considered as a possible explanation for both
negative parenting practices and poor peer relationships. Our findings suggest that negative
parenting practices might be, in part, maladaptive parenting responses to adolescents’
characteristics and behaviors, and these same characteristics and behaviors might be
responsible for problematic peer relationships. Thus, one cannot say simply that negative
parental behavior interferes with youths’ abilities to develop close, supportive peer
relationships; youths’ characteristics play important roles in relationships with both parents
and peers.
STUDY II
In recent years, there has been a growing awareness that reciprocal processes are
important for understanding interactions between parents and children. The idea that
parents react to children’s characteristics and children, in turn, react to parents’ behaviors
was suggested early on in coercion theory (Patterson, 1982). However, although reciprocal
effects are widely accepted in principle, they are still too seldom included in research
designs. This study is one of a growing number that includes and tests alternative directions
of effects.
There are several studies in the literature that have considered linkages among
youth characteristics, parenting, and peer relationships or interactions (Clark & Ladd, 2000;
Hinshaw, et al.,1997; Keown & Woodward, 2006; Lansford et al., 2003; Simons, et al.,
2001), but they differ from the present study in several important ways. Three of these
studies were cross-sectional, which makes it difficult to infer directions of effects. In
addition, most have looked at children rather than adolescents, which makes it difficult to
generalize to adolescent peer relationships. This study is unique in examining whether
youth characteristics might underlie both self- perceived negative parenting practices and
friends’ reported or self-perceived poor peer relationship quality, thus explaining the
relation between them.
One of the main strengths of this study was the use of peers’ reports in addition to
participants’ self-reports of relationship quality. This allowed us to identify the links that
are robust across informants as well as across waves. Considering only the findings that
reached significance at both waves and for both youths’ and peers’ reports of relationship
quality, the clearest conclusion is that they all involved links between psychopathy-like
traits and relationship quality. This suggests that these characteristics deserve more
attention in respect to peer relationships.
Both internalizing problems and psychopathy-like traits are positively linked to
negative parenting practices cross-sectionally and longitudinally, and in that way they are
similar. For internalizing problems unlike psychopathy-like traits, however, the relations
are bidirectional, suggesting that parents react negatively to internalizing problems and
their behaviors also make them worse. The more striking difference is that internalizing
problems are positively linked to support and trust in relationships both cross-sectionally
and longitudinally, but only from the youths’ own points of view. Perhaps this is not
surprising, since youths were reporting on the support and trust gained from their peers. It
might be that youths with internalizing problems get support from their friends but do not
provide a lot in return.
There are some reasons to be cautious about over-interpreting these results. It is
important to distinguish between effects over time during a period of several years and
original causes of behavior. In this study, we are dealing with adolescence and explaining
effects over time in adolescence. Parents and children have histories of interactions,
however, and negative parenting practices might have played a causal role in shaping those
characteristics earlier in the child’s life. This is possible, and it is intuitively appealing to
think that parents play a more active role in shaping the behavior of young children than
adolescents. But whatever has happened earlier, our data do show changes over time in
adolescence and directional or bidirectional effects. This could be important for
understanding and offering advice to parents of adolescents.
The reliance on youths’ reports can be seen as a limitation, particularly because
adolescents reported on their own traits as well as negative parenting. Some scholars have
argued that youths’ views of family interactions are the most accurate or valid (Glasgow,
Dornbush, Troyer, Steinberg, & Ritter, 1997; Niemi, 1974), but it is possible that youths
with certain traits misperceive parents’ behavior in systematic ways, and these
misperceptions explain the links between youth characteristics and negative parenting.
There are, however, at least two arguments for the validity of our negative parenting
measure. One is that the links we find between negative parenting and peer relationships
are similar to those reported previously with data provided by children, parents, observers,
and multiple raters (Fuligni & Eccles, 1993; Lasford, Criss, Pettit, Dodge, & Bates, 2003;
Schwartz, Dodge, Pettit, Bates, & The Conduct Prevention Research Group, 2000).
Another is that youths’ reports of negative parenting in our study correlate with parents’
reports of youth behaviors that parents would likely find frustrating, and thus might
respond to with angry outbursts or icy silence. These correlations are substantial, given that
they are between two different raters and for different behaviors and different actors. All in
all, however, we should be aware that these data represent youths’ perceptions of parenting.
The present study has a number of strengths, however. One is the large community
sample. Studies done previously in the negative parenting practices area have usually tested
younger kids or adults. Another strength is the use of peers’ independent reports of
relationships quality. We have been able to tease apart whether youths with certain
characteristics might perceive their friendships in biased ways or whether youth
characteristics influence their friends’ perspectives of their relationships. A third strength is
the longitudinal design, which allowed us to tease apart the directions of effects between
negative parenting practices and youth characteristics.
It is quite natural to think that negative parenting practices would affect many
aspects of children’s lives, one of them being peer relationships. This may be so. At the
same time, however, parents do seem to react to adolescent characteristics. Adolescents are
active agents with qualities that help to shape relationships with peers and parents.
STUDY II
References
Andershed, H., Hodgins, S., & Tengström, A. (in press). Convergent validity of the Youth
Psychopathic Traits Inventory (YPI): Association with the Psychopathy Checklist:
Youth Version (PCL:YV). Assessment.
Andershed, H., Kerr, M., Stattin, H., & Levander, S. (2002). Psychopathic traits in nonreferred youths: A new assessment tool. In E. Blaauw & L. Sheridan (Eds.),
Psychopaths: Current international perspectives (pp. 131- 158). The Hague:
Elsevier.
Anderson, K, E., Lytton, H., & Romney, D. M. (1986). Mothers’ interactions with normal
and conduct-disordered boys: Who affects whom? Developmental Psychology, 22,
604-609.
Asendorpf, J. B., & Wilpers S. (1998). Personality effects on social relationships. Journal
of Personality and Social Psychology, 74, 1531-1544.
Bagwell, C. L., Newcomb, A. F., & Bukowski, W. M. (1998). Preadolescent friendship and
peer rejection as predictors of adult adjustment. Child Development, 69, 140-153.
Buss, D. M. (1981). Predicting parent-child interactions from children’s activity level.
Developmental Psychology, 17, 59-65.
Clark, K. E., & Ladd, G. W. (2000). Connectedness and autonomy support in parent-child
relationships: Links to children’s socioemotional orientation and peer relationships.
Developmental Psychology, 36, 485-498.
Dekovic, M., & Meeus, W. (1997). Peer relations in adolescence: Effects of parenting and
adolescents’ self-concepts. Journal of Adolescence, 20, 163-176.
Dix, T., Ruble, D. N., Grusec, J. E., & Nixon, S. (1986). Social cognition in parents:
Inferential and affective reactions to children of three age levels. Child
Development, 57, 879-894.
Dolan, N. C., & Rennie, C. E. (2006a). Reliability, validity, and factor structure of the
Swedish Youth Psychopathic Traits Inventory in a UK sample of conduct
disordered boys. Journal of Forensic Psychiatry & Psychology, 17, 217-229.
Dolan, N. C., & Rennie, C. E. (2006b). Psychopathy Checklist: Youth Version and Youth
Psychopathic Traits Inventory: A comparison study. Personality and Individual
Differences, 41, 779-789.
Engels, R. C. M. E., Dekovic, M., & Meeus, W. (2002). Parenting practices, social skills
and peer relationships in adolescence. Social Behavior and Personality, 30, 3-18.
STUDY II
Faulstich, M. E., Carey, M. P., Ruggiero, L., Enyart, P., & Gresham, F. M. (1986).
Assessment of depression in childhood and adolescence: An evaluation of the
Center for Epidemiological Studies Depression Scale for Children (CES-DC).
American Journal of Psychiatry, 143, 1024-1027.
Fuligni, A. J., & Eccless J. S. (1993). Perceived parent-child relationships and early
adolescents’ orientation toward peers. Developmental Psychology, 29, 622-632.
Ge, X., Conger, R. D., Cadoret, R. J., Neiderhiser, J. M., Yates, W., Troughton, E., &
Stewart, M. A. (1996). The developmental interface between nature and nurture: A
mutual influence model of child antisocial behavior and parent behaviors.
Developmental Psychology, 32, 574-589.
Glasgow, K. L., Dornbush, S. M., Troyer, L., Steinberg, L., & Ritter, P. L. (1997).
Parenting styles, adolescents’ attributions, and educational outcomes in nine
heterogeneous high schools. Child Development, 68, 507-529.
Hinshaw, S. P., Zupan, B. A., Simmel, C., Nigg, J. T., & Melnick, S. (1997). Peer status in
boys with and without attention-deficit hyperactivity disorder: Predictions from
overt and covert antisocial behavior, social isolation, and authoritative parenting
beliefs. Child Development, 68, 880-896.
Isley, S. L., O’Neil, R., Clatfelter, D., & Parke, R. D. (1999). Parent and child expressed
affect and children’s social competence: Modeling direct and indirect pathways.
Developmental Psychology, 35, 547-560.
Keown, L. J., & Woodward, L. J. (2006). Preschool boys with pervasive hyperactivity:
Early peer functioning and mother-child relationship influences. Social
Development, 15, 23-45.
Kerr, M., & Stattin, H. (2003). Parenting of adolescents: Action or reaction? In A.C.
Crouter & A. Booth (Eds.) Children's influence on family dynamics: The neglected
side of family relationships. Chapter 9 (pp. 121-151) Mahwah, NJ: Lawrence
Erlbaum.
Kim, K. J., Conger, R. D., Elder, G. H., & Lorenz, F. O. (2001). Parent-adolescent
reciprocity in negative affect and its relation to early adult social development.
Developmental Psychology, 37, 775-790.
Lansford, J. E., Criss, M. M., Pettit, G. S., & Bates, J. E. (2003). Friendship quality, peer
group affiliation, and peer antisocial behavior as moderators of the link between
negative parenting and adolescent externalizing problems. Journal of Research on
Adolescence, 13, 161-184.
Mulhern, R. K. Jr., & Passman, R. H. (1981). Parental discipline as affected by the sex of
the parent, the sex of the child, and the child’s apparent responsiveness to discipline.
Developmental Psychology, 17, 604-613.
Neyer, F. J., & Asendorpf, J. B. (2001). Personality-relationship transaction in young
adulthood. Journal of Personality and Social Psychology, 81, 1190-1204.
Niemi, R. G. (1974). How family members perceive each other: Political and social
attitudes in two generations. New Haven, CT: Yale University Press.
Nolan, S. A., Flynn, C., & Garber, J. (2003). Prospective relations between rejection and
depression in young adolescents. Journal of Personality and Social Psychology, 84,
745-755.
Nurmi, J.-E. (1993). Self-handicapping and a failure-trap strategy: A cognitive approach to
problem behavior and delinquency. Psychiatria Fennica, 24, 75-85.
Nurmi, J.-E., Onatsu, T., & Haavisto, T. (1995). Underachievers’ cognitive and behavioral
strategies – Self handicapping at school. Contemporary Educational Psychology,
20, 188-200.
O’Connor, T. G., Deater-Deckard, K., Fulkel, D., Rutter, M., & Plomin, R. (1998).
Genotype-environment correlations in late childhood and early adolescence:
Antisocial behavioral problems and coercive parenting. Developmental Psychology,
34, 970-981.
Parker, J. G., & Asher S. R. (1993). Friendship and friendship quality in middle childhood:
Links with peer group acceptance and feelings of loneliness and social
dissatisfaction. Developmental Psychology, 29, 611-621.
Parker, J. G., Rubin, K. H., Price, J. M., & DeRosier, M. E. (1995). Peer relationships, child
development, and adjustment: A development psychopathology perspective. In D.
Cicchetti & D. Cohen (Eds.), Developmental Psychopathology (Vol. 2, pp. 96-161).
New York: Wiley.
Passman, R. H., & Blackwelder, D. E. (1981). Rewarding and punishing by mothers: The
influence of progressive changes in the quality of their sons’ apparent behavior.
Developmental Psychology, 17, 614-619.
Patterson, G. R. (1982). Coercive family process. Eugene, OR: Castalia.
Poythress, N. G., Dembo, R., Wareham, J., & Greenbaum, P. E. (2006). Construct Validity
of the Youth Psychopathic Traits Inventory (YPI) and the Antisocial Process
Screening Device (APSD) With Justice-Involved Adolescents. Criminal Justice and
Behavior, 33, 26-55.
STUDY II
Prinstein, M. J., Borelli, J. L., Cheah, Ch. S. L., Simon, V. A., & Aikins, J. W. (2005).
Adolescent girls’ interpersonal vulnerability to depressive symptoms: A
longitudinal examination of reassurance-seeking and peer relationships. Journal of
Abnormal Psychology, 114, 676-688.
Roberts, R. E., Lewinsohn, P. M., & Seeley, J. R. (1991). Screening for adolescent
depression: A comparison of depression scales. Journal of the American Academy
of Child and Adolescent Psychiatry, 30, 58-66.
Rogers, K. N., Buchanan, C. M., & Winchell, M. E. (2003). Psychological control during
early adolescence: Links to adjustment in differing parent/adolescent dyads. Journal
of Early Adolescence, 23, 349-383.
Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.
Rueter, M. A., & Conger, R. D. (1998). Reciprocal influences between parenting and
adolescent problem-solving behavior. Developmental Psychology, 34, 1470-1482.
Schwartz, D., Dodge, K. A., Pettit, G. S., Bates, J. E., & The Conduct Problems Prevention
Research Group (2000). Friendship as a moderating factor in the pathway between
early harsh home environment and later victimization in the peer group.
Developmental Psychology, 36, 646-662.
Simons, R. L., Chao, W., Conger, R. D., & Elder, G. H. (2001). Quality of parenting as
mediator of effect of childhood defiance on adolescence friendship choices and
delinquency: A growth curve analysis. Journal of Marriage and Family, 63, 63-79.
Skeem, J. L. & Cauffman, E. (2003). Views of the Downward Extension: Comparing the
Youth Version of the Psychopathy Checklist with the Youth Psychopathic Traits
Inventory. Behavioral Sciences & the Law. 21, 737-770
Stice, E., Ragan, J., & Randall, P. (2004). Prospective relations between social support and
depression: Differential direction of effects for parent and peer support? Journal of
Abnormal Psychology, 113, 155-159.
Straus, M. A., & Field, C. J. (2003). Psychological aggression by American parents:
National data on prevalence, chronicity, and severity. Journal of Marriage and
Family, 65, 795-808.
Vissing, Y. M., Straus M. A., Gelles, R. J., & Harrop, J. W. (1991). Verbal aggression by
parents and psychosocial problems of children. Child Abuse and Neglect, 15, 223–
238.
.33
.35
-.10
.17
-.01
.09
**
***
*
***
-.21 ***
.27 ***
-.13 *
-.04
.37 ***
.18 ***
-.07
.05
-.02
-
.48 ***
.48 ***
.25 ***
Self-reported
Internalizing Psychopathyproblems
like traits
*p < .05. **p < .01. ***p < .001. (PR) peer reported.
Self-reported
Negative parenting
Internalizing problems
Psychopathy-like traits
Support & trust
Conflict
Support & trust (PR)
Conflict (PR)
Negative
parenting
-.30 ***
.29 ***
-.13 *
-
***
***
***
***
-.17 **
.38 ***
.21
-.15
.26
-.25
& Conflict
-.20 ***
.02
-.26 ***
Support
trust
Table 1.
Intercorrelations between Measures at Time 1 (Above the Diagonal) and Time 2 (Below the Diagonal).
-.33 ***
-
-.09
-.07
-.18 **
.28 ***
-.15 **
.11
.03
.13
-.14
.32
-.34
-
*
**
***
***
*
Peer-reported
Support & Conflict
trust
.17**
.09**
-.01
.03
-.04
Support & trust
.19**
-.05b/.20***c
.14***
Conflict
.01
.09
-.04
Support & trust
Peer-reported
.06
.06
.10
Conflict
Note. ª Controlling for Time 1. bYouths (n = 211) who named the same person as their most important peer at both time points.
c
Youths (n = 442) who named different peers at both time points. **p < .01;***p < .001.
Negative parenting practices
Internalizing problems
Psychopathy-like traits
Time 1
Negative parenting
Time 2a
Self-reported
Peer relationship quality Time 2a
Table 2.
Beta Coefficients from Regression Models Predicting Changes from Time 1 to Time 2 in Negative Parenting and Relationship Quality
from Time-1 Measures
STUDY II
-.12 *
.18 ***
-.32 ***
.04
.04
-.14
.02
.09
-.19
.02
.07
.10 *
*
-.03
.06
-.19 **
Support & trust
.09 *
-.02
.24 ***
.10 *
.01
.21 ***
Conflict
.02
-.12
.32 *
.08
-.01
.02
.04
-.07
.15 *
Conflict
Peer-reported relationship quality
Note. ª Predicting Time 2 relationship quality controlling for Time 1 relationship quality. bLongitudinal analyses—only for the group
where the same peer chose the individual at both time points (n = 117). *p < .05. **p < .01. ***p < .001.
Variables entered simultaneously T2
Negative parenting practices
-.02
Internalizing problems
.16 ***
Psychopathy-like traits
-.16 ***
Variables entered simultaneously with longitudinal data ª
Negative parenting practices
-.03
Internalizing problems
.09 *
Psychopathy-like traits
.01
Variables entered simultaneously T1
Negative parenting practices
Internalizing problems
Psychopathy-like traits
Support & trust
Self-perceived peer relationship quality
Table 3.
Simultaneous Regression Models Predicting Peer Relationship Quality from Negative Parenting Practices and Youth Characteristics
Table 4.
Beta Coefficients from Regression Models Predicting Youth Characteristics from Selfperceived Parenting and Self- and Peer-rated Relationship Quality Longitudinally
Time 1
Negative parenting practices
Support & trust
Conflict
Support & trust (PR)
Conflict (PR)
Internalizing problems
Psychopathy-like traits
.18 ***
-.07 *
.03
-.02
.02
.01
-.05
.11 **
-.02
-.05
*p < .05. **p < .01. ***p < .001. aControlling for Time 1. (PR) peer-rated.
STUDY II
Time 2a
STUDY III
Klunserna Study
135
05-10-25, 10.13
STUDY III
STUDY III
STUDY III
STUDY III
STUDY III
STUDY III
STUDY III
STUDY III
STUDY III
STUDY III
STUDY III
STUDY IV
Klunserna Study
161
05-10-25, 10.15
Early Temperamental Unmanageability, Harsh Parenting
Profiles, and Adolescent Problem Behavior: A Mixture
Modeling Approach With Latent Parenting Classes
STUDY IV
Vilmante Pakalniskiene, Margaret Kerr, & Håkan Stattin
Örebro University
Abstract
Temperamentally unmanageable children are at risk for developing conduct problems, and
several studies have shown moderating effects of harsh parenting. Drawing on ideas from
research on physical punishment, we examined the possible roles that different
combinations of harsh parenting might play in the link between early unmanageable
temperament and later problem behaviors. We used prospective data from 3 months to 18
years in a sample of 212 children. Latent class analysis revealed different patterns of harsh
parenting, including physical punishment alone and in the context of a discordant motherchild relationship (harsh treatment). In mixture models, unmanageable temperament
increased children’s risk of experiencing physical punishment and, to a lesser extent, harsh
treatment. Children who experienced harsh treatment showed the highest levels of problem
behaviors in adolescence, and the experience of harsh parenting seemed to play a bigger
role than temperament in the development of problem behavior. Children who experienced
physical punishment were lower on problem behaviors, and temperament seemed primarily
responsible. These differences support the idea that physical punishment is most harmful in
development when combined with behaviors that signal rejection.
Keywords: harsh parenting, early unmanageability, conduct problems, norm violations
Early Temperamental Unmanageability, Harsh Parenting
Profiles, and Adolescent Problem Behavior: A Mixture
Modeling Approach
STUDY IV
Children’s conduct problems such as aggression and delinquency are quite stable
over time and difficult to change (e.g., Farrington, 1991; Kazdin, 1995; McCord, 1983),
and for this reason, researchers have invested much effort into understanding how they
develop. There is evidence that the process begins early with a child who is
temperamentally prone to anger, aggression, and/or opposition (Bates, 1989; Bates, Bayles,
Bennett, Ridge, & Brown, 1991; Bates, Pettit, Dodge, & Ridge, 1998; Caspi, 2000; Caspi,
Henry, McGee, Moffitt, & Silva, 1995; Caspi, Moffitt, Newman, & Silva, 1996; Earls &
Jung, 1987; Eisenberg, Fabes, Shepard, Murphy, Guthrie, Jones, et al., 1997; Garrison,
Earls, & Kindlon, 1984; Guerin, Gottfried, & Thomas, 1997; Henry, Caspi, Moffitt, &
Silva, 1996; Leve, Kim, Pears, 2005; Morris, Silk, Steinberg, Sessa, Avenevoli, & Essex,
2002; Stoolmiller, 2001). In all of these studies, early anger-prone, oppositional behavior,
which was assumed to be temperament based, was linked to conduct problems,
delinquency, or even criminal behavior later in life, thus suggesting that temperament lays
the developmental foundation for later problems.
Even though children with anger-prone, oppositional temperament, or what could be
called temperamental unmanageability, might be predisposed to develop conduct problems,
there is evidence that the ways in which their parents try to manage them can make the
development of conduct problems more or less likely. In particular, harsh parenting
behavior such as physical punishment or maternal rejection might strengthen
temperamental predispositions to develop aggressive conduct problems (e.g., Bates, Pettit,
Dodge, & Ridge, 1998; Leve, Kim, & Pears, 2005; Stoolmiller, 2001). Thus, for children
with unmanageable temperament, the course of development might be much more negative
for those who experience harsh discipline than for those who do not. Empirical tests of this
moderation idea using a standard interaction approach have produced mixed results. One
early study failed to find an interaction between temperamental unmanageability before age
3 and harsh parenting at age 3 in predicting criminal convictions by age 18 (Henry, Caspi,
Moffitt, & Silva, 1996). A recent study, however, reported a significant interaction between
temperamental unmanageability and harsh discipline, both at age 5, in predicting changes in
externalizing behavior over ages 5 to 17 (Leve, et al., 2005). It is difficult to predict and
find interaction effects (see, e.g., McClelland & Judd, 1993; Stoolmiller, 2001 for
discussions of this problem), so some researchers have used group- or person-centered
approaches (Bates, Pettit, Dodge, & Ridge, 1998, Stoolmiller, 2001). In one study,
temperamental unmanageability before age 2 or 5 predicted externalizing problems at ages
7-11 more accurately when the mother had been observed to be relatively low on restrictive
control early on (Bates et al., 1998). In the other study, mothers’ unskilled discipline at age
10 was a risk factor for growth in antisocial behavior from age 10 to age 14 only for boys
with high levels of age-10 temper tantrums (Stoolmiller, 2001). Thus, there is evidence that
harsh parenting moderates the link between temperamental unmanageability and later
conduct problems.
One limitation of this research, however, is that the measures of temperament were
either not very early (e.g., Stoolmiller, 2001; Leve et al., 2005) or were solely or partly
retrospective judgments of early temperament (Bates et at., 1998; Henry et al., 1996). One
would like to see this effect tested with prospective measures of early temperament.
Another potential limitation of this research is the use of single aspects of parenting.
This could be a limitation because there is evidence that combinations of harsh parenting
behaviors might be more important than single aspects. The evidence comes from research
on physical punishment, or spanking. In reports from one longitudinal sample, spanking
seemed to increase externalizing problems among European American but not African
American children (Deater-Deckard, Dodge, Bates, & Pettit, 1996; Lansford, DeaterDeckard, Dodge, Bates, & Pettit, 2004). The authors speculated that physical punishment
might be more expected for African American children than European American children,
and that African American parents might tend to use physical punishment in the context of
behaviors that communicate love, whereas European American parents might tend to use it
in the context of behaviors that communicate rejection. Thus, they proposed that the
negative behavioral effects of physical punishment might depend on: (a) how common it is
in the culture, in general, and (b) the emotional message communicated by parents’ other
behaviors.
There is supporting evidence from a couple of studies for the idea that the effects of
physical punishment depend on the message parents’ other behaviors communicate. In one
study, physical punishment was linked to maladjustment when it was perceived as rejection
(Rohner, Bourque, & Elordi, 1996). This study was cross-sectional, however, and a number
of measures of internal and external adjustment were combined into one omnibus
maladjustment measure. In a longitudinal study, however, spanking was associated with an
increase in behavior problems over time when maternal emotional support was low, but not
when it was high, and this pattern held for European American, African American, and
Hispanic children (McLoyd & Smith, 2002). This lends credence to Deater-Deckard and
STUDY IV
colleagues’ (1996; Lansford et al., 2004) speculation that the emotional context in which
mothers used spanking was responsible for the difference in outcomes between African
American and European American children in their sample. Furthermore, though early
temperament was not included in these studies, the results suggest that the developmental
outcomes of children with unmanageable temperament might depend not only on whether
they experience harsh parenting, as shown previously, but on the combination of harsh
parenting they experience. This is important, because there is also evidence in the literature
to suggest that in addition to being at risk for conduct problems, temperamentally
unmanageable children may be at risk for experiencing harsh parenting (Bates et al., 1998;
Stoolmiller, 2001). The question is whether they might be at risk for the same combination
of harsh parenting that, in turn, increases their risk of conduct problems. If so, then there
might be a subgroup for which targeted parenting interventions, if effective, would be
maximally beneficial.
In this study, we examine different latent combinations of harsh parenting behaviors
and model their relations to early temperamental unmanageability and adolescent problem
behaviors. We use data from a birth-to-adulthood longitudinal study that contains
prospective assessments of early temperament and later measures of harsh parenting and
conduct problems. The sample is culturally homogeneous, and physical punishment was
very common when the study began in the 1950s. Thus, the first feature of Deater-Deckard
and colleagues’ explanation—physical punishment is common and expected in the
culture—is constant across the sample. Using latent profile analysis, we identify different
patterns of harsh parenting that children have experienced. The harsh parenting measures
include striking and beating (measures of physical punishment) and discordant
relationships, defined as maternal insensitivity and mother-child conflict (which could
communicate rejection to the child). We use mixture modeling, which combines variableand person-centered approaches, to examine how distinct patterns of physical discipline and
discordant relationships relate to early unmanageable temperament and later conduct
problems and norm violations and how, apart from these links, early unmanageability
relates to later conduct problems and norm violations. Thus, we examine the hypothesis
that temperamentally unmanageable children who experience physical discipline in the
context of discordant relationships will have more behavior problems than those who
experience physical punishment in the context of good relationships. Finally, we use the
grown children’s recollections of maternal rejection to verify the assumption that physical
punishment in the context of poor parent-child relationships is experienced as rejection.
Method
Participants
Participants were born in a suburb of Stockholm in the mid 1950s, and they were
studied from birth into adulthood by researchers at the Clinic for the Study of Children's
Development and Health, Karolinska Hospital, Stockholm and the University of
Stockholm. The researchers intended to obtain a comprehensive picture of many aspects of
individual growth and development. Their aims included charting the course of physical
and psychological development and obtaining reasonably comprehensive life histories that
would be useful in many areas of research.
Every fourth pregnant woman who registered at the Solna prenatal clinic (in a
suburb of Stockholm) from April 1955 to April 1958 was invited to participate in the study.
Only 3% refused. In all, 212 children (122 boys and 90 girls) took part in the study. In
Sweden, virtually all pregnant women receive regular care at prenatal clinics such as the
one from which these participants were recruited; therefore, these children are
representative of the populations in Swedish urban communities on variables such as
parents' socioeconomic status, mother's marital status, parents' ages, sibling order,
gestational age, and birth weight (Karlberg et al., 1968). The sample also has a rate of
registered criminality similar to rates from comparable Swedish longitudinal samples
(Stattin & Klackenberg-Larsson, 1990). Children and their parents were examined four
times equally spaced (every three months) during their first year, twice (every six months)
during the second year, and annually (close to their birthdays) thereafter up to the age of 18.
Data were also collected at age of 25 years. One hundred and eighty one (85.4%) of the
children (104 boys and 77 girls) from the first wave participated at age 25. Nonparticipation
was due to change of residence, death, or lack of interest. We used logistic regression to
examine whether children’s gender, socioeconomic status, or any of the measures included
in this study predicted dropout by age 25. None of these measures predicted dropout among
participants. Thus, the age-25 nonparticipation did not seem to be biased.
At the time of the child’s birth, 192 of 212 mothers were married to the child’s
biological father (in a first marriage for both partners), 14 were cohabiting, and 5 were
single. Most of the mothers (166) did not work outside the home; 41 were employed
outside the home; the others were self-employed at home or away from home. The average
age of mothers was 27.4 (SD = 5.3); the average age of fathers was 30.2 (SD = 6.0). Parents
had been living together for an average of 3.6 years (SD = 3.5). In 180 families, both
parents were Swedish by birth; in 27 families, one was Swedish; and in 4 families, neither
was Swedish. According to the Graffar classification (Graffar, 1956; 1960), where
socioeconomic status (SES) is classified from 1 (highest) to 5 (lowest), about 4% were
classified as 1 (high SES); 20% were classified as 2 (medium-high SES); 31% were
classified as 3 (medium SES); 40% were classified as 4 (medium-low SES); and 4% were
classified as 5 (low SES).
Measures
Children’s Characteristics
Early unmanageability. For the measure of early unmanageability, we used
Adolescent conduct problems. For the measures of conduct problems, we used
behaviors that are included as diagnostic features of Conduct Disorder, Oppositional
Defiant Disorder, or both in the Diagnostic and Statistical Manual of Mental Disorders (4th
ed., [DSM-IV] American Psychiatric Association, 1994). When the child was 15 and 16
years, mother-rated behavior problems were measured by asking the following questions:
“Is he/she defiant when being rebuked?” “Can you trust him/her not do things which he/she
should not do?” (reversed), “Does he/she break things willfully?” “Does he/she argue to
have his/her own way?” “If you (Mother) chastise him/her, does he/she seem to mind?”
(reversed), “If you (Mother) chastise him/her, will he/she do the same thing again the same
day?” “If father chastises him/her, does he/she seem to mind?” (reversed), “If Father
chastises him/her, will he/she do the same thing again the same day?” “Does he/she stay out
without your approval?” “Is he/she disobedient on purpose?” “Does he/she tell fibs to get
out of trouble?” “Does he/she take things he/she knows he/she should not have?” “Does
he/she get really furious?” and “Does he/she get irritated over trifles?” There were five
response options ranging from “never” to “always” to all the questions. Higher scores
indicate more problem behavior. The alpha reliability for this scale was .89. Additional
reliability and validity of the measure has been reported previously for this sample (see
STUDY IV
temper tantrums and resistance to control (see Stattin, Janson, Klackenberg-Larsson, &
Magnusson, 1995, for a validation of this measure). We combined mothers’ ratings from 3
months to 3 years. When the child was 3 – 12 months, the questions were about angerprone temperament: “Does he/she often get angry?” “Does he/she often get extremely
angry?” The alpha reliability for this scale was .81. When the child was 18 months – 3
years several age-appropriate unmanageability items were added: “Does he/she want to get
his/her own way?” “Is he/she often disobedient with you?” “Is he/she a noisy child?” “Is
he/she a destructive child?” The alpha reliability for this scale was .69. The correlation
between the two scales was .26 (p < .001).
Stattin, Janson, Klackenberg-Larsson, & Magnusson, 1995, for a validation of this
measure).
Adolescent norm violations. In addition to mother-reported conduct problems, in
the mixture modeling we used youths’ and interviewer’s reports of behaviors that are listed
as associated features of Conduct Disorder, Oppositional Defiant Disorder, or both in the
DSM-IV (American Psychiatric Association, 1994). They are truancy, alcohol drinking,
drug use, and early initiation of sexual intercourse. For the analysis we created a mean
score of these four standardized items. The alpha reliability for these four items was .71.
This adolescent- and interviewer-rated norm violations measure was significantly
correlated with mother-rated conduct problems at 15-16 years, r = .32, p < .001. The
truancy measure was a judgment made by those who interviewed the adolescents and their
parents when the adolescents were 16 years old. Based on information from both the youths
and their parents, the interviewers rated truancy on a four-point scale ranging from (1) “no
sign of being tired of school or truancy” to (4) “a lot of truancy from school.” Alcohol
drinking at age 18 was measured by total alcohol consumption per month (beer, wine and
spirits). The questions were “how much wine do you drink per month?”, “how much spirits
do you drink per month?”, “how much beer do you drink per month?” The response scale
for wine consumption was (1) none to (9) 10 glasses per month, the response scale for
spirits consumption was (1) none to (9) 18 glasses (15cl) per month, and the response scale
for beer consumption was (1) none to (9) more than 80 45cl cans per month. At the age of
17, youths reported on their prior drug use (hash, amphetamine, LSD, opium, and other
drugs). A composite measure of drug use was formed on a 7-point scale, based on the
frequency of use of any of these drugs.” At age 25, participants answered the question at
what age they had their first intercourse. There were fifteen response options ranging from
“at the age of 11” to “at the age of 25.”
Harsh Parenting
For the measures of harsh parenting we used mothers’ reports of their striking and
beating (hitting with an object), and interviewers’ ratings of discordant relationships
between mothers and children. In Table 1 are the numbers of mothers who reported using
any physical discipline or were judged as having any degree of discord in the mother-child
relationship at different age periods.
Striking. At each assessment, mothers were asked whether (and how often) they
struck their children. When children were 6 to 9 years, there were six response options,
ranging from “never” to “on many occasions every day.” Because of a lower incidence of
striking older children, the response options were changed from age 10 to age 12; there
were five response options, ranging from “never” to “daily” (see Stattin, Janson,
Klackenberg-Larsson, & Magnusson, 1995, for a validation of this measure). The periodto-period correlations for these measures ranged from .45 to 56.
Beating. When the child was 6 to 9 years, mothers were asked whether they had
given the child a real beating. For children ages 10 to 12 years, mothers reported about
beating frequency. There were five response options, ranging from “never” to “once a day.”
We dichotomized the measure at each age to make them comparable over ages. The
resulting dichotomous measures distinguish beating from no beating (see Stattin, Janson,
Klackenberg-Larsson, & Magnusson, 1995, for a validation of this measure). The periodto-period correlations for beating ranged from .31 – 39.
Discordant mother-child relationships. After each interview occasion from the
Maternal Rejection
For the measure of mother’s rejecting behavior we used the (grown up) children’s
retrospective reports given at age 25. They were instructed to think about how they
perceived their mothers when they were 12 years of age and younger and to evaluate the
following statements about their mothers: “demanded more of me than one should from the
child,” “was really interested in what I did and how I felt” (reversed), “had very few rules
for me,” “made me feel wanted and needed” (reversed), “nagged at and quarreled with me
when I behaved badly,” “made me feel that what I did was important to her” (reversed),
“did not spend more time with me when it was necessary,” “encouraged me to take my own
initiatives” (reversed), “did not want to me to bring my friends home,” “spoke to me in a
warm and affectionate manner” (reversed). There were four response options, ranging from
“does not apply at all” to “applies perfectly.” Higher scores indicate more rejecting
behavior. The alpha reliability for this scale was .82.
STUDY IV
children’s ages 6 to 12 years, the interviewers made judgments of the quality of motherchild relationships. The interviewers used a 3-point Likert scale with the scale points
described as follows: (1) good relationships—no conflict or maternal insensitivity; (2) some
evidence of conflict, disagreement, or insensitivity, or inconsistent relationship quality—
sometimes good and other times bad; and (3) pronounced conflict or apparent insensitivity
(see Stattin & Klackenberg, 1992, for a validation of this measure) The period-to-period
correlations for discordant relationships ranged from .44 to 66.
Data Analysis
For all analyses, we used Mplus 4.0 (Muthén & Muthén, 2006). Missing or
incomplete data and participant attrition are problems in all longitudinal research (Hanson,
Tobler, & Graham, 1990). Missingness, which was due to not having answers to some
questions, was addressed by using the full information maximum likelihood technique,
which is thought to provide less biased estimates than listwise or pairwise deletion (Shafer
& Graham, 2002). The proportion of missing values may be calculated with a covariance
“coverage” matrix (Muthén & Muthén, 2006), which provides an estimate of available
observations for each pair of variables. The minimum coverage necessary is .10. In this
study, coverage ranged from .76 to .98, which means that the covariance of each set of two
variables in the matrix covered 76% to 98% of cases.
Mixture Modeling
We performed mixture modeling to identify whether children’s early temperament
would relate to patterns of harsh parenting later on, whether the patterns of harsh parenting
would relate to levels of adolescent behavior problems, and whether, independent of these
links, early temperament would be related to later behavior problems. First, we tested the
latent class model to identify the patterns of harsh parenting that children experienced.
Latent class modeling refers to modeling with categorical latent variables that represent
subgroups where group membership is not known but is inferred from the data. Latent
profile analysis, which is a type of latent class analysis, is a specific statistical modeling
method developed to identify distinct subgroups according to selected characteristics and
predict class membership from continuous measures. This method offers an alternative to
cluster analytic techniques. We will use the term latent class analyses when talking about
distinct groups predicted from continuous measures such as striking, beating, and
discordant relations.
Second, we tested mixture models, as shown in Figure 1, where the child’s early
temperament served as a covariate and adolescent problem behaviors (conduct problems in
one model and norm violations in the other) served as distal outcomes. The mixture model
estimated latent class analysis together with covariance and a distal outcome. This model
helped to test whether children who experienced certain patterns of harsh parenting had
higher levels of unmanageable temperament early on (A in Figure 1) and more conduct
problems later in life (B in Figure 1). The link between class membership and distal
outcomes represents the magnitude of the outcomes for each of the harsh parenting classes.
To test whether the means on the distal outcomes varied significantly across classes, we ran
the models with and without holding the means equal across classes to get chi-square (two
times the log likelihood) difference tests. The pathways between early unmanageability and
the distal outcomes show to what extent early unmanageability is related to the occurrence
of adolescent problem behavior, once the model controls for the risk associated with class
membership (C in Figure 1). All these links in the mixture models were estimated
simultaneously.
Results
Harsh Parenting Behaviors and Their Relations to Early
Unmanageability and Adolescent Problem Behavior
STUDY IV
To test the model shown in Figure 1, we first examined the latent classes of
mother’s striking, beating, and discordant relationships at ages 6-12. Models with different
numbers of classes were compared with the Bayesian information criterion (BIC), the
sample-size adjusted Bayesian information criterion (SSABIC), the Akaike information
criterion (AIC), and the Entropy criterion. Lower scores of BIC, SSABIC, and AIC
represent better fitting models (Muthén & Muthén, 2000; Schwartz, 1978). Entropy refers
to the average classification accuracy in the assignment of participants to classes. Entropy
values range from zero to 1, with values closer to 1 indicating better classifications of
individuals to specific classes (Bauer & Currant, 2003). For this model, the five-class
solution had the best fit indices, but one class consisted of only 6 people (see Table 2).
Inspection of the four- and five-class solutions revealed that the class with 6 people was
very similar to the largest class, which had the lowest levels of all three harsh parenting
variables. This 6-person class had just slightly higher levels of mother’s striking and
beating. We chose the four-class solution because: (a) very small classes can create
problems with analyses and interpretation of results (Nylund, Asparouhov, & Muthén,
2006); (b) theoretically, we were most interested in comparing those who experienced
physical punishment alone with those who experienced it in the context of discordant
relationships, and those two groups existed in the four-class solution; and (c) even though
entropy was somewhat lower than with the three-class solution, other indicators suggested
that the four-class solution was reasonable. The final model estimated four classes of
mother’s harsh treatment (see Figure 2).
The largest class accounted for 49.7% of the sample (n = 90). As shown in Figure 2,
this normative class had low levels of all three harsh parenting variables: striking, beating,
and discordant relationships (M = -.305, SE = .084 for mother’s striking; M = -.635, SE =
.042 for beating; and M = -.384, SE = .042 for discordant relationships). The second largest
class, a physical punishment class, accounted for 30.9% of the sample (n = 56) and had
high levels of mother’s striking (M = .355, SE = .097) and beating (M = .859, SE = .073)
and low levels of relationship discord (M = -.271, SE = .051). The third class, a discordant
relationships class, accounted for 11.6% of the sample (n = 21) and had high levels of
discordant mother-child relationships (M = 1.363, SE = .283), and low levels of striking (M
= -.231, SE = .277) and beating (M = -.415, SE = .155). The final class, a harsh treatment
class, accounted for 7.8% of the sample (n = 14) and had high levels of striking (M = .821,
SE = .228), beating (M = 1.023, SE = .088), and discordant relationships (M = 1.587, SE =
.150). Thus, even though striking, beating, and discordant relationships are related to each
other, children experienced different combinations over ages 6-12. Almost one third of
children in the sample experienced striking and beating with no notable signs of discordant
relationships with their mothers. For a few, however, physical punishment occurred in the
context of a discordant mother-child relationship.
In the process of building the mixture models, the covariate (unmanageable
temperament) and the distal outcomes (mother-reported conduct problems in one model
and self- and interviewer-reported norm violations in the other) were added to the latent
class model. The latent classes were regressed on early unmanageability, and then problem
behavior (either conduct problems or adolescent norm violations) were regressed on early
unmanageability and the latent classes (Log likelihood = -762.822; BIC = 1684.891, SSA
BIC= 1589.845, AIC = 1585.643, entropy = .835, number or parameters = 30 for model
with conduct problems as a distal outcome; Log likelihood = -711.946; BIC = 1583.146,
SSA BIC= 1488.100, AIC = 1483.898, entropy = .828, number or parameters = 30 for
model with norm violations as a distal outcome).
Does early unmanageability increase the risk of experiencing
specific combinations of harsh parenting?
In the part of the model that examined relations between children’s early
unmanageability (age 3 months–3 years) and combinations of harsh parenting behavior
(age 6–12 years), latent class membership was regressed on early unmanageability (A in
Figure 1). In this comparison, one class serves as the baseline, or reference, category. We
chose the normative class, which had low levels of mother’s striking, beating, and
discordant relationships, as the reference category for the other classes. Results are
expressed in odds ratios as in logistic regression and presented in the columns labeled A in
Table 3. The results were similar in the two models—one with conduct problems as the
distal outcome and the other with norm violations as the distal outcome. Children with high
levels of early unmanageability were more likely to be in the physical punishment class
than in the normative class (OR = 2.78, p < .05, 95% CI – 1.49-5.21, OR = 3.05, p < .05,
95% CI – 1.32-6.96, for models with conduct problems and norm violations, respectively,
as distal outcomes). Children with high levels of unmanageability also showed a tendency
to be in the harsh treatment class rather than the normative class (OR = 1.78, p < .10, 90%
CI – 1.02-3.10, OR = 1.36, p < .10, 90% CI – 1.01-1.92, for models with conduct problems
and norm violations, respectively, as distal outcomes), but they were not at increased risk of
being in the discordant relationships class (OR = .85, 95% CI – .47-1.42, OR = .88, 95% CI
– .50-1.57, for models with conduct problems and norm violations, respectively). Thus, for
models with both problem-behavior outcomes, unmanageable temperament increased
children’s risk of experiencing physical punishment and to a lesser extent physical
punishment in the context of discordant relationships later in childhood.
In these models, the links between class membership and distal outcomes represent
the magnitude of the outcomes for each of the harsh parenting classes (B in Figure1). To
test whether the means of the distal outcomes varied significantly across classes, we ran the
models with and without holding the means equal across classes to get chi-square (two
times the log likelihood) difference tests.
According to chi-square difference testing for the model with conduct problems as
the outcome variable (second column in Table 3), children in the normative and discordant
relationships classes had significantly lower levels of conduct problems in adolescence than
children in the other classes (M = -.214, SD = .107, M = -.158, SD = .324, for the normative
and discordant relationships classes, respectively) and children who experienced harsh
treatment were significantly more likely than children in all of the other classes to have
conduct problems in adolescence (M = 1.265, SD = .278). Children who experienced
physical punishment alone had significantly lower levels of conduct problems than children
in the harsh treatment class, but significantly higher levels of conduct problems than those
in the normative and discordant relationships classes (M = .065, SD = .123). Turning to the
model with norm violations as the distal outcome (fifth column in Table 3), children in the
normative class had significantly lower levels of norm violations than children in the other
classes (M = -.105, SD = .075) and children who experienced harsh treatment had
significantly higher levels than children in the other classes (M = 1.002, SD = .375).
Children who experienced physical punishment alone were similar to those who
experienced discordant relationships (M = .011, SD = .136, M = .076, SD = .192, for the
physical punishment and discordant relationships classes, respectively). Thus, the pattern of
STUDY IV
Are different combinations of harsh parenting related to levels of
later problem behaviors?
results is largely similar for both problem-behavior outcomes. The only difference between
the two models was for those who experienced only discordant relationships. They were
similar to the normative group on conduct problems but similar to the physical punishment
group on norm violations. Otherwise, in both models children who experienced physical
punishment in the context of discordant mother-child relationships had more problem
behaviors later on than those who experienced physical punishment alone, and those who
experienced physical punishment alone had more problem behaviors than those who did not
experience either type of harsh parenting.
Links between early unmanageable temperament and adolescent
problem behaviors
In the mixture models we also examined to what extent children’s early
unmanageability was related to later conduct problems or norm violations apart from the
risk associated with their harsh parenting class membership (C in Figure 1). For the
expression of this path in the model we use standardized estimates provided by Mplus
(equivalent to standardized estimates provided by Amos or Lisrel). It is possible to refer to
these estimates as betas (regular standardized coefficients or linear regression coefficients).
For the significance levels we used Z statistics, so values that exceed +1.96 or fall below 1.96 are significant at p < .05 and values that exceed +1.65 or fall below -1.65 are
significant at p < .10).
Results from the mixture model with conduct problems as the outcome (third
column in Table 3) showed only one class for which early temperament was linked to later
conduct problems after controlling for the risk associated with the class membership.
Among children in the physical punishment class, higher early unmanageability was
significantly related to risk of conduct problems (Est. = .229, z = 1.978). Notably, among
children in the harsh treatment class, where conduct problems were highest, there was no
significant link between early temperament and conduct problems apart from the risk
associated with harsh parental treatment (Est. = -.103, z = -.403). For the normative and
discordant relationships classes, early unmanageability was not significantly related to
conduct problems (Est. = .126, z = 1.309 and Est. = -.070, z = -.241 for the normative and
discordant relationships classes, respectively). Results from the mixture model with norm
violations as the outcome (sixth column in Table 3) showed similar results. The physical
punishment class was the only class for which early temperament was significantly linked
to later norm violations after controlling for the risk associated with the class membership
(Est. = .254, z = 2.007). For the normative class, however, there was a tendency (Est. =
.146, z = 1.627). But again, for the class with the highest levels of norm violations—the
Harsh Parenting and Children’s Perceptions of Maternal Rejection
To infer children’s experiences of these harsh parenting behaviors as rejecting, we
tested an additional mixture model that was identical to those reported above, except that
the distal outcome was children’s age-25 retrospective reports of their mothers’ rejecting
behavior during their upbringing (Log likelihood = -731.176; BIC = 1621.601, SSA BIC=
1526.555, AIC = 1522.353, entropy = .837, number or parameters = 30). Significance
testing revealed that children who experienced mother’s harsh treatment or discordant
relationships remembered their mothers as more rejecting during childhood than children in
either of the other classes (M = .699, SD = .298, M = .681, SD = .316, for the harsh
treatment and discordant relationships classes, respectively). Children who experienced
only physical punishment, however, remembered their mothers as significantly less
rejecting than children in the harsh treatment and discordant relationships classes, but
significantly more rejecting than those in the normative class (M = .062, SD = .152). Not
surprisingly, children in the normative class recalled the least maternal rejection of all the
classes (M = -.325, SD = .112). In sum, children who experienced mother’s physical
punishment in the context of discordant relationships tended to remember their mothers as
more rejecting than children who experienced physical punishment in the context of
generally good relationships.
STUDY IV
harsh treatment class—early unmanageability was not related to risk of norm violations
over and above the risk associated with harsh treatment (Est. = .073, z = .146). Finally,
among children who experienced discordant relationships, higher early unmanageability
was not related to risk of norm violations (Est. = .093, z = .975). Comparing the two groups
of the most theoretical interest, then, harsh treatment, or physical punishment in the context
of discordant mother-child relationships, was related to higher levels of later conduct
problems and norm violations than physical punishment in the context of good
relationships. What is more, for the harsh treatment group unlike the physical punishment
group, the experience of harsh parenting seemed to play a bigger role than temperamental
unmanageability in the development of problem behavior. These group differences are
consistent with the idea that physical punishment has a different meaning in the context of
good parent-child relationships than in the context of discordant relationships. If it could be
assumed that the children with discordant relationships had the experience of rejection, then
this would support Deater-Deckard and colleagues’ (1996; Lansford et al., 2004)
suggestion that physical punishment is related to bad outcomes if a child experiences it in
the context of behaviors that communicate rejection.
Discussion
Previous studies have shown that among children who are temperamentally difficult
to manage, the course of development might be very different for those who experience
harsh parenting than for those who do not (Bates et al., 1998; Leve et al., 2005; Stoolmiller,
2001), but previous studies also suggest that combinations of harsh parenting behaviors—
physical punishment in the context of behaviors that communicate rejection—might be
more important than individual harsh parenting behaviors—physical punishment alone
(Deater-Deckard, Dodge, Bates, & Pettit, 1996; Lansford, Deater-Deckard, Dodge, Bates,
& Pettit, 2004; McLoyd & Smith, 2002; Rohner, Bourque, & Elordi, 1996). Using a
mixture modeling approach with latent classes representing different harsh parenting
profiles, we modeled the relations between temperamental unmanageability, harsh
parenting profiles, and later problem behaviors. Taken together, the results show that
different harsh parenting profiles play different roles in the link between early
temperamental unmanageability and later problem behavior. Unmanageable temperament
puts children at risk for physical punishment. For children who experience physical
punishment, both it and their own unmanageability seem to play separate roles in their later
problem behaviors, which are elevated, but not the highest in the sample. Early
temperament also, but to a lesser extent, increases the risk of experiencing physical
punishment together with discordant mother-child relationships. Although children in this
harsh parenting profile have the highest levels of problem behaviors, their behavior
problems are not linked to early temperament apart from the experience of harsh parenting.
Because of the tenuous link between temperament and this harsh parenting combination, it
seems reasonable to conclude that for this group harsh parenting plays a bigger role than
temperamental unmanageability does in the development of later problems. Thus, our
results add new information to the earlier findings that harsh parenting moderates the link
between early temperament and problem behavior outcomes (Bates et al., 1998; Leve et al.,
2005; Stoolmiller, 2001), and they lend credibility to the suggestion (Deater-Deckard et al.,
1996; Lansford et al., 2004) that physical punishment can have differential effects
depending on whether the child perceives the parent as rejecting. Our findings are not
entirely consistent, however, with previous findings that when physical punishment is used
frequently, it is less likely to be linked to children’s adjustment problems (Lansford et al.,
2005). In this sample, physical punishment was very frequently used, and yet it was linked
to higher levels of conduct problems and norm violations, even when it was not
experienced together with discordant relationships.
STUDY IV
The latent class part of this analysis could be referred to as a person-centered
approach, which implies that considering the whole person, or at least a pattern of the
person’s characteristics or behaviors, will yield a different understanding of the person’s
functioning than one would get by studying relations among variables in a sample of the
population (Bergman, Magnusson, & El-Khouri, 2003; Bergman & Trost, 2006;
Magnusson & Stattin, 2006). In this case, the combination of the person- and variablecentered approaches in the mixture modeling helped to reveal fundamental differences
among the families in the different harsh parenting profiles in the role that temperamental
unmanageability plays in harsh parenting and later problem behavior. The striking and
beating that one group of mothers did might have been a qualitatively different activity than
the striking and beating that the other group did, because relationship discord might have
been an indicator of maternal psychopathology or dire problems in the home such as
substance abuse or domestic violence. Children who experienced discordant relationships
later remembered their mothers as rejecting whether they used a lot of physical punishment
or not, and that suggests that there were profound differences between mothers who used
physical punishment alone and those who used it in the context of discordant relationships.
There are undoubtedly other ways in which the understanding of these processes
could be improved. For instance, one thing that we have not considered in this study is that
children with different temperaments might respond differently to the same parenting
behaviors. This idea has support from several studies (e.g., Belsky, 1997; Wachs &
Gandour, 1983). Although this was not our question for this study, it will ultimately be an
important additional issue to consider in this developmental process and to add to future
models.
One potential limitation of this study is that it began in the mid 1950s. The question
is what the cultural views of physical punishment were at the time, because that would
determine how children viewed striking and beating. In part, an answer can be inferred
from the sheer numbers of mothers who reported striking and beating their children. For
striking, when children were 6 and 7 years old, it was virtually all of the mothers, to one
degree or another. When the children were 9 years old, a third of mothers reported having
given the child “a real beating.” Thus, we think it is safe to conclude that the children in
this study did not view physical punishment as aberrant, and we assume that this explains
why children who received physical punishment in the context of harmonious mother-child
relationships did not remember their mothers as rejecting. We also assume that if this study
could be repeated in Sweden today, the findings concerning physical punishment might be
very different. Sweden was the first country in the world to outlaw physical punishment of
children. The law went into effect in 1979, so a whole cohort of Swedes has now grown to
adulthood amidst public awareness campaigns concerning this. Although some parents
undoubtedly still use physical punishment, they are probably more out of control when they
do it than the parents in our sample were when they struck or beat their children. Moreover,
because Swedish children know that striking and beating is illegal, if they experience it
they are likely to perceive their parents’ intentions and feeling toward them much
differently than the children in our sample did.
The present study has a number of strengths. Perhaps the main strength is the
prospective, long-term, longitudinal design with data collected at short intervals, beginning
before the children’s first birthdays and following the children into young adulthood. These
data allowed us to examine connections between harsh parenting, very early unmanageable
temperament, and later problem behavior in a more elaborate way than ever before reported
in the literature. Another strength was the reliance on a community sample which did not
differ much from other Swedish children of the same cohort generally (Karlberg et al.,
1968). A further strength of the study was the use of data from different reporters,
particularly for the problem-behavior outcomes. The links between mother–rated conduct
problems and temperament and physical punishment could be the result of rater bias, but in
this case the results were almost identical using youths’ and interviewers’ judgments of
norm violations. These findings give us confidence that the results of these models are not
just due to rater bias, but are tapping into actual developmental processes.
What implications do our findings have for the debate that has been ongoing in the
literature concerning physical punishment (see, e.g., Baumrind, 1997; Deater-Deckard &
Dodge, 1997; Gershoff, 2002; Holden, 2002; Larzelere, 2000) and its implications for
policy makers? On one hand, the evidence seems to be mounting that under the right
cultural and family conditions physical punishment does not necessarily produce more
aggressive or problematic behavior. This could be taken as an argument against changing
the tradition of physical punishment in countries where it is still legal. On the other hand,
the results also show that under less than optimal family conditions, physical punishment
might contribute to the development of behavior problems that will interfere with the goals
of society and the individual’s own enjoyment of life, and this should arouse concern for
policymakers and parents themselves.
References
STUDY IV
American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental
Disorders (4th ed.) Washington, DC, USA: Author.
Bates, J. E. (1989). Application of temperament concepts. In G. A. Kohnstamm, J. E. Bates,
& M. K. Rothbart (Eds.), Temperament in childhood (pp. 321-355). Chichester,
England: Wiley.
Bates, J. E., Bayles, K., Bennett, D. S., Ridge, B., & Brown, M. M. (1991). Origins of
externalizing behavior problems at eight years of age. In D. Pepler & K. Rubin
(Eds.), Development and treatment of childhood aggression (pp. 93-120). Hillsdale,
NJ: Erlbaum.
Bates, J. E., Pettit, G. S., Dodge, K. A., & Ridge, B. (1998). Interaction of temperamental
resistance to control and restrictive parenting in the development of externalizing
behavior. Developmental Psychology, 34, 982 – 995.
Bauer, D. J., & Currant, P. J. (2003). Distributional assumptions of growth mixture models:
Implications for overextraction of latent trajectory classes. Psychological Methods,
8, 338 – 363.
Baumrind, D. (1997). Necessary distinctions. Psychological Inquiry, 8, 176 – 182.
Belsky, J. (1997). Variation insusceptibility to environmental influences: An evolutionary
argument. Psychological Inquiry, 8, 182-186.
Bergman, L. R., Magnusson, D., & El-Khouri, B. (2003). Studying individual development
in an interindividual context: A person oriented approach. Mahwah, NJ: Lawrence
Erlbaum.
Bergman, L. R., & Trost, K. (2006). The person-oriented versus the variable-oriented
approach: Are they complementary, opposites, or exploring different worlds?
Merrill-Palmer Quarterly, 52, 601-632.
Caspi, A. (2000). The child is father of the man: Personality continuities from childhood to
adulthood. Journal of Personality and Social Psychology, 78, 158 – 172.
Caspi, A., Henry, B., McGee, R., Moffitt, T., & Silva, P. (1995). Temperamental origins of
child and adolescent behavior problems: From age 3 to age 15. Child Development,
66, 55 – 68.
Caspi, A., Moffitt, T. E., Newman, D. L., & Silva, P. A. (1996). Behavioral observations at
age 3 years predict adult psychiatric disorders. Archives of General Psychiatry, 53,
1033-1039.
Deater-Deckard, K., & Dodge, K. A. (1997). Spare the rod, spoil the authors: Emerging
themes in research on parenting and child development. Psychological Inquiry, 8,
230 – 235.
Deater-Deckard, K., Dodge, K. A., Bates, J. E., & Pettit, G. S. (1996). Physical discipline
among African American and European American mothers: Links to children’s
externalizing behaviors. Developmental Psychology, 32, 1065 – 1072.
Earls, F., & Jung, K. (1987). Temperament and home environment characteristics as causal
factors in the early development of childhood psychopathology. Journal of the
American Academy of Child and Adolescent Psychiatry, 26, 491 – 498.
Eisenberg, N., Fabes, R., Shepard, S. A., Murphy, B. C., Guthrie, I. K., Jones, S., et al.
(1997). Contemporaneus and longitudinal prediction of children’s social functioning
from regulation and emotionality. Child Development, 68, 642 – 664.
Farrington, D. P. (1991). Childhood aggression and adult violence: Early precursors and
life coutcomes. In D. J. Pepler & K. H. Rubin (Eds.), Development and treatment of
childhood aggression (pp. 5 – 29). Hillsdale, NJ: Earlbaum.
Garrison, W., Earls, F., & Kindlon, D. (1984). Temperament characteristics in the third
year of life behavioral adjustment at school entry. Journal of Clinical Child
Psychology, 13, 298-303.
Gershoff, E. T. (2002). Corporal punishment, physical abuse, and the burden of proof:
Reply to Baumrind, Larzelere, and Cowan (2002), Holden (2992), and Parke
(2002). Psychological Bulletin, 4, 602 – 611.
Graffar M (1956) Une methode de classification sociale d echantillons de population.
Courrier, 332,133 – 8.
Graffar, M. (1960). Social study of the samples. In S. Karger (Ed). Modern problems in
pediatrics. Basel.
Guerin, D. W., Gottfried, A. W., & Thomas, C. W. (1997). Difficult temperament and
behavior problems: A longitudinal study from 1.5 to 12 years. International Journal
of Behavioral Development, 21, 71-90.
Hanson, W. B., Tobler, N. S., & Graham, J. W. (1990). Attrition in substance abuse
prevention research: A meta-analysis of 85 longitudinally followed cohorts.
Evaluation Review, 14, 677 – 685.
Henry, B., Caspi, A., Moffitt, T. E., & Silva, P. A. (1996). Temperamental and familial
predictors of violent and nonviolent criminal convictions: Age 3 to age 18.
Developmental Psychology, 32, 614 – 623.
STUDY IV
Holden, G. W. (2002). Perspectives of the effects of corporal punishment: Comment on
Gershoff (2002). Psychological Bulletin, 128, 590 – 595.
Karlberg, P., Klackenberg, G., Engström, I., Klackenberg-Larsson, I., Lichtenstein, H.,
Stensson, J., & Svennberg, I. (1968). The development of children in a Swedish
urban community: A prospective, longitudinal study – Parts I-VI. Acta Paediatrica
Scandinavica Supplement, Whole No. 187.
Kazdin, A. E. (1995). Conduct disorders in childhood and adolescence. Thousand Oaks,
CA, USA: Sage.
Lansford, J. E., Chang, L., Dodge, K. A., Malone, P. S., Oburu, P., Palmérus, K. et al.
(2005). Physical discipline and children’s adjustment: Cultural normativeness as a
moderator. Child Development, 76, 1234 – 1246.
Lansford, J. E., Deater-Deckard, K., Dodge, K. A., Bates, J., & Pettit, G. S. (2004). Ethnic
differnces in the link between physical discipline and later adolescent externalizing
behaviors. Journal of Child Psychology and Psychiatry, 45, 801-812.
Larzelere, R. E. (2000). Child outcomes of nonabusive and customary physical punishment
by parents: An updated literature review. Clinical Child and Family Psychology
Review, 3, 199 – 221.
Leve, L. D., Kin, H. K., & Pears, K. C. (2005). Childhood temperament and family
environment as predictors of internalizing and externalizing trajectories from age 5
to 17. Journal of Abnormal Child Psychology, 33, 505 – 520.
Nylund, K. L., Asparouhov, T., & Muthén, B. O. (2006). Deciding on the number of
classes in the latent class analysis and growth mixture modeling: A Monte Carlo
simulation study. Manuscript submitted for publication.
Magnusson, D., & Stattin, H. (2006). The person in context: A holistic-interactionistic
approach. In R. M. Lerner and W. Damon (Eds.), Handbook of child psychology (6th
edition). Volume 1, Theoretical models of human development (pp. 400 – 464).
Hoboken, NJ, USA: Wiley.
McClelland, G. H., & Judd, C. M. (1993). Statistical difficulties of detecting interactions
and moderator effects. Psychological Bulletin, 114, 376 – 390.
McCord, J. (1983). A longitudinal study of aggression and antisocial behavior: In K. T.
VanDusen & S. A. Mednick (Eds.), Prospective studies of crime and delinquency
(pp. 269 – 275). Boston: Kluwer-Nijhoff.
McLoyd, V. C., & Smith, J. (2002). Physical discipline and behavior problems in African
American, European American, and Hispanic children: Emotional support as a
moderator. Journal of Marriage and Family, 64, 40 – 53.
Morris, A. S., Silk, J. S., Steinberg, L., Sessa, F. M., Avenevoli, S., & Essex, M. J. (2002).
Temperamental vulnerability and negative parenting as interacting of child
adjustment. Journal of Marriage and the Family, 64, 461-471.
Muthén, B. O., & Muthén, L. K. (2000). Integrating person-centered and variable-centered
analyses: Growth mixture modeling with latent trajectory classes. Alcoholism:
Clinical and Experimental Research, 24, 1 – 10.
Muthén, L. K., & Muthén, B. O. (2006). Mplus users guide. Los Angeles, CA: Muthén and
Muthén.
Rohner, R. P., Bourque, S. L., & Elordi, C. A. (1996). Children’s perceptions of corporal
punishment, caretaker acceptance, and psychological adjustment in a poor, biracial
Southern community. Journal of Marriage and the Family, 58, 842 – 852.
Schwartz, G. (1978). Estimating the dimension of the model. The Annals of Statistics, 6,
461 – 464.
Shafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art.
Psychological Methods, 7, 147 – 177.
Stattin, H., Janson, H., Klackenberg-Larsson, I., & Magnusson, D. (1995). Corporal
punishment in everyday life: An intergenerational perspective. In J. McCord (Ed.),
Coercion and punishment in long-term perspectives (pp. 315-347). Cambridge:
Cambridge University Press.
Stattin, H., & Klackenberg-Larsson, I. (1990). The relationship between maternal attributes
in the early life of the child and the child’s future criminal behavior. Development
and Psychopathology, 2, 99 – 111.
Stattin, H., & Klackenberg, G. (1992). Discordant relations in intact families:
Developmental tendencies over 18 years. Journal of Marriage and the Family, 54,
940-956.
Stoolmiller, M. (2001). Synergistic interaction of child manageability problems and parentdiscipline tactics in predicting future growth in externalizing behavior for boys.
Developmental Psychology, 37, 814 – 823.
Wachs, T. D., & Gandour, M. (1983). Temperament, environment and six month cognitiveintellectual development: A test of the organismic-specificity hypothesis.
International Journal of Behavioral Development, 6, 135 – 152.
Table 1
Numbers of Mothers at Each Age Who Reported Using Any Physical Discipline or Were
Judged as Having Any Degree of Discord in the Mother-Child Relationship (N = 212).
Striking
Beating
Discordant relations
6y
7y
8y
9y
10y
11y
12y
169
56
11
163
52
13
158
31
18
153
73
25
123
49
22
102
30
23
68
20
24
STUDY IV
Table 2
Fit Indices for Latent Class Solutions
Number of
Log
parameters likelihood
Number of
latent classes
1 class
2 classes
3 classes
4 classes
5 classes
6 classes
6
10
14
18
22
26
-639.147
-577.853
-538.513
-500.154
-482.139
-524.804
BIC
SSABIC
AIC
Entropy
1310.144
1208.789
1151.341
1117.090
1102.274
1145.158
1291.135
1177.107
1106.986
1047.389
1091.921
1108.130
1290.295
1175.706
1105.025
1044.308
1016.279
1085.609
.893
.964
.889
.882
.833
Harsh treatment
Physical punishment
Discordant relationships
Normative
b
1.78 †
2.78 *
.85
Reference
Latent parenting class
*p < .05; †p < .10
1
Odds ratios; 2 Mean values; 3 Estimates
a
Significantly higher than all others;
A
Temperament
predicting
latent class1
c
c
b
a
-.10
.22 *
-.07
.13
C
Temperament
predicting
conduct
problems 3
c
1.36 †
3.05 *
.88
Reference
1.00
.01
.08
-.11
c
b
b
a
B
Mean norm
violations 2
Norm violations
.07
.25 *
.09
.15 †
C
Temperament
predicting
norm
violations 3
Significantly lower than all others
A
Temperament
predicting
latent class 1
Significantly different from highest and lowest;
1.27
.07
-.16
-.21
B
Mean conduct
problems2
Conduct problems
Table 3. Summary of Mixture Modeling Results for Models with Conduct Problems and Norm Violations, Respectively, as Distal
Outcomes. Column Headings A, B, and C Refer to Pathways Shown in Figure 1.
STUDY IV
Figure Captions
Figure 1. Conceptual latent classes of mothers’ behavior model with risk factor
associated with latent classes and prediction of distal outcomes.
Figure 2. Model-estimated means for the four-class solution.
Covariate:
- Early unmanageability
3m-3y
A
Striking
6 – 12 years
C
Class
Membership
Beating
6 – 12 years
B
Discord
relations
6 – 12 years
STUDY IV
Outcomes:
Conduct problems/
Norm violations
28
2
1.5
1
Striking
Beating
Discordant relations
0.5
0
-0.5
Normative
-1
Physical
Discordant
punishment relationships
Harsh
treatment
STUDY V
Klunserna Study
171
05-10-25, 10.16
Children’s Temperamental Unmanageability, Harsh
Parenting, and Quality of Romantic Relationships in
Adulthood from a Longitudinal Perspective
Vilmante Pakalniskiene
Örebro University
STUDY V
Abstract
Temperamentally unmanageable children or children who experience negative
parenting are at risk of having problems in romantic relationships later in life. It has been
suggested that parents’ values and attitudes toward marriage might also affect their
children’s romantic relationships. Under the assumption that the best understanding might
come from combining these factors, in this study I examined the possible roles that
different combinations of harsh parenting might play in the link between early
unmanageable temperament and parents’ attitudes on one hand and relationship quality
later in life on the other. I used prospective data from 3 months to 35 years in a sample of
212 children. Latent class analysis revealed different patterns of harsh parenting. In a
mixture model, unmanageable temperament increased children’s risk of experiencing harsh
treatment. Children who experienced a combination of physical punishment and discordant
relationships (harsh treatment) had the lowest quality romantic relationships later in life
children’s temperament and the experience of harsh parenting seemed to play similar roles
in the development of problems in relationships. Children who experienced discordant
relationships also had similarly low quality relationships, and parenting seemed primarily
responsible. Also, for these children, parents’ attitudes towards marital conflict affected
later romantic relationships. Thus, the results suggest that behaviors that create discord or
conflict in the family might be more damaging than other types of parents’ behaviors.
Keywords: harsh parenting, early unmanageability, romantic relations, marital conflict
Children’s Temperamental Unmanageability, Harsh
Parenting, and Quality of Romantic Relationships in
Adulthood from a Longitudinal Perspective
STUDY V
Eventually, everyone wants to have a partner in life. Generally, people think that
having a romantic partner will make their life much better or happier. Research also
suggests that people who have a romantic partner have better health, are happier, or have
higher self-esteem (see, e.g., Hibbard & Pope, 1991; Kiecolt-Glaser & Newton, 2001; Ross,
Mirowsky, & Goldsteen, 1990). However, some people have a romantic partner but are not
happy; in such cases, they will have difficulties in the relationship with their partner. The
question is why some people have difficulties in their romantic relationships and are not
satisfied with romantic relationship quality, whereas others do not. In the literature from
different theoretical perspectives, several possible explanations have been tested.
From the perspective of temperament and personality theory, one explanation is that
traits and behaviors existing long before the partners first meet contribute to the way
romantic partners get along over time. Even children’s temperament and behavior styles at
age 3 have been linked to experiences in relationships, such as intimacy, power balance,
mutual interests, and partner violence, at age 21 (Newman, Caspi, Moffitt, & Silva, 1997).
And a history of childhood tantrums has been linked to divorce at midlife, at age 40 (Caspi,
Elder, & Bem, 1987). Temperament is thought of as the foundation for later personality,
and it is believed to be elaborated over time into a stable behavioral disposition (e.g., Caspi,
2000; Caspi & Silva, 1995; Eisenberg, Fabes, Guthrie, & Reiser, 2000; Rutter, 1987;
Snyder & Ickes, 1985). From this perspective, it follows that researchers interested in older
children or adolescents will examine the effects of personality traits on romantic
relationships later in life. Empirical evidence suggests that negative emotionality,
neuroticism, or aggressiveness during childhood, adolescence or adulthood may have longterm effects on romantic or marital relationships (e.g., Blum & Mehrabian, 1999;
Bouchard, Lussier, & Sabourin, 1999; Donnellan, Conger, & Bryan, 2004; Huston &
Houts, 1998; Kelly & Conley, 1987; Kinnunen & Pulkkinen, 2003; Watson, Hubbard, &
Wiese, 2000). Thus, stable personality characteristics or even early temperamental
predispositions may affect future romantic relationships.
From a learning theory perspective, research has suggested that the nature of child
upbringing may have long-term consequences for romantic or marital relationships
(Andrews, Foster, Capaldi, & Hops, 2000; Capaldi & Clark, 1998; Flouri & Buchanan,
2002; Franz, McClelland, & Weineberger, 1991; Linder & Collins, 2005). In particular,
parent-child closeness in adolescence has been linked to the quality of relationships with
partners in midlife, at age 33 (Flouri & Buchanan, 2002). And poor parenting practices,
such as monitoring and discipline, or aversive communication during adolescence, have
been found to have strong associations with physical aggression towards partners in young
adulthood (Andrews, Foster, Capaldi, & Hops, 2000; Capaldi & Clark, 1998). Perhaps the
clearest evidence that parental behavior influences the child’s future relationships comes
from studies on child abuse and harsh parenting. Adults who have been abused or
experienced harsh parenting as children are not satisfied with their relationships, do not
have positive perceptions of their current romantic partners (Colman & Widom, 2004;
Finkelhor, Hotaling, Lewis, & Smith, 1989; Fleming, Mullen, Sibthorpe, & Bammer,
1999), are fearful of having a partner (Davis & Petretic-Jackson, 2000), or have trouble
maintaining intimate or romantic relationships (Colman & Widom, 2004; Felitti, 1991;
Fleming et al., 1999). Thus, it seems that negative or harsh parental behavior, especially
during late childhood and adolescence, has long-term consequences for the child’s future
relationships.
A learning perspective also suggests that parents’ own marital satisfaction, values,
and attitudes towards marriage or romantic relationships may affect their children’s
romantic relationships, as well as children’s characteristics or parent-child relationships
(e.g., Amato & Booth, 2001; Conger, Cui, Bryan, & Elder, 2000; Sanders, Halford, &
Behrens, 1999). Social learning theory offers a background for this way of thinking,
suggesting that children learn a variety of interpersonal behaviors or attitudes through their
observation of adults (Bandura, 1977). And children have frequent opportunities to observe
their parents during childhood and adolescence, thus giving them possibilities to get to
know their parents’ behaviors and attitudes in the family. More specifically, children with
parents who have troubles in their relationships observe less positive behaviors than other
children, and may use these types of behaviors later in life with their own romantic
partners. Similarly, children with parents who think that conflict is a natural or necessary
part of any relationship can have this attitude later in life, and experience various problems
in their romantic relationships. Thus, it might be that not only parents’ behavior with a
child, but also parents’ behavior to each other or attitudes, affect the child’s romantic
relationships later in life.
In reality, children’s temperament, harsh parenting, and parents’ relationships are
probably not independent of each other as factors affecting adult romantic relationships.
Children with angry, oppositional temperaments are more likely than others to experience
harsh parenting (e.g., Bates, Pettit, Dodge, & Ridge, 1998; Pakalniskiene, Kerr, & Stattin,
2007; Stoolmiller, 2001), and harsh parenting seems to escalate adolescents’ problem
STUDY V
behaviors (e.g., Caspi, 2000; Leve, Kim, & Pears, 2005; Morris, Silk, Steinberg, Sessa,
Avenevoli, & Essex, 2002; Pakalniskiene et al., 2007; Stoolmiller, 2001). This, in turn, may
be linked to adult romantic relationships (Karney & Bradbury, 1995; Kinnunen &
Pulkkinen, 2003). Thus, children’s temperament and harsh parenting are related to each
other in affecting adult romantic relationships. It is also reasonable to suppose that parents’
marital conflict spills over into their relationships with their children (Fauber, Forehand,
Thomas, & Wierson, 1990), and is experienced by the child as conflict or disharmony in the
parent-child relationship. This disharmony in relationships, if the child interprets it as
parental rejection, can determine whether harsh parenting practices, such as physical
punishment, result in increased aggression and other problem behaviors (Deater-Deckard,
Dodge, Bates, & Pettit, 1996; Lansford, Deater-Deckard, Dodge, Bates, & Pettit, 2004;
McLoyd & Smith, 2002; Rohner, Bourque, & Elordi, 1996; Pakalniskiene et al., 2007). It
seems that disharmony in parent-child relationships is more important than isolated aspects
of negative or harsh parenting in determining the nature of adult romantic relationships.
Although it would make for a complex model, the best understanding might come from
combining all these factors – children’s temperament, harsh parenting, and parents’
relationships.
There are several studies that have combined these factors – children’s
characteristics, parenting, and parents’ relationships – in examining future romantic
relationships. A couple of longitudinal studies (Capaldi & Clark, 1998; Donnellan, LarsenRife, & Conger, 2005) have shown that both parenting behaviors toward the child and
adolescent characteristics are important, directly or indirectly, in predicting romantic
relationships later on. However, parents’ marital negativity or parental dyadic aggression
do not seem to affect children’s later relationships with adult partners. Results from other
longitudinal studies are mixed, showing that either parents’ behaviors alone (Conger et al.,
2000) or parents’ marital discord alone (Amato & Booth, 2001) predict children’s marital
quality or romantic relationship quality in early adulthood. These studies, however, have
not included children’s characteristics. There is also empirical evidence suggesting that
conflict-ridden parent-adolescent interactions and adolescents’ depression or antisocial
behavior may predict satisfaction, communication, and aggression in marital or dating
relationships later in life (Andrews et al., 2000; Kim, Conger, Elder, & Lorenz, 2001). Even
though only a few studies have examined the combination of children’s characteristics,
parenting, and parents’ relationships in the prediction of later romantic relationships, results
from other studies indicate that all these three factors in one way or another are important
predictors. However, all these studies have focused on adolescents and young adults. They
have not examined the influence of very early temperament or harsh parenting
combinations on mid-adulthood romantic relationships.
Taken together, it seems that adult relationship satisfaction may be affected by
many factors. From several theoretical perspectives, adult relationships may be influenced
by characteristics of the child, parent-child relationships, and parents’ marital satisfaction.
Previous studies indicate that various children’s or adolescents’ characteristics, parents’
marital relations, or parent-child relationships in one way or another are important in
predicting romantic relationships later in life. In addition, parents’ marital relations may
influence parent-child relationships, and be experienced by the child as conflict or
disharmony; if the child interprets this as parental rejection, such disharmony can modify
the effects of negative parenting. Thus, disharmony in the parent-child relationships may be
an important aspect with regard to the nature of the adult romantic relationship. Though it
would make for a complex model, the best understanding might come from combining
these factors. One previous study, modeled the links between early temperament, different
combinations of harsh parenting, and adolescent problem behavior (Pakalniskiene et al.,
2007), and the results indicated that the combination of physical punishment and discordant
mother-child relationships was linked to adolescent problem behavior, and early
temperament was not clearly related to this harsh parenting profile. In this study, I use the
same longitudinal dataset to test a model that is similar in that I identify harsh parenting
profiles and relate them to early temperament, but different in that I consider the role of
early parental conflict in addition to children’s temperament and I look at the grown
children’s own partner relationships in middle adulthood as the outcome of interest.
In this study, I use longitudinal data from infancy to middle adulthood to model
relations between early temperamental unmanageability and parents’ attitudes toward
marital conflict, childhood experiences of different combinations of harsh parenting, and
middle adult partner relationship quality. Following the idea that disharmony in parentchild relationships could be important for later relationship quality, the different
combinations of harsh parenting behaviors that children experienced included combinations
of physical punishment and relationship discord. To examine relationships between patterns
of harsh parenting behaviors, early temperamental unmanageability, parents’ attitudes
towards marital conflict, and relationship quality in adulthood, I used a mixture model,
which combines variable- and person-centered approaches. This model examined whether
children who experience various combinations of harsh parenting differ with regard to early
unmanageability, parents’ attitudes, and romantic relationship quality later in life.
Method
Participants
STUDY V
All analyses are based on data from a Swedish longitudinal study that started in the
mid 1950s. Participants in this study were investigated by researchers at the Clinic for the
Study of Children's Development and Health, Karolinska Hospital, Stockholm and the
University of Stockholm from birth into adulthood. Every fourth pregnant woman who
registered at the Solna prenatal clinic (in a suburb of Stockholm) from April 1955 to April
1958 was invited to participate in a long-term pediatric study (in Sweden, virtually all
pregnant women receive regular care at prenatal clinics). Only 3 percent of those who were
asked refused to participate. In all, 212 children (122 boys and 90 girls) took part in the
study. Comparisons on variables such as parents’ socioeconomical status (SES), age, and
marital status; sibling order; gestational age; and birth weight have shown the sample to be
representative of populations in other Swedish urban communities (see Karlberg,
Klackenberg, Engström, Klackenberg-Larsson, Lichtenstein, Stensson, & Svennberg, 1968;
Stattin & Klackenberg-Larsson, 1993). At the time of the child’s birth, 192 of the 212
mothers who participated in the study were married to the child’s biological father (in a
first marriage for both partners), 14 were cohabiting, and 5 were single. Most of the
mothers (166) did not work outside the home; 41 were employed outside the home; the
others were self-employed at home, or away from home. The average age of mothers was
27.4 (SD = 5.3); the average age of fathers was 30.2 (SD = 6.0). Parents had been living
together for an average of 3.6 years (SD = 3.5). In 180 families, both parents were Swedish
by birth; in 27 families, one was Swedish; and in 4 families, neither was Swedish.
According to the Graffar classification (Graffar, 1956; 1960), where socioeconomic status
is classified from 1 (highest) to 5 (lowest), about 4% were classified as 1 (high SES); 20%
were classified as 2 (medium-high SES); 31% were classified as 3 (medium SES); 40%
were classified as 4 (medium-low SES); and 4% were classified as 5 (low SES). Children
and their parents were examined four times, equally spaced (every three months) during
their first year, twice (every six months) during the second year, and annually (close to their
birthdays) thereafter up to the age of 18. Data were also collected at the average ages of 21,
25, and 35 years.
At the age of 35, 96 of the 212 participants were married, 65 were cohabitating, and
34 had a romantic partner, but were not living together. The ages of the participants’
partners varied from 21 to 52 years, with a mean of 35.5 (SD = 5.7). Partners had been
living together for an average of 10.2 years (SD = 5.2). Partners who were not living
together had previously been together for an average of 3.5 years (SD = 2.3). Thirty-seven
participants had at least one child; 68 reported having two children, 29 three children, and
10 four or more. One hundred eighty-five participants were working at age 35. Thirty-eight
percent of participants and 35% of participants’ partners had a university degree at age 35.
Measures
Early Unmanageability
For the measure of early unmanageability, I used mothers’ reports of children’s
early unmanageability and resistance to control, making two assessments: 3 – 12 months
and 18 months – 3 years. Child characteristics were assessed with age-appropriate items
(for validation of this measure, see Stattin, Janson, Klackenberg-Larsson, & Magnusson,
1995). When the child was 3 – 12 months, the behavior problem measures included the
following questions: “Does he/she often get angry?” “Does he/she often get extremely
angry?” The alpha reliability for this scale was .81. When the child was 18 months – 3
years, behavioral problems were addressed by the following questions: “Is he/she a noisy
child?” “Does he/she often get angry?” “Does he/she often get extremely angry?” “Is
he/she a destructive child?” “Does he/she want to get his/her own way?” “Is he/she often
disobedient?” The alpha reliability for this scale was .69. For the analysis, I standardized
scores within time points to create a common metric, and the measures were collapsed into
one age period.
Harsh Parenting
For the measures of harsh parenting I used mothers’ reports of their striking (milder
corporal punishment) and beating (stronger corporal punishment), and interviewers’ ratings
of discordant relationships between mothers and children, which were given on each
assessment from 6 to 12 years. For the analysis, mean scores for striking, beating, and
discordant relationships for all the age periods from 6 to 12 years were computed. The
mean scores represent the average level of each variable over this period. As reported
previously (Pakalniskiene et al., 2007), the numbers of mothers who engaged in any
striking ranged from 169 to 68 over ages 6-12, beating ranged from 56 to 73 over ages 6-9
and from 49 to 20 over ages 10-12.
Striking. To measure striking, mothers stated whether or not (and how often) they
struck their child. When children were 6 to 9 years, there were six response options,
ranging from “never” to “on many occasions every day.” Because of a lower incidence of
striking older children, the response options were changed from age 10 to age 12; there
were five response options, ranging from “never” to “daily.” The average period-to-period
correlation for the measures of striking ranged from .45 to .56.
Beating. When the child was 6 to 9 years, mothers stated whether or not they had
given their child a real beating. For children ages 10 to 12 years, mothers reported beating
frequency. There were five response options, ranging from “never” to “once a day.” To
keep consistency in measures over time we dichotomized this measure. The average periodto-period correlation for beating ranged from .31 to .39.
Discordant Mother-Child Relationships. After each interview occasion from
the child’s ages 6 to 12 years, the interviewers made judgments of the quality of the
mother-child relationship. The interviewers used a 3-point Likert scale with the scale points
described as follows: (1) good relationship – no conflict or maternal insensitivity; (2) some
evidence of conflict, disagreement, or insensitivity, or inconsistent relationship quality –
sometimes good and other times bad; and, (3) pronounced conflict or apparent insensitivity
(see Stattin & Klackenberg, 1992, for a validation of this measure) The period-to-period
correlations for discordant relationships ranged from .44 to 66.
Relationship Quality with a Partner
STUDY V
Participants evaluated 10 items describing the relationship with their romantic
partner at age 35. The questions were: “Does your partner talk with you about his/her
problems?” (reversed), there were four response options, ranging from “yes, always” to
“never;” “how warm do you feel towards your partner?” (reversed), there were five
response options, ranging from “very much” to “not at all;” “how do you get along with
your partner?,” there were five response options, ranging from “bad” to “very good;” “how
often do you get really mad at your partner?” (reversed), there were five response options,
ranging from “never” to “often;” “how would you describe your husband/wife?,” there
were five response options, ranging from “only negative features” to “only positive;” “do
you and your husband/wife have any interests in common?,” there were five response
options, ranging from “no interests in common, no possibilities for recreation together” to
“totally share each other’s activities always with the same pleasure;” “if you would give a
picture of your relationships how would you describe the atmosphere at home?,” there were
six response options, ranging from “very disharmonious, divorce atmosphere” to “very
harmonious, we share the same attitudes, open, warm home atmosphere;” “how often do
you cuddle? If you think one month back, how often did you spontaneously kiss or hug
each other?,” there were five response options, ranging from “not at all” to “daily, almost
daily;” “how is your sexual life? Are you well adapted sexually to each other? Do you
function well together?, “there were six response options, ranging from “have no sexual life
or very seldom” to “very well adapted;” “does you partner give you encouragement and
support when you have troubles at work?” (reversed), there were five response options,
ranging from “I get all the help I need” to “my partner is more of an obstacle.” For the
analyses we computed a mean score of all the items. For the latent profile analysis, we
standardized all the question scores to create a common metric and created a scale from the
standardized items, with higher scores indicating better relationship quality. The alpha
reliability for this scale was .81.
Attitudes towards Marital Conflict
Both mothers and fathers responded to 5 statements describing attitudes towards
marital conflict. The statements were: “people who believe that they can manage their
marriage without any arguments do not know anything about reality;” “sometimes it is
necessary for the wife/husband to disagree with what they say to get their voice heard;” “it
does not matter how much partners like each other, there are always differences between
them that cause irritation and quarrels;” “there are some thing that cannot be made up with
just a discussion;” and, “when two people with a will of their own get married it is natural
that they will quarrel.” There were four response options, ranging from “strongly disagree”
to “strongly agree.” Higher scores indicated positive attitudes towards marital conflict. For
the analyses we created a mean score of all the items. The alpha reliability for this scale
was .75.
Data Analysis
For all the analyses, I used Mplus 4.0 (Muthén & Muthén, 2006) software with
FIML (full information maximum likelihood) estimation. The FIML estimation allowed me
to use the complete sample in this study, even when some data were missing. The reason
for missing cases was due mostly to some data not being available for a specific wave or
due to not having answers to some questions, rather than due to participants’ attrition.
Using ordinary t-tests, I compared participants with incomplete data (87) and participants
with complete data (125) on the variables used in these analyses. The two groups did not
differ significantly on any of the variables used. Thus, these missing data can be considered
missing at random. There is evidence that FIML estimation provides less biased estimates
than listwise or pairwise deletion (Shafer & Graham, 2002).
Mixture Modeling
Results
In this study, harsh parenting behaviors and their relationships to early
unmanageability, parents’ attitudes, and relationship quality with a partner in adulthood
were tested in the model presented in Figure 1. First, the latent classes of mother’s harsh
behaviors such as striking, beating, and discordant mother-child relationships during late
childhood and early adolescence at ages 6-12 were tested. Following the idea that
covariates could improve model fit and also improve accuracy of assignment individuals to
certain classes or profiles (Muthèn, 2003), early unmanageability and parental attitudes
towards marital conflicts were included in model building and testing latent classes. Models
with increased numbers of classes were compared (Table 1) using the Bayesian information
criterion (BIC), the sample-size adjusted Bayesian information criterion (SSABIC), the
Akaike information criterion (AIC), and the Entropy criterion. Lower scores on the BIC,
STUDY V
To identify the patterns of harsh parenting that children had experienced latent class
model was chosen. Latent class modeling refers to modeling with categorical latent
variables representing groups generated from the data. Latent profile analysis, which is a
type of latent class analysis, predicts latent class membership from continuous indicators
(Muthén & Muthén, 2006), while latent class analysis predicts class membership from
categorical data. Even though class membership in this study was predicted from
continuous data, I will use the term latent class analysis, because this term is mainly used in
mixture modeling.
In this study, I combined the latent class modeling with early covariates and distal
outcomes. Models of this type are called mixture models (Muthén & Muthén, 2006). The
mixture modeling tested in this study is presented in Figure 1 and estimated
simultaneously. Based on the previous research, the child’s early temperament and parents’
marital conflict served as early covariates; and relationship with a partner at age 35 served
as the distal outcome. This model helps to test whether children who have higher levels of
unmanageable temperament early on are likely to experience certain patterns of harsh
parenting (A in Figure 1) and whether these experiences of harsh parenting are
systematically related to relationship quality with a partner later in life (B in Figure 1). The
pathways between the early covariates and the distal outcome represent the extent to which
early unmanageability and parents’ attitudes towards marital conflict are related to partner
relationship quality later in life, after controlling for the risk associated with experiencing
harsh parenting (C in Figure 1).
SSABIC, and AIC represent better fitting models (Muthén & Muthén, 2000; Schwartz,
1978). Entropy refers to the average classification accuracy in the assignment of
participants to classes or profiles, and higher scores represent greater accuracy (Bauer &
Currant, 2003). Entropy values can range from 0 to 1, with values closer to 1 indicating
better classifications of individuals into specific classes (Bauer & Currant, 2003). The fir
indices for the classes are very similar to those reported previously (Pakalniskiene et al.,
2007), with small changes reflecting the addition of parental marital conflicts. Although the
BIC value indicated that models with three or four classes have very similar fit, the AIC
indicated that four-class solution fits better than three-class solution. However, the fiveclass solution had the best of all four fit indices. Comparing three-, four-, and five-class
solutions it was decided that in order to have a meaningful, adequate size of each class, the
four-class solution should be chosen, Very small classes can create problems with analysis
and the interpretation of results (Nylund, Asparouhov, & Muthén, 2006).
In the next step, the adulthood outcome (relations with a partner at age 35) was
added to the model (Log likelihood = -643.257; BIC = 1482.919, SSA BIC= 1365.696,
AIC = 1360.513, entropy = .824). Table 2 shows the mean values of mothers’ behaviors for
the four class solution with the adulthood outcomes included in the model. Four-class
solution model presents very similar profiles to reported earlier (Pakalniskiene et al., 2007).
Four different classes were identified: a normative class consisting of children (n = 103)
who very seldom experienced harsh parenting; a physical punishment class, consisting of
children (n = 62) who frequently experienced mother’s striking and beating, but not
discordant relationships; a discordant relationships class, consisting of children (n = 22)
who experienced discordant mother-child relationships and who very seldom experienced
mother’s striking or beating; and a harsh treatment class, consisting of children (n = 15)
who frequently experienced mother’s striking, beating, and discordant relationships. It
seems that children during late childhood and early adolescence could experience various
types mother’s behaviors. It could be only physical punishment, physical punishment
together with conflicted relationships, or only conflicted mother-child relationships. Chisquare analysis failed to detect significant gender differences in the four classes. This
suggests that, in each class, there are similar proportions of girls and boys.
Parents’ attitudes towards marital conflict, children’s early
unmanageability, and the risk of experiencing specific mothers’
harsh behavior
Previous studies suggested that parent-child relationships or parents’ behavior could
be influenced by the child’s temperament or even parents’ marital satisfaction. To estimate
Class membership and adulthood outcomes
It seems that unmanageable temperament increased children’s risk of experiencing
certain parents’ behavior. The question is whether certain parents’ behavior could affect
children’s later relationships. To test whether the romantic relationship quality varied
across classes (B in Figure1), I ran the models with and without holding the means of
relationship quality equal across classes to get the chi-square (twice the log likelihood)
differences. Because I made multiple comparisons, I used a more conservative significance
level (p < .01) than the conventional significance level (p < .05). Significance testing
revealed (third column in Table 3) that children who experienced discordant relationships
and harsh treatment had significantly lower quality relationships in adulthood than children
in the normative and physical punishment classes (M = -.361, SD = .112, M = -.344, SD =
.130, for the discordant relationships and harsh treatment classes, respectively). However,
STUDY V
the relationships between children’s temperament, parents’ marital satisfaction, and
parents’ behavior latent class membership was regressed on parents’ attitudes towards
marital conflict and children’s early temperament (A in Figure 1). The risk for
experiencing certain mothers’ behavior is given in reference to the normative class, which
had low levels of mother’s striking, beating, and discordant relationships, because in this
comparison, one of the classes serves as the reference category. Results are presented in
odds ratios and presented in the columns labeled A in Table 3. Relative to children in the
normative class, children with higher levels of early unmanageability were more likely to
be in the harsh treatment class (OR = 1.96, p < .05, 95% CI – 1.02-3.75) or showed a
tendency to be in the physical punishment class (OR = 1.50, 90% CI – 1.04-2.18).
However, children with higher levels of unmanageability were not at increased risk of
being in the discordant relationships class (OR = .84, 95% CI – .39-1.84). Across all
classes, parents’ attitudes towards marital conflict at age 4 did not increase the risk for
children of experiencing certain parents’ behavior (OR = 1.04, 95% CI – .47-2.28, OR =
1.11, 95% CI – .74-1.67, OR = 1.08, 95% CI – .66-1.77 for discordant relationships,
physical punishment, and harsh treatment class, respectively). Taken together,
unmanageable temperament increased children’s risk of experiencing physical punishment
in the context of discordant relationships and, to a lesser extent, physical punishment in the
context of good relationships later in childhood. It seems that children’s unmanageable
temperament could evoke mothers’ physical punishment. However, parents’ attitudes
towards marital conflict did not increase children’s risk of experiencing certain parental
behaviors.
children who did not experience mother’s harsh behavior or who experienced only mother’s
physical punishment had significantly higher quality relationships with their partners in
adulthood than other children (M = .056, SD = .064, M = .142, SD = .090 for the normative
and physical punishment classes, respectively). Thus, children who experienced physical
punishment in the context of conflicted mother-child relationships or experienced
conflicted relationships alone had worse relationships later in life than other children.
Contribution of early unmanageable temperament and parents’
attitudes towards marital conflict to the prediction of adulthood
relationships
It was found that unmanageable temperament increased children’s risk of
experiencing certain parents’ behavior. In this model, the links from children’s early
unmanageability and parents’ attitudes towards marital conflict to later relationship quality
were also examined for each of the classes (C in Figure 1).
Results are presented in the last two columns in Table 3. Results are expressed in
standardized estimates provided by the program. For the significance levels, Z statistics
were used, so values that exceed +1.96 or fall below -1.96 are significant at p < .05, and
values that exceed +1.65 or fall below -1.65 are significant at p < .10. Results showed that
high early unmanageability was significantly related to the likelihood of having a poor
relationship with a partner at age 35 over and above the risk associated with class
membership (Est. = -.443, z = -3.952) for children who experienced harsh treatment; there
was no significant link between parents’ attitudes towards marital conflict and relationship
quality (Est. = .215, z = 1.161). For children in the discordant relationships class, where the
quality of relationships with a partner at age 35 was also the lowest, early temperament did
not contribute to the likelihood of having poor relationship quality in adulthood, over and
above the risk associated with class membership (Est. = -.144, z = -.776). However, there
was a significant link between parents’ attitudes towards marital conflict and relationship
quality apart from the risk associated with the class membership (Est. = -.294, z = -1.979).
For children in the normative and physical punishment classes, neither early
unmanageability nor parents’ marital conflicts added to the likelihood of having poor
relationship quality in adulthood over and above the risk associated with class membership.
Thus, for children who experienced harsh treatment, temperamental unmanageability
seemed to play a role in the development of poor relationships. Also, it seems that parents’
positive attitudes towards marital conflicts could be related to relationships in adulthood,
but only for children who experienced discordant mother-child relationships. What is more,
children who experienced discordant relationships or mother’s physical punishment in the
context of discordant relationships had poor romantic relationships later in life. Even
though children’s temperamental unmanageability may be a risk factor for experiencing
mother’s physical punishment or harsh treatment, the children who experienced mother’s
physical punishment in the context of good relationships had much better relationship
quality than children who experience conflicts at home. These differences suggest that
conflicted attitudes and conflicted atmosphere at home is related to satisfaction later in life,
while child’s temperament could be more related to parents’ physical punishment.
Discussion
STUDY V
Previous studies have suggested that children’s characteristics, or early
temperament (e.g., Kim et al., 2001; Newman et al., 1997), and negative or harsh parenting
(e.g., Andrews et al., 2000; Capaldi & Clark, 1998; Flouri & Buchanan, 2002; Franz et al.,
1991) may affect romantic relationships in young adulthood. Also, previous studies suggest
that parents’ values, attitudes toward marriage or parents’ marital conflict may affect their
children’s romantic relationships. In addition, parents’ marital conflict might influence
parent-child relationships and be perceived by the child as disharmony; if the child
interprets this as parental rejection, such disharmony can modify the effects of harsh
parenting (Deater-Deckard, et al., 1996; Lansford, et al., 2004; McLoyd & Smith, 2002;
Rohner, et al., 1996). It seems that best understanding might come from combining these
factors. Even though there have been studies combining all or some of the factors, they
have focused on adolescents’ and young adults’ characteristics or relationships, and do not
examine the influence of very young children’s characteristics or various negative
parenting combinations on mid-adulthood romantic relationships. Using a mixture
modeling approach, with latent classes representing different parenting profiles, I modeled
the relations between temperamental unmanageability, parents’ attitudes, harsh parenting
profiles, and later romantic relationship quality. Taken together, the results show that
various parenting profiles play different roles in the links between early temperamental
unmanageability, parents’ attitudes towards marital conflict, and later romantic relationship
quality.
In the previous study, the same parenting profiles proved to be important in
understanding adolescent problem behavior (Pakalniskiene et al., 2007). But this study
added to that by exploring a completely new domain of social adjustment: adult
relationship quality. In this study, also the new covariate - parents’ attitudes towards marital
conflicts – was added to the latent class model presented in the previous study
(Pakalniskiene et al., 2007). The results suggested that parents’ attitudes could influence
their own behavior as well as their children’s attitudes towards relationships in the future.
All in all, this study and the previous study suggest that these parenting profiles have
implications not only for adolescence, but also for adulthood.
From the perspective of temperament and personality theory, it seems that
unmanageable temperament contributes to the way romantic partners get along over time,
and also contributes to the way parents get along with their children. First, unmanageable
temperament contributes to the way romantic partners get along over time. For children
who experience physical punishment in the context of discordant relationships, their own
unmanageability seems to play a role in their later romantic relationships. It seems that
early temperament plays a role in the later romantic relationships of some children, but not
of others. Second, early unmanageable temperament puts children at risk of physical
punishment, together with a discordant mother-child relationship, and to a lesser extent
increases the risk of experiencing physical punishment in the context of a good motherchild relationship. It seems that early characteristics of the child also affect the ways in
which parents behave with their children. Thus, unmanageable temperament contributes to
the ways in which some children get along with other people, either their parents, or their
romantic partners.
From a learning theory perspective, the nature of child upbringing may also have
consequences for romantic relationships. For children who experience physical punishment
in the context of discordant relationships, parents’ behaviors seem to play a role in their
later romantic relationships. The same tendency applies to children with a discordant
relationship profile. It seems that experiencing conflict relationships, with or without
physical punishment, during childhood and adolescence will lead to poor romantic
relationships in adulthood. These results suggest that emotional context in the family might
be more potentially damaging than any other type of parental behavior, such as physical
punishment. In support of this idea, our results suggest that children who experience
physical punishment in the context of good relationships, and also children who do not
experience mothers’ harsh behaviors, have romantic relationships of a very similar quality
later in life. Thus, it seems that physical punishment alone does not interfere with
relationship quality, but emotional context, created by other parental behavior in the family,
is an important factor with regard to later relationships. It may be that children from
different groups interpret their parents’ behaviors differently. For example, children who
experience harmonious relationships, or even physical punishment in harmonious
relationships, might interpret their mother’s behavior as reflecting their mother’s wish to
make them better persons; these children might know and feel that their parents love them.
STUDY V
And children who experience discordant relationships or physical punishment may interpret
their mother’s behaviors as rejection. Children’s interpretations of their mothers’ behaviors
may change the course of their later romantic relationships.
Also, from a learning theory perspective, certain parental attitudes may also
influence what will happen later in a child’s life. Mainly, it has been suggested in previous
studies that parents’ marital satisfaction will influence later romantic relationships (e.g.,
Amato & Booth, 2001; O’Leary & Cascardi, 1998; Stocker & Richmond, 2007). However,
it has also been suggested that an orientation towards a romantic partner that has been
learned during childhood will eventually affect adult relationships (Waller & Shaver,
1994). It seems that parents influence the attitudes and behaviors of their children towards a
romantic partner. Nevertheless, parents’ attitudes towards marital conflict may affect later
romantic relationships only in the case of children who have experienced discordant
relationships. If children know that marital conflict is an accepted part of relationships, and
also experience conflict in mother-child relationships, they might learn that conflict is
necessary in any type of relationship. By contrast, for other children, who also experience
physical punishment, their unmanageable temperament may play a more important role in
the development of relationships than their parents’ attitudes. Whatever the reason, it
seems, from a learning theory perspective, that certain parental attitudes during early
childhood and parent-child relationships will affect their children’s lives in adulthood.
One possible limitation of the study is that our data were collected in the mid-1950s
and onwards. The question is whether children’s view of striking or beating differs from
these days. We assume that if this study was repeated in Sweden today, the findings
concerning physical punishment would be very different. Sweden was the first country
worldwide to outlaw the physical punishment of children. The law came into force in 1979,
so a whole cohort of Swedes has now grown into adulthood amidst public awareness of this
issue. However, the conditions at the time our sample was taken were very similar to those
in North America, where most harsh or negative parenting studies have come from.
Similarity between Sweden and North America can be inferred from the sheer numbers of
mothers who report striking and beating their children in both countries. In Sweden, for
striking, when children in our sample were 6 or 7 years-old, it was virtually all of the
mothers, to one degree or another. When the children were 9 years-old, a third of mothers
reported having given their child “a real beating.” In nationally representative samples from
North America, 85% of parents use corporal punishment, such as slapping, spanking,
hitting, or shaking, with their 6 year old children. Despite declining after age 6, over half of
American parents hit their children at age 12 (Straus & Stewart, 1999). Thus, we think it is
safe to conclude that children’s views of physical punishment in our sample are similar to
those of children in countries where physical punishment is still legal, and even
encouraged. This suggests that the findings concerning physical punishment in Sweden
may be very similar to findings from countries where corporal punishment is legal.
The present study has a number of strengths. Perhaps the main strength is its
prospective, long-term, longitudinal design, with data collected before the children’s first
birthday and the follow-up of the children into adulthood. These data allowed us to
examine connections between harsh parenting, very early unmanageable temperament,
parents’ attitudes, and later relationship quality in a more elaborate manner than ever before
reported in the literature. Another strength lies in its use of a community sample, which, in
general, does not differ much from other Swedish children of the same cohort (Karlberg et
al., 1968). A further strength of the study consists in its usage of data from different
reporters. This gives us confidence that the results of our models are not just due to rater
bias, but tap into actual developmental processes.
Parents want their children to establish and maintain good relationships with other
people, and be happy in romantic relationships. However, depending on the child’s
temperamental predispositions or behaviors, and parents’ behaviors or attitudes, this may
not be accomplishable. Having a child who is temperamentally prone to anger or aggression
is a challenge to some parents, and they can use physical punishment or create conflict in
parent-child relationships. It seems that physical punishment does not necessarily make a
child’s life worse. However, emotional context in the family may contribute to the
development of various problems that will interfere with the goals of society and the
individual’s own enjoyment of life, which arouses concern. It seems that parents need to be
aware that their ways of thinking, feeling, and behaving can affect their child’s future.
References
STUDY V
Amato, P. R., & Booth, A. (2001). The legacy of parents’ marital discord: Consequences
for children’s marital quality. Journal of Personality and Social Psychology, 81,
627-638.
Andrews, J. A., Foster, S. L., Capaldi, D., & Hops, H. (2000). Adolescent and family
predictors of physical aggression, communication, and satisfaction in young adult
couples: A prospective analysis. Journal of Consulting and Clinical Psychology, 68,
195-208.
Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice-Hall.
Bates, J. E., Pettit, G. S., Dodge, K. A., & Ridge, B. (1998). Interaction of temperamental
resistance to control and restrictive parenting in the development of externalizing
behavior. Developmental Psychology, 34, 982 – 995.
Bauer, D. J., & Currant, P. J. (2003). Distributional assumptions of growth mixture models:
Implications for overextraction of latent trajectory classes. Psychological Methods,
8, 338 – 363.
Blum, J., & Mehrabian, A. (1999). Personality and temperament correlates of marital
satisfaction. Journal of Personality, 67, 93-125.
Bouchard, G., Lussier, Y., & Sabourin, S. (1999). Personality and marital adjustment:
Utility of the Five-Factor Model of Personality. Journal of Marriage and the
Family, 61, 651-660.
Capaldi, D. M., & Clark, S. (1998). Prospective family predictors of aggression toward
female partners for young at-risk males. Developmental Psychology, 34, 1175-1188.
Caspi, A. (2000). The child is father of the man: Personality continuities from childhood to
adulthood. Journal of Personality and Social Psychology, 78, 158 – 172.
Caspi, C., & Silva P. (1995). Temperamental qualities at age three predict personality traits
in young adulthood: Longitudinal evidence from a birth cohort. Child Development,
66, 486-498.
Caspi, A., Elder, G. H., & Bem, D. J. (1987). Moving against the world: Life-course
patterns of explosive children. Developmental Psychology, 23, 308-313.
Colman, R. A., & Widom, C. S. (2004). Childhood abuse and neglect and adult intimate
relationships: a prospective study. Child Abuse and Neglect, 28, 1133-1151.
Conger, R. D., Cui, M., Bryan, C. M., & Elder, G. H. (2000). Competence in early adult
romantic relationships: A developmental perspective on family issues. Journal of
Personality and Social Psychology, 79, 224 – 237.
Davis, J. L., & Petretic-Jackson, P. A. (2000). The impact of child sexual abuse on adult
interpersonal functioning: A review and synthesis of the empirical literature.
Aggression and Violent Behavior, 5, 291-323.
Deater-Deckard, K., Dodge, K. A., Bates, J. E., & Pettit, G. S. (1996). Physical discipline
among African American and European American mothers: Links to children’s
externalizing behaviors. Developmental Psychology, 32, 1065 – 1072.
Donnellan, M. B., Conger, R. D., & Bryan, C. M. (2004). The Big Five and enduring
marriages. Journal of Research in Personality. 38, 481-504.
Donnellan, M. B., Larsen-Rife, D., & Conger, R. D. (2005). Personality, family history,
and competence in early adult romantic relationships. Journal of Personality and
Social Psychology. 88, 562-576.
Eisenberg, N., Fabes, R. A., Guthrie, I.K., & Reiser, M. (2000). Dispositional emotionality
and regulation: Their role in predicting quality of social functioning. Journal of
Personality and Social Psychology, 78, 136-157.
Fauber, R., Forehand, R., Thomas, A. M., & Wierson, M. (1990). A mediational model of
the impact of marital conflict on adolescent adjustment in intact and divorced
families: The role of disrupted parenting. Child Development, 61, 1112-1123.
Felitti, V. J. (1991). Long-term medical consequences of incest, rape, and molestation.
Southern Medical Journal, 84, 328-331.
Finkelhor, D., Hotaling, G. T., Lewis, I. A., & Smith, C. (1989). Sexual abuse and its
relationship to later sexual satisfaction, marital status, religion, and attitudes.
Journal of Interpersonal Violence, 4, 379-399.
Fleming, J., Mullen, P. E., Sibthorpe, B. & Bammer, G. (1999). The long-term impact of
childhood sexual abuse in Australian women. Child Abuse and Neglect, 23, 145159.
Flouri, E., & Buchanan, A. (2002). What predicts good relationships with parents in
adolescence and partner in adult life: Findings from the 1958 British birth cohort.
Journal of Family Psychology, 16, 186-198.
Franz, C. E., McClelland, D. C., & Weineberger, J. (1991). Childhood antecedents of
conventional social accomplishments in mid-life adults: A 35-year prospective
study. Journal of Personality and Social Psychology, 60, 586-595.
Graffar M (1956) Une methode de classification sociale d echantillons de population.
Courrier, 332,133 – 8.
Graffar, M. (1960). Social study of the samples. In S. Karger (Ed). Modern problems in
pediatrics. Basel.
STUDY V
Haapasalo, J., & Tremblay, R. E. (1994). Physically aggressive boys from ages 6 to 12:
Family background, parenting behavior, and prediction of delinquency. Journal of
Consulting and Clinical Psychology, 62, 1044-1052.
Hanson, W. B., Tobler, N. S., & Graham, J. W. (1990). Attrition in substance abuse
prevention research: A meta-analysis of 85 longitudinally followed cohorts.
Evaluation Review, 14, 677 – 685.
Hibbard, J. H., & Pope, C.R. (1991). Effects of domestic and occupational roles on
morbidity and mortality. Social Science and Medicine, 32, 805–811.
Huston, T. L., & Houts, R. M. (1998). The psychological infrastructure of courtship and
marriage. The role of personality and compatibility in romantic relationships. In T.
N. Bradbury (Ed.), The developmental course of marital dysfunction (pp.114-161).
Cambridge: Cambridge University.
Karlberg, P., Klackenberg, G., Engström, I., Klackenberg-Larsson, I., Lichtenstein, H.,
Stensson, J., & Svennberg, I. (1968). The development of children in a Swedish
urban community: A prospective, longitudinal study – Parts I-VI. Acta Paediatrica
Scandinavica Supplement, Whole No. 187.
Karney, B. R., & Bradbury, T. N. (1995). The longitudinal course of marital quality and
stability: A review on theory, methods, and research. Psychological Bulletin, 118, 334.
Kelly, E. L., & Conley, J. J. (1987). Personality and compatibility: A prospective analysis
of marital stability and marital satisfaction. Journal of Personality and Social
Psychology, 52, 27-40.
Kiecolt-Glaser, J. K., & Newton, T. L. (2001). Marriage and health: His and hers.
Psychological Bulletin, 127, 472-503.
Kim, K. J., Conger, R. D., Elder, G. H., & Lorenz, F. O. (2001). Parent-adolescent
reciprocity in negative affect and its relation to early adult social development.
Developmental Psychology, 37, 775-790.
Kinnunen, U., & Pulkkinen, L. (2003). Childhood socio-emotional characteristics as
antecedents of marital stability and quality. European Psychologist, 8, 223-237.
Larzelere, R. E., & Kuhn, B. R. (2005). Comparing child outcomes of physical punishment
and alternative disciplinary tactics: A meta-analysis. Clinical Child and Family
Psychology Review, 8, 1-37.
Linder, J. R., & Collins, W. A. (2005). Parent and peer predictors of physical aggression
and conflict management in romantic relationships in early adulthood. Journal of
Family Psychology, 19, 252-262.
Lansford, J. E., Deater-Deckard, K., Dodge, K. A., Bates, J., & Pettit, G. S. (2004). Ethnic
differnces in the link between physical discipline and later adolescent externalizing
behaviors. Journal of Child Psychology and Psychiatry, 45, 801-812.
Leve, L. D., Kim, H. K., & Pears, K. C. (2005). Childhood temperament and family
environment as predictors of internalizing and externalizing trajectories from age 5
to 17. Journal of Abnormal Child Psychology, 33, 505 – 520.
McLoyd, V. C., & Smith, J. (2002). Physical discipline and behavior problems in African
American, European American, and Hispanic children: Emotional support as a
moderator. Journal of Marriage and Family, 64, 40 – 53.
Morris, A. S., Silk, J. S., Steinberg, L., Sessa, F. M., Avenevoli, S., & Essex, M. J. (2002).
Temperamental vulnerability and negative parenting as interacting of child
adjustment. Journal of Marriage and the Family, 64, 461-471.
Muthén, B. O., & Muthén, L. K. (2000). Integrating person-centered and variable-centered
analyses: Growth mixture modeling with latent trajectory classes. Alcoholism:
Clinical and Experimental Research, 24, 1 – 10.
Muthén, L. K., & Muthén, B. O. (2006). Mplus users guide. Los Angeles, CA: Muthén and
Muthén.
Newman, D. L., Caspi, A., Moffitt, T. E., & Silva P. A. (1997). Antecedents of adult
interpersonal functioning: Effects of individual differences in age 3 temperament.
Developmental Psychology, 33, 206-217.
Nylund, K. L., Asparouhov, T., & Muthén, B. O. (2006). Deciding on the number of
classes in the latent class analysis and growth mixture modeling: A Monte Carlo
simulation study. Manuscript submitted for publication.
O’Leary, K. D., & Cascardi, M. (1998). Physical aggression in marriage: A developmental
analysis. In T. N. Bradbury (Ed.), The developmental course of marital dysfunction
(pp.344-374). Cambridge: Cambridge University.
Pakalniskiene, V., Kerr, M., & Stattin, H. (2007). Early temperamental unmanageability,
harsh parenting profiles, and adolescent problem behavior: A mixture modeling
approach with latent parenting classes. Manuscript submitted for publication.
Pettit, G. S., Laird, R. D., Dodge, K. A., Bates, J. E., & Criss, M. M. (2001). Antecedents
and behavior-problem outcomes of parental monitoring and psychological control in
early adolescence. Child Development, 72, 583–598.
Rohner, R. P., Bourque, S. L., & Elordi, C. A. (1996). Children’s perceptions of corporal
punishment, caretaker acceptance, and psychological adjustment in a poor, biracial
Southern community. Journal of Marriage and the Family, 58, 842 – 852.
STUDY V
Ross, C. E., Mirowsky, J., & Goldsteen, K. (1990). The impact of the family on health: The
decade in review. Journal of Marriage and the Family, 52, 1059-1078.
Rutter, M. (1987). Temperament, personality, and personality disorder. British Journal of
Psychiatry, 150, 443-458.
Sanders, M. R., Halford, W. K., & Behrens, B. C. (1999). Parental divorce and premarital
couple communication. Journal of Family Psychology, 13, 60-74.
Schwartz, G. (1978). Estimating the dimension of the model. The Annals of Statistics, 6,
461 – 464.
Shafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art.
Psychological Methods, 7, 147 – 177.
Snyder, M., & Ickes, W. (1985). Personality and social behavior. In G. Lindzey, & E.
Aronson (Eds.), Handbook of social psychology, 3rd ed., Vol. 2: 883-947. New
York: Random House.
Stattin, H., Janson, H., Klackenberg-Larsson, I., & Magnusson, D. (1995). Corporal
punishment in everyday life: An intergenerational perspective. In J. McCord (Ed.),
Coercion and punishment in long-term perspectives (pp. 315-347). Cambridge:
Cambridge University Press.
Stattin, H., & Klackenberg-Larsson, I. (1990). The relationship between maternal attributes
in the early life of the child and the child’s future criminal behavior. Development
and Psychopathology, 2, 99 – 111.
Stattin, H., & Klackenberg, G. (1992). Discordant relations in intact families:
Developmental tendencies over 18 years. Journal of Marriage and the Family, 54,
940-956.
Stocker, C. M., & Richmond, M. K. (2007). Longitudinal associations between hostility in
adolescents’ family relationships and friendship and hostility in their romantic
relationships. Journal of Family Psychology, 21, 490- 497.
Stoolmiller, M. (2001). Synergistic interaction of child manageability problems and parentdiscipline tactics in predicting future growth in externalizing behavior for boys.
Developmental Psychology, 37, 814 – 823.
Straus, M. A., & Stewart, J. H. (1999). Corporal punishment by American parents: National
data on prevalence, chronicity, severity, and duration, in relation to child and family
characteristics. Clinical Child and Family Psychology Review, 2, 55-70.
Waller. N. G., & Shaver, P. R. (1994). The importance of nongenetic influences on
romantic love styles: A twin-family study. Psychological Science, 5, 268-274.
Watson, D., Hubbard, B., & Wiese, D. (2000). General traits of personality and affectivity
as predictors of satisfaction in intimate relationships: Evidence from self- and
partner-ratings. Journal of Personality, 68, 413-449.
Table 1
Fit Indices for Latent Class Solutions
Number of
Log
parameters likelihood
Number of latent
classes
1 class
2 classes
3 classes
4 classes
5 classes
6 classes
6
10
14
18
22
26
-639.147
-577.853
-538.513
-524.804
-500.154
-492.139
BIC
SSABIC
AIC
Entropy
1310.174
1218.789
1151.341
1145.188
1117.090
1112.274
1291.135
1157.107
1106.986
1108.130
1047.389
1081.921
1290.295
1175.706
1105.025
1095.609
1044.308
1016.279
.893
.964
.853
.889
.842
STUDY V
Table 2
Means of Mothers’ Behavior for Four Latent Class Solution
Striking
Class
Normative class
Physical punishment class
Discordant relationships class
Harsh treatment class
-.304
.360
-.213
.820
Beating
-.638
.861
-.405
1.035
Discordant relationships
-.382
-.274
1.362
1.596
1.96 *
1.50 †
.84
Reference
Latent parenting class
Harsh treatment
Physical punishment
Discordant relationships
Normative
1.08
1.11
1.04
Reference
A
Marital conflict
predicting latent
class1
-.34
.14
-.36
.06
a
b
a
b
B
Mean
relationship
quality2
*p < .05; †p < .10
1
Odds ratios; 2 Mean values; 3 Estimates
a
Significantly higher than all others; b Significantly lower than all others
A
Temperament
predicting latent
class1
-.44 ***
.10
-.14
-.06
C
Temperament
predicting
relationship quality 3
Relationship quality at age 35
.22
.09
-.29 *
-.05
C
Marital conflict
predicting
relationship quality 3
Table 2. Summary of Mixture Modeling Results for Models with Conduct Problems and Norm Violations as Distal Outcomes. Column
Headings A, B, and C Refer to the Pathways Shown in Figure 1.
STUDY V
Figure Captions
Figure 1. Conceptual latent classes of mothers’ behavior model with risk factors
associated with latent classes and the prediction of distal outcomes.
Covariates:
- Early unmanageability 3m –
3y
- Mother-reported attitudes
towards marital conflict 4y
A
Striking
6 – 12 years
C
Class
Membership
Beating
6 – 12 years
STUDY V
B
Discordant
relations
6 – 12 years
Outcome:
Relations with a
partner at age 35
Publications in series Örebro Studies in Psychology
1. Andershed, Henrik, Antisocial Behavior in Adolescence – The Role of Individual
Characteristics. 2002.
2. Trost, Kari, A new look at parenting during adolescence: Reciprocal interactions
in everyday life. 2002.
3. Jensen, Eva, (Mis)understanding and Learning of Feedback Relations in a Simple
Dynamic System. 2004.
4. Wester Herber, Misse, Talking to me? – Risk Communication to a diverse
Public. 2004.
5. Boersma, Katja, Fear and avoidance in the development of a persistent
musculoskele tal pain problem. Implications for secondary prevention. 2005.
6. Jansson, Markus, Insomnia: Psychological Mechanisms and Early Intervention.
A Cognitive-Behavioral Perspective. 2005.
7. Lindblom, Karin, Utbrändhet i normalbefolkningen. Arbets- och individfaktorers
relation i utvecklingen mot eller återhämtning från utbrändhet. 2006.
(Vetenskaplig uppsats)
8. Almqvist, Lena, Children’s Health and Developmental Delay: Positive
Functioning in Every-day Life. 2006.
9. Johansson, Peter, Understanding psychopathy through the study of long-term
violent offenders. 2006.
10. Persson, Andreas, Leisure in Adolescence: Youth’s activity choices and why they
are linked to problems for some and not others. 2006.
11. Ojala, Maria, Hope and worry: Exploring young people’s values, emotions, and
behavior regarding global environmental problems. 2007.
12. larsson, Mats, Human Iris Characteristics as Biomarkers for Personality. 2007.
13. pakalniskiene, Vilmante, Harsh or Inept Parenting, Youth Characteristics and
Later Adjustment. 2008.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement