"For a better life..." A study on migration and health in Nicaragua; Cecilia Gustafsson $(function(){PrimeFaces.cw("Tooltip","widget_formSmash_items_resultList_43_j_idt799_0_j_idt801",{id:"formSmash:items:resultList:43:j_idt799:0:j_idt801",widgetVar:"widget_formSmash_items_resultList_43_j_idt799_0_j_idt801",showEffect:"fade",hideEffect:"fade",target:"formSmash:items:resultList:43:j_idt799:0:fullText"});});

"For a better life..." A study on migration and health in Nicaragua; Cecilia Gustafsson $(function(){PrimeFaces.cw("Tooltip","widget_formSmash_items_resultList_43_j_idt799_0_j_idt801",{id:"formSmash:items:resultList:43:j_idt799:0:j_idt801",widgetVar:"widget_formSmash_items_resultList_43_j_idt799_0_j_idt801",showEffect:"fade",hideEffect:"fade",target:"formSmash:items:resultList:43:j_idt799:0:fullText"});});
“For a better life…”
A study on migration and health in Nicaragua
Cecilia Gustafsson
Department of Geography and Economic History
Umeå University, Sweden
GERUM 2014:2
GERUM-Kulturgeografi 2014:2
Institutionen för geografi och ekonomisk historia, Umeå Universitet
Department of Geography and Economic History, Umeå University
901 87 Umeå, Sverige/Sweden
Tel: +46 90 786 63 62
Fax: +46 90 786 63 59
Web: http://www.geoekhist.umu.se
E-mail: [email protected]
This work is protected by the Swedish Copyright Legislation (Act 1960:729)
ISBN: 978-91-7601-192-8
ISSN: 1402-5205
© 2014 Cecilia Gustafsson
Cover photos: ©Tom Dowd│Dreamstime.com
Electronic version: http://umu.diva-portal.org/
Printed by: Print & Media
Umeå, Sweden 2014
I owe many thanks to a great number of people for having been able to conduct
this research and write this thesis. It has been a rather long, often joyful but
also occasionally very arduous, journey that began with a phone call in 2005
from the person who would become my main supervisor – Gunnar Malmberg
– asking if I would like to be part of a research proposal on the theme of
migration and health in Nicaragua. So, to begin with, thank you so much,
Gunnar, for giving me that opportunity and for believing in me!!! Especially
seeing as I’d never been to Latin America, didn’t speak Spanish, and had no
experience of working with statistical data… Now, almost ten years later, the
project has finally come to an end! I know that part of your decision to call me,
Gunnar, was based on a conversation with my former graduate thesis
supervisor and subsequent PhD co-supervisor, Aina Tollefsen. Had it not been
for you, Aina, I don’t think Gunnar would’ve made that call, and I also don’t
think I would’ve ventured on this journey at all. So, a great thank you, Aina,
for your encouragement that began during the work on my graduate thesis
and has continued ever since. You two – Gunnar and Aina – have made the
best team of supervisors a PhD candidate could wish for. Thanks ever so much
for your encouragement, support, patience, and for sharing your intellectual
expertise and specific skills with me during countless hours of talks, or handson work at the computer and with the manuscript. You both have also joined
me in my travels to Nicaragua; thank you for the fun times and for the work
we did there together!
Key in this whole process has of course been the Department of Geography
and Economic History at Umeå University. I would like to thank the whole
Department for making this project possible from the start, as well as for the
support over the years. Great thanks to all the Heads – Kerstin Westin, Urban
Lindgren, Dieter Müller, Einar Holm, and Ulf Wiberg – and special thanks to
Kerstin and Dieter for reading and commenting on the first manuscript of this
thesis. Thank you also Rikard Eriksson and Emma Lundholm, for reading and
approving the final version of the thesis. Many thanks to the “TA staff” over
the years – Lotta Brännlund, Erik Bäckström, Ylva Linghult, Fredrik Gärling,
Maria Lindström, and Margit Söderberg, for easing the PhD work process and
teaching. To all my former and current co-doctoral students – thank you so
much for the fun times and the warm support! I’ve really enjoyed getting to
know you all and being part of the PhD group! Special greetings to the
“Monday-lunch group”: Jenny, Erika, Linda, Katarina and Madeleine (I’m
finally one of you now!). And, to Anne Ouma and Erika Sörensson for sharing
an interest in development geography.
Two actors to whom I am deeply indebted and grateful for making this
research possible are the Centre for Demographic and Health Research
(CIDS) and the organization CHICA in León and Cuatro Santos, Nicaragua.
Thank you ever so much for initiating and carrying through the research
together with my colleagues and me. Without your collaboration, a large part
of the research would not have been possible to conduct to begin with!
Quisiera agradecer a todos los que han colaborado con este trabajo en el
CIDS y CHICA! Special thanks to the heads of CIDS, Rodolfo Peña and Eliette
Valadarez, and to Elmer Zelaya at CHICA. And many thanks to all other staff
members at CIDS and CHICA who helped with the study and with other
practical matters –Andrés Herrera, Wilton Pérez, Mariano Salazar, Claudia
Obando Medina, William Hugarto, Margarita Chévez, Maria Mercedes Orozco
Puerto, Aleyda Fuentes, Francisca Trujillo, Maria Teresa Orozco, Doña
Azucena, Marlon Meléndez, Rolando Osejo, Alland Delgado and Ramiro
Bravo, as well as the numerous fieldworkers involved in the survey study.
Thank you also, Yamileth Gutiérrez, for transcribing the interviews. And, to
Mariela Contreras, Uppsala University, for the time we spent together in field
and for letting me use some of your photos in the thesis. I am also deeply
indebted and grateful to the Nicaraguan men and women who participated in
the interviews and survey in this study. Even though many of you are not likely
to be reached by these words, I would like to express my deepest gratitude to
you all, for taking the time and effort to share your experiences with me.
Muchisimas gracias a todos los que participaron en el estudio!!!
Several other researchers at Umeå University have also been supportive
during the research process. Many thanks to the group of Swedish-Nicaraguan
researchers at, or connected to, the Division of Epidemiology – particularly
Ann Öhman, Kjerstin Dahlblom, Gunnar Kullgren, and Ulf Högberg – for
sharing your expertise on Nicaragua, the collaboration between Umeå-León,
and health surveys. Thanks also to Hans Stenlund and Erling Lundevaller for
sharing your statistical knowledge. Also, a big thank you! to Linda Berg,
UCGS, for reading the first manuscript of this thesis, and for providing many
good ideas for how to improve the text.
I would also like to thank the research funder, the Swedish International
Development Cooperation Agency/Department for Research Cooperation
(Sida/Sarec), for the initial grant. Thanks also to JC Kempes Minnes
Stipendiefond for smaller grants over the years.
Last but not least, a great thank you to my beloved family and to my dearest
friends for standing by and cheering me on all these years. I hope my mind
will be a little less occupied with work from now on so that I can dedicate it
more to you…
Cecilia Gustafsson, December 2014, Umeå
Table of Contents
List of figures
List of tables
1. Introduction
La Americana
La Bestia
Points of departure and focus of the thesis
The migration-health nexus
Health geography
The “Western” bias in research on migration and health
The migration-health nexus within the context of social transformations
and social inequalities
Aim and research questions
Framing the study: the research collaboration, and the Health and
Demographic Surveillance Systems in León and Cuatro Santos
Outline of the text
2. Theoretical framework
Geographical and sociological perspectives on health
Putting health into place
A holistic/integrative perspective on health
Social and critical perspectives on health
Embodiment, emotions and health
Stress, health and coping
Health care
A social transformation and relational perspective on migration
Migration, social transformations and development processes
Mobile livelihoods
Translocal geographies and transnational social spaces
The interrelations between migration and health
The migration-health nexus as a bi-directional process
The “globalized” body
Migrant health
Transnational families and health
Recapitulation: a critical framework for analysing the migration-health nexus
3. Materials and methods
A mixed-methods case study
Case study methodology
Mixed-methods research
The fieldwork
Getting to know the field, and holding test interviews
The interview study
The interviewees
The interview situation
Qualitative research approaches and methods of analysis: the biographical
approach and constructivist grounded theory
The two-step survey study
The HDSS in León and Cuatro Santos
Survey step 1: singling out individuals
Survey step 2: construction of sample and questionnaire
The survey procedure
The survey data and statistical analysis
The last fieldtrip: feedback and follow-up
Reflections on conducting mixed-methods research
4. Nicaraguan landscapes: “La vida es dura”
Crucial moments in the past: socio-economic transformations 1520-2006
The colonial era and the post-independence period
The Somoza dynasty and the Sandinista revolution
The Sandinista years and the Contra war
The Conservative era and the return of Daniel Ortega
Living conditions during the fieldwork period
The Ortega administration
The socio-economic situation
Migration patterns
The study settings of León and Cuatro Santos
Introduction to the empirical chapters
The complexity of migration-health relations
Mobile livelihoods, migrant health and translocal lives
Vulnerability, suffering and coping
5. Mobile livelihoods and health dynamics
Prior experiences of migration
Qualitative results: migration biographies and networks
Survey results: migration networks, migration histories and intentions
for future migration
Summarizing comment
Motivations for moving and staying
Survey results: stated motives behind intended moves
Qualitative results: the troubles making a living and striving for a better life
Particular health concerns as motivating factors
Social support, remittances and health
Survey results: help within social networks
Qualitative results on remittances
Who receives remittances? Results of the survey study
Summary and conclusions
6. Health on the move
The journey
Passing through the jungle
“Illegal” border crossings
“Legal” border crossing
Life in the new place
New environments
Working and living conditions
Access to health care and medicine
Returning “home”
Happy returns
Ambivalent returns
“Shameful” return
Results of the survey study: the migrants’ situation abroad
Summary and conclusions
7. Coping with translocal lives
Divided families
Emotional impacts of separation
Changes in family relations
Survey results: migration and self-rated health
Parenting and caring at a distance – tensions and coping strategies
Trying to maintain relations
Making plans
Sending dollars shows care
Contact within transnational social spaces
Summary and conclusions
8. Concluding discussion
Tracing health within the migration process
Migration, health and social transformations in Nicaragua
Complex migration-health relations – the importance of contextualization
and social differences
The embeddedness of health in mobile livelihoods
The importance of social networks and translocal social support for health
The stresses of migration – migrants’ vulnerability and suffering
The health effects of separation and coping strategies
Advantages and disadvantages
Resumen en español
Sammanfattning på svenska
Appendix: Survey questionnaire
List of figures
Figure 1:
Map of Nicaragua and the study settings
p. x
Figure 2:
Study areas in León municipality, 2006
p. 68
Figure 3:
Study areas in urban León, 2006
p. 68
Figure 4:
The HDSS in Cuatro Santos, 2005
p. 69
Figure 5:
The two-step survey
p. 75
Figure 6:
Family members in other places (who)
p. 132
Figure 7:
Family members abroad (country of residence)
p. 133
Figure 8:
Place of birth
p. 135
Figure 9:
Expressed intentions to move
p. 136
Figure 10:
Stated motives behind intensions to move
p. 139
Figure 11:
Perceived social support
p. 175
Figure 12:
Type of help received
p. 177
Figure 13:
Use of money remittances
p. 177
Figure 14:
Origin of money remittances
p. 178
Figure 15:
Sender of money remittances
p. 178
Figure 16:
Received help during illness period
p. 179
Figure 17:
Origin of help during illness period
p. 180
Figure 18:
Provider of help during illness period
p. 180
Figure 19:
Contact with emigrated relatives (frequency)
p. 266
List of tables
Table 1:
The interviewees
p. 56
Table 2:
The study population and sample frame
p. 72
Table 3:
The sample
p. 73
Table 4:
The respondents
p. 77
Table 5:
Variables in the data
p. 79
Table 6:
p. 80
Table 7:
Nicaragua’s modern history; selected indicators
and major events
p. 122
Table 8:
Location of dispersed family members
p. 133
Table 9:
Immigration status of emigrated relatives
p. 134
Table 10:
Migration history
p. 135
Table 11:
Exchanges of help
p. 175
Table 12:
Type of help during illness
p. 181
Table 13:
Logistic regression: “Remittance-receiver”
p. 191
Table 14:
Logistic regression: “Remittance-receiver”
p. 191
Table 15:
Logistic regression: “Remittance-receiver”
p. 192
Table 16:
Logistic regression: “Remittance-receiver”
p. 192
Table 17:
Logistic regression: “Remittance-receiver”
p. 193
Table 18:
Self-rated physical and mental health
p. 256
Table 19:
Logistic regression: “Good self-rated physical health”
p. 257
Table 20:
Logistic regression: “Good self-rated physical health”
p. 258
Table 21:
Logistic regression: “Good self-rated mental health”
p. 259
Table 22:
Way of contact with emigrated relatives
p. 266
Constructivist Grounded Theory
Coordinator of Austria’s development co-operation
(Coordinación de Hermanamientos e Iniciativas
de la Cooperación Austríaca)
Centre for Demographic and Health Research
(Centro de Investigación en Demografía y Salud)
Commission on Social Determinants of Health
Grounded Theory
Human Development Index
Human Development Report
Health and Demographic Surveillance System
Initiative for Heavily Indebted Poor Countries
International Monetary Fund
National Institute for Information and Development
(Institutio Nacional de Información de Desarrollo)
International Organization for Migration
Millennium Development Goal
Mixed-methods (research)
Non-Governmental Organization
Structural Adjustment Programme
United Nations
United Nations Development Programme
World Bank
World Health Organization
Figure 1: Map of Nicaragua and the study settings.
This part includes four chapters that together set the scene for
the empirical study. The first chapter introduces the thesis and
the research topic, the second provides the theoretical
framework and presents previous research, the third describes
the study’s empirical material and methods of analysis, and the
fourth presents the context of the study – Nicaragua and the two
study settings of León and Cuatro Santos.
View of León, Church of el Calvario.
Photo: Otto Dusbaba, Dreamstime.com
Urban centre of San Pedro, Cuatro Santos.
The Cathedral of León.
La Americana
In the documentary film “La Americana”1, a Bolivian woman named Carmen tells
her story of working as an undocumented immigrant in the United States. Carmen
had gone to the US because her daughter Joanna had been injured in a traffic
accident when she was a little girl. High expenses for Joanna’s health care led to
serious debt for Carmen, and she saw no other way out than to go to the US to work.
She left Joanna in the care of her grandmother and travelled via Mexico to the
American border, which she crossed hidden in the back seat of a car. Carmen’s hopes
were to earn a great deal of money to repay the debt and cover her daughter’s present
and long-term medical needs. After working six years in New York, she decided to
return to Bolivia for Joanna’s fifteenth birthday. Upon her return, though, Carmen
soon realized that the money she had earned in the US wouldn’t last long due to the
high medical costs and living expenses. Still, she said she didn’t regret going back,
since both she and her daughter had suffered a great deal emotionally during their
La Bestia
In April 2013, on IOM’s web page2, Niurka Piñeiro told the story of José Luis
Hernandez, a 19-year-old Honduran, who “had lost a leg, an arm and four fingers of
the other hand after falling off of La Bestia, or the Beast, as Central American
migrants aptly name the train that leaves the southern Mexican city of Arriaga and
travels north to Reynosa, just across the border from McAllen, Texas”. José’s goal,
recounts Piñero, was “to help my family build our own house, maybe even buy a car.
I just wanted a better life. And with that dream I left my home; the dream of helping
my family. And here I am a burden to my family”. José believed he had fallen from
the train after falling asleep on the roof, but, as Piñero denotes in her article, “many
other migrants say that if you don’t pay US$100 or more to the members of the
‘maras’ or gangs that hop on and off La Bestia they will push you off the moving
1 By People’s Television, directed by Nicholas Bruckman (2008). See: www.la-americana.com.
2 See the International Organization for Migration’s (IOM) blog, “The Migration Blog: Read all about it”,
12/04/2013: http://weblog.iom.int/beast-turns-dreams-nightmares. The situation for migrants travelling with
La Bestia is also vividly portrayed in the award-winning film “Sin Nombre” (2009) by Cary Joji Fukanaga.
The connections between migration and health that are suggested by Carmen
and José Luis’ stories are examples of what this thesis is about. Their accounts
capture many important dimensions also relevant in this study’s context of
understanding the migration-health nexus: the practice of mobile livelihoods
(migratory lifestyles) for making a living in low-income countries, the
difficulty affording health care and medicine in countries with non-inclusive
health care systems, the importance of social networks for care and money,
the dangers during transit and illegal border crossings, the complications of
living and working without legal documents in a new country, and the
psychological costs of family separation. On the following pages, these aspects
of migration-health relations – and many more – will be explored and
analysed in the context of Nicaragua. Nicaragua is a country where migration
is a predominant feature with deep historical roots. Migrant workers’
remittances have over the years become an increasingly important source of
income for the population, partly used to pay for health care and medicine as
the public sector is unable to provide adequate services for all.
Points of departure and focus of the thesis
Health and migration are intimately linked. Given that migration is an inherently
social and geographical process, and that health and health care are socially and
geographically patterned, this is hardly surprising. Yet much more work needs to be
done to clarify the relationships. (Gatrell & Elliott, 2009: 178)
The migration-health nexus
It is commonly acknowledged that there are intimate linkages between
migration3 and health4. These linkages are relatively well-researched, within
both the medical and social sciences (see e.g. Evans 1987; Carballo & Mboup
2005; Jatrana, Graham & Boyle 2005; Gatrell & Elliott 2009; Schaerström,
Rämgård & Löfman 2011); yet, as I will return to in a moment, far more
research is indeed needed in order to disentangle these intricate relations.
As Gatrell and Elliott (2009) mention in the quote above, migration and
health are social and geographical processes that naturally influence one
another. The act of moving influences all areas of life, including health; and
3 In this thesis migration is defined as moves undertaken by individuals to new places of residence for any
length of time (see Chapter 2 for further discussion). Internal migration implies moves within the borders of a
country, whereas international migration refers to movements across national borders (see e.g. Boyle, Halfacree
& Robinson 1998).
4 In this thesis I apply a holistic, social and critical understanding of health, much related to the World Health
Organization’s (WHO) definition that reads “a state of complete physical, mental and social well-being and not
merely the absence of disease or infirmity” (WHO 2006: 1) (see Chapter 2 for further discussion).
health, as part of life, naturally also influences migration patterns. Thus,
health is also a geographical process. As people move between physical and
social milieus, their health, as well as their access to health care, may be highly
influenced. Moreover, migrants’ family members may also be affected in the
process. Hence, “health […] conditions are powerfully entangled with people’s
trajectories into, within and out of, different spaces and places” (Smith &
Easterlow 2005: 185). These relations between migration and health are at the
core of what I call the migration-health nexus in this thesis.
A common and useful way to conceptualize the migration-health nexus is to
look at it from two sides; that is, to distinguish between, how migration affects
health (M  H) on the one hand, and how health affects migration (H  M)
on the other (Jatrana, Graham & Boyle 2005, with reference to Hull 1979).
One example of the first type of interaction (M  H) is how a person’s or
group’s life and health situation in previous places of residence (e.g. countries,
towns or smaller communities) may influence the person’s/group’s current
health situation at the destination, or influence the health systems in
destination countries (commonly referred to as migrants’ health footprints).
Another example is how the act of moving per se may influence the health
status of migrants (the stresses of migration). A final example of M  H
interactions concerns how the establishment and living conditions in the
destination country may influence the migrant’s health situation and his/her
access to health care (life after migration). Turning to the second side of the
migration-health nexus (H  M), one important example is how the health
situation of migrants may influence their propensity to migrate (denominated
health selectivity in migration). Another example is how health problems may
spur migration in order to reach health care or social support (migration for
care) (ibid; see also e.g. Gatrell & Elliott 2009). As the examples above show,
the migration-health nexus is commonly viewed as bi-directional, in the sense
that migration and health may affect each other. This thesis deals with the bidirectional connections between migration and health, i.e. both the diverse
impacts of migration on health, and the different impacts of health on
migration, in the case of Nicaragua.
The existing literature on the linkages between migration and health can be
divided into studies investigating health in relation to either internal
migration (migration within countries), or international migration (migration
across national borders). According to McKay, Macintyre and Ellaway (2003),
the research on internal migration and health is dominated by studies on
geographical variations in health, with the aim to determine the main
predictor of health outcome (e.g. place of birth or place of residence). Many
studies are also conducted on the effects of moving between areas of different
character (e.g. from rural to urban areas, or from well-off to more deprived
areas). A third area focuses on selective migration, that is, the movement of
“healthy” or “unhealthy” migrants. The research on international migration
and health, on the other hand, is primarily focused on comparing the health
patterns of migrant groups to those of the host population, or to non-migrants
still residing in the sending country (for example studies on mental health and
mortality due to, e.g., cardiovascular disease and cancer). This thesis explores
health in relation to both internal migration (moves within Nicaragua) and
international migration (moves across the border of Nicaragua), and thus
involves both of these vast research fields.
The picture in the literature remains unclear as to the effects of migration on
health, and vice versa. The vast variation in migration patterns, in migrant
groups, and in research design makes it difficult to draw any overall
conclusions. The effects seem to depend on “who is migrating, where they
migrate from, where they migrate to, and what health outcome is measured”
(McKay, Macintyre & Ellaway 2003: 18; see also e.g. Schaerström, Rämgård
& Löfman 2011). In this study, I do not attempt to establish whether or not
migration is beneficial for health, since the effects are so diverse and contextdependent. Instead, this thesis aims to explore and analyse the manifold
relations between migration and health that exist in the case of Nicaragua,
and the surrounding factors that are of importance for the enactment of
these relations. The ambition of this study is therefore to capture the ways in
which different kinds of migration experiences – in terms of, for example,
economic circumstances, household formations, family relations, gender and
immigration status – relate to health during different stages of the migration
Much contemporary migration in Nicaragua is practised as a strategy for
making a living, that is, as part of people’s livelihoods. In order to characterize
these movements I apply the concept of “mobile livelihoods” (e.g. Olwig &
Sørensen 2002), which emphasizes the embeddedness of migration in lives
and livelihoods, and the importance of multiple geographical settings for
making a living5. This concept is closely aligned with recent trends in
migration studies that emphasize the processual and relational nature of
migration. Migration is therefore understood in this thesis as a relational
process that binds together everyday lives across spaces, places and scales,
thus creating “translocal geographies”6 (Brickell & Datta 2011). Based on this
5 Another similar concept is that of “multi-local livelihoods” (Thieme 2008), which also stresses the importance
of migration, other types of mobility, and multiple geographical settings in people’s strategies for making a
living. See Chapter 2 for further discussion, including the distinction between “migration” and “mobility”.
6 The concepts of “translocality”, “translocalism”, and “translocal geographies” highlight the geographies of
everyday lives across spaces, places and scales, without giving preference to any particular situatedness (for
instance the nation, as in “transnationalism”) (Brickell & Datta 2011). “Transnationalism” (see e.g. Vertovec
2009), as well as the idea of “transnational social space” (Faist 2000), are also important conceptualizations in
understanding, I aim to investigate health in relation to the whole process of
migration – health is, thus, “traced” within the migration process. I therefore
make use of the frameworks developed by Haour-Knipe (2013) and
Zimmerman, Kiss and Hossain (2011) for analysing migration-health
relations during the entire migration process (these frameworks, as well as the
theoretical concepts, are described more fully in Chapter 2). The thesis
consequently analyses relations between migration and health in places of
origin, during travel, at the destination and after return. Moreover, the study
examines the situation and consequences for both migrants and family
members to migrants (“left-behinds”7), and for the relation between the two,
within the surrounding local and global context (I thus also follow the call by
Toyota, Yeoh & Nguyen, 2007, to bring the left-behinds into migration
studies). The study thus analyses migration-health relations from both an
individual and a broader structural perspective. By means of this approach, it
has been possible to place the study participants’ accounts of migration and
health into the wider context of local and global socio-economic and political
power relations that structure the migration processes under investigation in
this thesis (cf. Paerregaard 2008).
Health geography
Within geography, migration-health relations – as well as health and health
care in general – are primarily investigated within the sub-fields of medical
and health geography (for overviews of the research field, see Gatrell & Elliott
2009; Brown, McLafferty & Moon 2010; Anthamatten & Hazen 2011;
Schaerström, Rämgård & Löfman 2011). Medical geography draws inspiration
from the medical/epidemiological tradition as well as cultural ecology, and is
mainly concerned with the spatial patterning of disease, illness and medical
care (Mayer 2010; Rosenberg & Wilson 2005). Research within medical
geography that examines migration-health relations has consequently often
followed traditional epidemiological approaches that generally tend to focus
on the analysis of disease and illness among migrants in destination countries,
either at the time of their arrival or over time, in comparison with populations
in the host, or sending, countries (Gushulak & MacPherson 2006a,b).
Health geography – from which this thesis draws the most inspiration –
evolved in the late 1980s, in connection to the “cultural turn” in the social
the thesis, as they emphasize that migration is a process in which migrants interact and identify with multiple
nations, states and/or communities. See Chapter 2 for further discussion.
7 Although I use the term “left-behinds” I would like to stress that these persons, in general, are not passive
“victims” left behind by the “active” migrating family members (for example, passive recipients of the migrants’
remittances), but instead often actively involved in e.g. migration decisions (see e.g. Toyota, Yeoh & Nguyen,
2007) .
sciences when some medical geographers (most importantly Robin Kearns)
argued for a shift in focus within medical geography (for discussions on the
development of the field, see e.g. Kearns 1993; Rosenberg 1998; Kearns &
Moon 2002; Pearce 2003; Rosenberg & Wilson 2005; Moon 2009). Although
health geography is still closely related to medical geography – through the
shared interest in geographical variations in health and health care, for
instance – there are certain contrasts that are important to acknowledge in
order to understand where this thesis is positioned. As Robin Kearns called
for (for instance in his influential article from 1993), much of today’s research
in health geography is concerned with a holistic model of health, which
favours aspects of positive health and wellness (instead of mortality and
morbidity), as well as with a social model of health that acknowledges the
influence of economic, political, cultural and social factors on health.
Furthermore, health geography often takes a more critical stand towards
health issues, stressing aspects of inequalities/inequities in health, and the
importance of power relations in producing and reproducing these
differences. Following these advancements, this thesis critically analyses
migration-health relations in Nicaragua based on a holistic and social
understanding of health.
Within health geography, the key geographical concepts of place and space
have also gained a more prominent position, and the field is now characterized
by a “place awareness”. A relational view on place and space has consequently
been favoured, instead of the geometric space generally applied within
medical geography, concerned with distance and location (e.g. Rosenberg &
Wilson 2005; Moon 2009) (see Chapter 2 for further discussion on the
relational perspective). Through the new place awareness, health geography
now often stresses the importance of the local context, and of relations
between individuals and the local and the wider contexts, for understanding
health (Parr & Butler 1999). In line with the above, this thesis uses a relational
perspective on space and place, and thereby acknowledges the importance of
the relations between the individual and the surrounding social contexts for
understanding and analysing migration-health relations. One ambition of the
thesis is consequently also to “place” the migration-health nexus in the case
of Nicaragua in context.
Besides stressing “place awareness, a critical position, and an engagement
with sociocultural theory” (Moon 2009), health geography is also more
pluralistic than medical geography with regard to research methodology, and
includes not only qualitative and quantitative but also mixed-methods studies
(on qualitative approaches in health geography, see the special issue in The
Professional Geographer 1999, vol. 51, no. 2; see also e.g. Elliott & Gillie 1998;
and Dyck & Dossa 2007). In recent years, there have also been advancements
to incorporate ideas from the “mobilities” turn (e.g. Urry 2000, 2007) into the
field of health geography, with research on other types of mobility than
migration, such as travel/tourism, virtual mobility (e.g. mobilities of
information), and mobilities of care/carers (see for example Gatrell 2011). In
line with recent trends in health geography, this thesis uses a mixed-methods
approach to study the migration-health nexus, combining qualitative and
quantitative data materials. It also investigates health in relation to both
migration of a more permanent sort as well as to other types of mobility (e.g.
temporary migrant work).
The “Western” 8 bias in research on migration and health
Much of the international literature on migration and health has a Western
focus. This is clearly seen in the review by McKay, Macintyre and Ellaway
(2003), mentioned earlier, in which most studies had been conducted within
Europe, North America and Oceania9. Even though the studies often include
diverse immigrant groups – and are “global” in that sense – there are generally
fewer studies that take the South as its actual empirical base, and few are,
furthermore, published in academic journals in English. The research within
medical and health geography is also largely “an Anglo-American affair” (Parr
2004: 247) rather than a global issue. Much of the scientific debate and the
majority of research within the field has the English-speaking, Western world
as its audience, as well as its empirical base (see, e.g., Phillips & Rosenberg
2000; Kearns & Moon 2002; Jatrana, Graham & Boyle 2005; Hunter 2010).
Even though there is a range of quantitative studies conducted within medical
geography that also explore conditions in developing regions (in recent years
studies on HIV/AIDS, for instance) (Gushulak & MacPherson 2006a,b),
generally much less research has been done in Asia, Africa, and South
America. One consequence of this is that many research issues concerning
migration and health in Third World countries remain uninvestigated, or
poorly investigated due to limitations in data (Jatrana, Graham & Boyle 2005;
Konseiga et al. 2009; Adazu et al. 2009). Moreover, due to the diversity of
migration patterns and surrounding circumstances, previous results and
theoretical explanations from Western studies should naturally be explored
and validated in new settings (Hadley 2010; Gushulak & MacPherson
2006a,b). Hence, the knowledge about migration and health in Third World
countries is often sparse and fragmented, and there is still a need to conduct
8 The different terms I use for denominating regions and countries – e.g. “Western”/“North”/“Developed” and
“Third World”/“South”/“Developing” – are mere descriptive terms of global patterns of “development” (see e.g.
Potter et al. 2008; and Chant & McIlwaine 2009). “Development” is defined further on in the text.
9 McKay, Macintyre and Ellaway (2003) only mention one study from the Central American setting (Moss et
al. 1992). This indicates that a relatively small share of studies on migration and health have been done in the
region, and that those conducted often remain inaccessible to the English audience, or might not be digitalized.
empirical studies on migration and health in diverse socio-cultural
environments (Jatrana, Graham & Boyle 2005).
Previous research in the study setting
According to Cabieses et al. (2013), research on migration and health in the
Latin American context is limited, and much of it is also outdated. They
therefore conclude that “[t]here is an urgent need for better understanding of
the living conditions and health of migrant populations in Latin America”, and
that one area that specifically needs to be highlighted is the “the study of
migration as a dynamic and complex process inextricably connected with
broader economic, social, and international factors” (p. 72; my emphasis).
According to the authors, this would lead to stronger theoretical
understandings of the migration-health process, better data, and better
support for policy-makers in the region. This thesis consequently aims to
study the migration-health nexus in Nicaragua as a dynamic and complex
process connected to broader contextual factors.
According to my own review – which I make no claims is exhaustive – the
existing literature on migration and health in Latin America primarily
investigates nine different areas of study that in some way connect migration
and health, for example HIV/AIDS, mental health, mortality patterns, access
to and use of health care, remittances, and vulnerability. Most of the studies
focus on either migrants in South America, migrants of South American
descent, or Mexican migrants. Though some research has been done on
Nicaraguan migration patterns (migration within, from or to Nicaragua),
there are few published accounts of migration-health relations concerning
Nicaraguan migrants. The majority of the existing studies have used
qualitative research approaches (based on a limited number of participants),
and have mostly focused on emigrated Nicaraguans (primarily Nicaraguans
living in or travelling to Costa Rica). Additionally, they have tended to have a
particular health concern in focus (e.g. reproductive health or HIV/AIDS).
Furthermore, there are only a few unpublished reports and undergraduate
theses that have analysed migration data from the Nicaraguan Health and
Demographic Surveillance Systems (HDSS) (see below), on which this study
is partly based. All this points to a further need for more research on
migration-health relations in the Nicaraguan setting.
The migration-health nexus within the context of social
transformations and social inequalities
A complex dynamic of social transformations and social inequalities
surrounds the migration-health nexus, and serves as the foundation and point
of departure for this thesis’ investigation of migration-health relations in the
context of Nicaragua.
The migration patterns we observe in the world today – unquestionably a
pervasive feature of contemporary times10 – are part of globally encompassing
processes of economic, political, social, and cultural character, generally
termed processes of “globalization” (see e.g. Jensen & Tollefsen 2012;
Eriksson 2007; Potter et al. 2008; and Bauman 2000). Migration movements
both produce and are produced by these processes of global
interconnectedness, and take place in a context of vast socio-economic
inequalities and global and local power relations that influence people’s living
conditions and opportunities in life (see e.g. UNDP 2009). These processes
also have historical antecedents; therefore, I believe it is essential to
“historicise the present” (Mirza 2009: 6). I consequently discuss historical
developments with relevance for my research questions in the thesis (Chapter
4), in order to analyse how present migration-health processes are influenced
by practices in the past.
Several scholars also argue that migration should be understood and analysed
as part of broader (global) social transformations (see e.g. Castles 2010;
Davies 2007; Portes 2009). In relation to this the “migration-development
nexus” has received much attention, in both research (see e.g. Geiger & Pécaud
2013; de Haas 2012; Faist, Fauser & Kivisto 2011; Glick Schiller & Faist 2010)
and international forums (e.g. the United Nation’s [UN] High-level Dialogue
on International Migration and Development, and the Global Forum on
Migration and Development)11. Many different aspects of this nexus have been
investigated, and it has been viewed both optimistically and pessimistically
over the years. Research within the field stresses that migration and
“development”12 interrelate in manifold ways. One important research
10 The United Nations (UN) estimate the number of international migrants (i.e. persons moving across national
borders) at over 231 million, equalling about 3% of the world’s population (UN DESA, online database, accessed
2014-02-16). This is not even a third of all internal migrants (persons moving within countries); a number
estimated at 76o million in 2005, according to the UN (UN DESA 2013).
11 See: UN DESA, High-Level Dialogue on International Migration and Development, Internet (accessed 201401-07), and the Global Forum on Migration and Development (GFMD), Internet (accessed 2014-01-07).
12 “Development” – in the meaning human development – is a value-laden concept with many definitions. In
this thesis, I follow the UN’s definition which, in short, states that human development is about “the expansion
of people’s freedoms to live long, healthy and creative lives” (UNDP 2010: 2) (see Chapter 2 for further
question concerns the interactions between levels of development and
migration patterns, and another fundamental question is, in brief, whether or
not migration is beneficial for development. Two issues with relevance for the
migration-health nexus are whether countries lose or gain human capital (e.g.
labour resources) due to migration (commonly discussed as “brain drain” and
“brain gain”, respectively), and whether remittances (the money migrant
workers send home to their family members) work as incitements for
development, and thereby may improve living conditions in the migrant
sending countries (see Chapter 2 for further discussion on migration and
Health must also be placed in relation to socio-economic transformations (see
e.g. Kawachi & Wamala 2007; Gushulak & MacPherson 2006a,b; and Lee &
Collin 2005). Health is commonly regarded as key for achieving and
sustaining development13, and accordingly much research has been conducted
on the bi-directional interconnections between health and development over
the years (see, e.g., Ashtana 2009; Ruger 2003; Phillips & Verhasselt 1994).
One large research area focuses on the epidemiological transition, i.e. the
changes in health patterns said to take place in relation to socio-economic
development14 (McCracken 2009); and many studies investigate development
in relation to, for example, maternal and child health, communicable diseases
(such as HIV/AIDS), and health care systems/provision, as well as health in
relation to aspects such as poverty and structural adjustment policies (the last
of these areas will be discussed further in Chapter 2). Furthermore, since
diseases are no longer confined within national borders due to increasing
mobility and global communication, the character of disease, as well as its
treatment, has become global – indeed, “the body has been globalised”
(Turner 2004: 236). For example, a process of “medical globalization” has
taken place on a world scale since the 1950s, in which Western medicine has
come to dominate over indigenous forms of medicine (often referred to as
13 Health is an important aspect of human development, according to the UN’s definition (see Footnote 12).
Three of the Millennium Development Goals (MDG) – a set of goals adopted by the UN at the Millennium
Summit in 2000 in order to improve the living conditions for all inhabitants of the world – are in fact directly
aimed at improving health issues; i.e. child mortality (MDG4), maternal health (MDG5), and HIV/AIDS,
malaria and tuberculosis (MDG6) (see UN 2013). In the Millennium Declaration (see
http://www.un.org/en/ga/search/view_doc.asp?symbol=A/RES/55/2), the UN also acknowledges that
globalization is key in the process, since its uneven effects, and costs, must change for an inclusive and equitable
world to be realized. See Pérez (2012) on the progress of MDG1 and 4 in León and Cuatro Santos, Nicaragua.
14 It is argued that the health patterns (health standards, disease burdens, mortality causes, etc.) of populations
change as societies develop. According to the original idea of epidemiological transition theory (cf. Omran 1971),
the burden of infectious diseases is said to be more common in developing countries, while chronic health
problems – so-called welfare diseases, e.g. cardiovascular diseases – are more common in developed countries
(McCracken 2009). Criticism of this theory has nevertheless been raised (see e.g. McCracken 2009), and today
there is evidence that developing countries experience a so-called “double burden” of disease, whereby the
population suffers from both infectious and chronic diseases simultaneously (see e.g. Agyei-Mensah & de-Graft
Aikins 2010, for the case of Ghana).
traditional medicine, or traditional medicinal knowledge, TMK). Problems
with over-use (of medicine and treatment) have occasionally been reported in
relation to this. Moreover, the “globalization of the body” has also produced
responses in policy and research regarding migrants. In this context,
migration has primarily been viewed in two ways – with focus on the threats
and risks it entails, or on migrants’ rights and entitlements to health and
health care. The tendency to regard migrants as potential disease-spreaders is
highly connected to what the anthropologist Mary Douglas (1966)
denominated “fear of pollution”, and what post-colonial scholars today
theorize as the “othering” of migrants and their “different” bodies (see e.g.
Ahmed 2000; Sandoval-García 2004). The rights-based approach to health
instead argues for health as a human right, and emphasizes entitlements to
health that are – or should be – equal for all human beings around the world,
including migrants. This “health-for-all” approach has its origins in the
Declaration of Alma-Ata from 1978, and the World Health Organization’s
(WHO) subsequent adoption of the “Health for all” strategy in 198115;
moreover, it was the foundation for WHO’s work with the Commission on
Social Determinants of Health (CSDH), whose final report from 2008
(CSDH/WHO 2008) clearly pointed out the vast inequalities in health that
exist both the global and the local level (that is, between countries/regions of
the world, and between different social groups within countries), which they
explain to a great extent with social factors16 (see also Pearce & Dorling 2009).
In sum, this thesis analyses migration-health relations in the case of
Nicaragua in the context of global socio-economic transformations,
particularly in relation to the debates on migration, development, and health.
The concept of mobile livelihoods is applied in order to emphasize how
migration, and consequently also migration-health relations, are embedded in
people’s strategies for making a living within the “globalized” labour market.
The role remittances play in people’s livelihoods – and whether they have any
possibility of improving living conditions, education, and health – is
scrutinized. Additionally, I look at what role health plays in migration
decisions, and in the development potentials of migration. Furthermore, the
thesis places the analysis of migration and health in Nicaragua in relation to
the globalization of the body, the right-based approach to health, and the view
15 The Declaration of Alma-Ata was adopted by the international community at the International Conference
on Primary Health Care in 1978 (see: http://www.who.int/publications/almaata_declaration_en.pdf).
Preparations before the conference were led by the Director-General of WHO, Halfdan Mahler. Mahler’s vision
of “Health for all by the year 2000” was adopted by the conference, and successively became the leitmotif of
WHO’s work (see also: http://www.who.int/social_determinants/resources/action_sd.pdf; and box 1.1, p. 6,
in http://apps.who.int/iris/bitstream/10665/85761/2/9789240690837_eng.pdf).
16 The Commission furthermore stressed that the widespread, global inequalities in health are avoidable, and
consequently a matter of social justice. Therefore, they preferably speak of inequities in health, i.e. inequalities
that are avoidable, unjust, or unfair (CSDH/WHO 2008).
of migrants as health threats, for example by investigating international
migrants’ health problems and access to health care, and by scrutinizing the
“othering” of Nicaraguan migrants.
Aim and research questions
The overall aim of this thesis is to critically explore and analyse relations
between migration and health, what I call the migration-health nexus, in the
contemporary Nicaraguan context. Based on a mixed-methods approach and
fieldwork in León and Cuatro Santos, Nicaragua, the thesis aims to provide
answers to the following research questions (chapter(s) in which a question is
primarily addressed in parentheses):
1) How can the dynamics between migration and health be understood in
the Nicaraguan context? (Chapters 4 and 5)
In what ways are Nicaraguan migration patterns and health trends
related to past and present socio-economic transformations?
How are these migration patterns and health trends related to
social differentiation (based on e.g. gender, ethnicity, class,
and immigration status)?
2) In what ways do health issues influence Nicaraguan men’s and
women’s migration strategies? (Chapter 5)
How are health concerns integrated into motives for
migration, staying and returning?
For what reasons are remittances sent, and how are
health issues related to these remittance patterns?
3) In what ways does migration affect men’s and women’s lives and health
situations in the different places and during the different phases
involved in the migration process? (Chapters 6 and 7)
How do different kinds of migration experiences affect the migrant,
(e.g. socially, economically, healthwise, and emotionally)?
How do different kinds of migration experiences affect the family
members of migrants (the “left-behinds”)?
Framing the study: the research collaboration, and the
Health and Demographic Surveillance Systems in León and
Cuatro Santos
This study was initiated within the framework of a long, ongoing research
collaboration between Umeå University in Sweden and León University
(UNAN-León) in Nicaragua17. As part of this collaboration, two Demographic
and Health Surveillance Systems (HDSS) were set up during the 1990s and
the 2000s; the first in the municipality of León, which is situated on the Pacific
Coast and harbours the second largest town in Nicaragua (León) (see Figure
1, p. x). The HDSS-León was initiated at the end of the 1990s, and is managed
by the Centre for Demographic and Health Research (CIDS). The second
surveillance site was initiated at the beginning of the 2000s in Cuatro
Santos18, an area in the northern part of Chinandega consisting of four
predominantly rural municipalities (San Pedro, San Francisco, Cinco Pinos
and Santo Tomás) (see Figure 1). The HDSS in Cuatro Santos is managed by
the organization CHICA19, in close collaboration with CIDS. Like other HDSS
sites20, the HDSS in León and Cuatro Santos regularly (often annually) gather
population-based data, with the ambition to monitor demographic processes
in the population, and to conduct epidemiological and public health research.
In Nicaragua, studies have been made on, for example, reproductive health
(Zelaya Blandón 1999), child health (Pérez 2012), intra-familiar violence
(Salazar Torres 2011; Ellsberg 2000; Valladares Cordoza 2005), and mental
health (Obando Medina 2011; Herrera Rodríguez 2006; Caldera Aburto
2004). (See Chapter 3 for a more detailed account of the contents of the HDSS
and how they developed).
The empirical material in this thesis, consisting of survey and interview data,
was gathered through fieldwork in the above-mentioned settings between the
years 2006 and 2008, with a follow-up visit in 2013 (see Chapter 3). The
quantitative part of the study (and to some extent the qualitative part as well)
was carried out within the frames of the two surveillance systems in León and
17 The Division of Epidemiology and Public Health Sciences at Umeå University played an important role in
the collaboration with UNAN-León, Nicaragua from the start in the 1980s, together with other departments at
Umeå University. The Department of Women’s and Children’s Health at Uppsala University was also an
important actor in the process. The largest funder of the research collaboration was the Swedish International
Development Cooperation Agency (SIDA).
18 Cuatro Santos is a figurative name for four municipalities in the department of Chinandega that all have
“San” or “Santo” in their names: San Pedro del Norte, San Francisco del Norte, San Juan de Cinco Pinos, and
Santo Tomás del Norte (in short: San Pedro, San Francisco, Cinco Pinos and Santo Tomás).
19 CHICA also coordinates and runs development projects in the area of Cuatro Santos.
20 There is a global network of HDSS sites, the INDEPTH Network, all situated in low- and middle-income
countries (the INDEPTH Network homepage, accessed 2013-04-26). At the time of the fieldwork the HDSS in
León was part of this network, but today it is no longer an INDEPTH site. The HDSS in Cuatro Santos has never
been a part of the network.
Cuatro Santos, in close collaboration with the organizations CIDS and CHICA.
This gave me access to a unique set of data on migration events, which is very
rare in low-income countries in the South. I also got to work in a research
environment with expertise on public health issues, and with many years of
experience conducting survey studies in the setting. Moreover, through
including both study settings I could explore the same issues in two rather
distinct places (e.g. rural and urban), which of course broadened the analytical
base. Nevertheless, there were of course also constraints involved with the
research approach, which will be discussed next.
Even though I had access to a unique set of data on migration events, a
limitation involved with conducting the study within the frames of the HDSS
in León and Cuatro Santos was that the study populations were pre-defined
and that data had been collected by others than myself over the years, which
meant that I had no control over previous work with sample selection and data
collection. Still, the HDSS are both well-designed and well-managed (see
Chapter 3). A perhaps greater limitation was that, even though the HDSS had
collected data on migration events for many years, the sizes of groups with
different migration characteristics within the populations – from which the
sample for this study was drawn – were sometimes small. Those categorized
as In-migrants (persons who had moved into the study areas) were
particularly few, and, in the analysis (i.e. the regression analysis), this group
proved to be too small to produce sound results. Another limitation involved
with using the HDSS was that persons who had moved out of the surveyed
areas (so-called “out-migrants”) could not be included in our survey, which
meant that I could only survey the out-migrants’ family members who still
resided in the HDSS areas. Moreover, people with illnesses, who were
identified in the first step of our survey, also numbered rather few for
particular migration categories (e.g. In-migrants and Left-behinds), even in
the León-setting. It was thus not possible to select the sample randomly, but
this was solved through selective sampling and applying appropriate weights
for each sample group in the statistical analysis. Furthermore, through
applying a mixed-methods approach, the quantitative study could also be
complemented with qualitative data, just as the qualitative study could be
enriched by the statistical information provided by the survey study. (See
Chapter 3 for further details on the methods applied).
Outline of the text
The next chapter in this introductory section provides the theoretical
framework that informs the analysis. The third chapter describes and
discusses the empirical material the study is buildt on, as well as the ways in
which it was gathered and analysed. The fourth chapter introduces the
contextual setting, Nicaragua and the two study settings León and Cuatro
Santos, in which the migration-health relations in this study are placed.
Thereafter follows the second part of the thesis, including three empirical
chapters presenting the findings of the study. Chapter 5 is dedicated to the
practice of mobile livelihoods and their relations to health, and presents for
instance health-related motivations for moving and staying. Chapter 6 is
concerned with the consequences of migration, and implications on health,
for the migrant. Chapter 7, the last empirical chapter, looks at the
consequences of migration on social relations, and thus focuses on the
relationships between migrants and left-behinds. Lastly, in the concluding
part of the thesis (Chapter 8), the findings of the study are summarized and
discussed in relation to the theoretical framework and previous research.
Mountain view, Cuatro Santos.
Alfombras – sawdust carpets with religious motives.
Easter 2007, Subtiava, León.
Theoretical framework
This chapter outlines the theoretical framework used in the thesis for
analysing the relations between migration and health in the Nicaraguan case.
The chapter first presents theoretical ideas and concepts concerning health
and migration, respectively, that I have made use of for analysing the
migration-health nexus. Thereafter follows an outline of some fundamental
understandings of this nexus, and a discussion of particular issues concerning
migration-health relations, such as migrant health, and transnational
families. The framework for analysing the migration-health nexus is
summarized at the end of the chapter.
Geographical and sociological perspectives on health
This thesis analyses health from a geographical and sociological perspective.
Both health geography and the sociology of health and illness are based on a
holistic (integrative), social and critical understanding of health (see below).
Health geography is also characterized by a “place awareness”, and gives
prominence to a relational view on space and place.
Putting health into place
Starting in the 1990s, a process of “putting health into place” occurred within
medical geography, which led to a reinvention of the discipline. This was a
highly necessary process, according to Kearns and Gesler (1998), since
“diseases, service delivery systems, and health policies are socially produced,
constructed, and transmitted” (ibid. p. 5; my emphasis). In the process of
putting health into place the unproblematic, geometric space generally
applied within medical geography – concerned with distance and location –
was criticized for conceiving space “as a mere blank surface on which [one]
uncritically [could] […] map medical and ‘deviant’ subjects” (Parr & Butler
1999: 11). The concept of place was subsequently more greatly acknowledged,
with the result that other spatialities, such as the body, received somewhat less
attention. Parr and Butler (1999) nevertheless highlight the importance of the
new place awareness: “[t]he retheorisation of place in medical geography as a
complex material, sociological, experiential and philosophical phenomena is
crucial to thinking through how the local is involved in the making of and
experience of different mind and body states (through place-based
understandings of health, illness and the body, as well as appreciating the
wider spaces of more structural contexts and responses to such phenomena)”
(ibid. p. 11).
According to the relational perspective on space and place, space is not seen
merely as a map surface on which places are located and things take place, or
as something “outside of place”, something “out there”, or “up there” (Massey
2005: 185). Instead, space is viewed as where the social is constructed, as the
product of social relations, as “our constitutive interrelatedness” (Massey
2005: 195). Put differently, space is the sum of the interactions and
interrelations between heterogeneous existences, or multiple trajectories,
which interrelate in a continuous and unending process (Massey 1994, 2005).
In the words of Doreen Massey (2005: 61), space is “the sphere of the
continuous production and reconfiguration of heterogeneity in all its forms –
diversity, subordination, conflicting interests”. Space thus also contains
material practices of power; the spatial may, hence, be seen as “a cartography
of power” (ibid. p. 85). From the relational perspective, places are,
furthermore, not regarded “as points or areas on maps, but as integrations of
space and time; as spatio-temporal events” (ibid. p. 130; italics in original).
Places are thus no more concrete, grounded or bounded than space is; nor are
they where “real life” goes on. Instead, places are open (thus connected to the
wider setting), and made up of a collection of all the spatial interrelations, as
well as the non-relations (the exclusions), that go on at a particular location,
at a particular point in time. Place is thus understood as “a moment within
power-geometries, as a particular constellation within the wider topographies
of space” (ibid. p. 131).
Relational thinking thus stresses how spaces and places emerge through
connections with other spaces and places on multiple scales, in contrast to
seeing spaces as static areas and places as fixed centres of meaning (Andrews
et al. 2012). Through applying the relational perspective on place and space
within health geography, the importance of the relations between individuals,
the local contexts and wider, structural contexts for health is thus emphasized
(Parr & Butler 1999). This view is foundational in this thesis, and has
influenced me to look at both migration and health, and the relations between
the two, from a relational perspective.
A holistic/integrative perspective on health
The holistic, or integrative, view of health stressed within health geography, is
also emphasized by the WHO, which defines health as “a state of complete
physical, mental and social well-being and not merely the absence of disease
or infirmity” (WHO 2006: 1). This definition has nonetheless been criticized
since coming into use in the 1940s, for example for its vagueness and
broadness (Callahan 2012). In my opinion it is still useful, since it shares
common ground with the “biopsychosocial”21 approach to health, which
extends the biomedical model of health that generally connects ill-health to
specific diseases and their pathological processes (White 2005). The
biopsychosocial approach stresses the importance of an integrated, or holistic,
approach to health and illness, and considers other factors than simply
biological processes for understanding health and illness, such as the social
context and psychological well-being (e.g. thoughts and emotions, which are
believed to be important both for health and for the ability to cope with illhealth and other difficult situations in life). Hence, the biopsychosocial
approach “incorporates thoughts, feelings, behavior, their social context, and
their interactions with both physiology and pathophysiology” (White 2005:
preface). Closely related to the biopsychosocial model of health is the concept
of “mind/body health”. Within mind/body medicine, the “Cartesian dualism”
– i.e. the doctrine that separated mind/soul and body – is questioned, and
instead the unity of the body is emphasized (Dreher 2004). The dynamics that
make physical and mental health influence each other are a central theme
within the field. An important study area, highly relevant in this thesis,
concerns the connections between stress and physical ill-health (see e.g.
Vitetta et al. 2005). Since migration is a well-known stressor (see further on
in the text), this research field is crucial in the analysis.
As health geography espouses a holistic understanding of health, it thus
favours other aspects of health than mortality and morbidity, such as positive
health and wellness (e.g. Rosenberg & Wilson 2005; Moon 2009). The holistic
view is also stressed within the sociology of health and illness. Within this field
health is seen as a complex phenomenon, composed not only of an objective,
pathological dimension, but also shaped by subjective and social factors (e.g.
social class, gender and ethnicity) (Wainright 2008). Health is furthermore
seen as influenced by processes on the micro, meso and macro levels. Hence,
the individual and his/her biological, cognitive, and emotional characteristics
are placed within the broader framework of social networks, social relations,
21 The “bio” (in biopsychosocial) stands for the biological, e.g. tissue changes; the “psycho” relates to the
psychological, e.g. personal growth and development; and the “social” concerns the social aspects, for example
a person’s current life situation (Shorter 2005; referred by White 2005).
social integration and social capital, the respective contents of which – for
example social ties and social support – all influence health. The meso and the
micro levels are in turn seen as influenced by relations at the macro level – the
social-structural conditions of society, for instance social and economic
inequality (Turner 2004). Based on these understandings, this thesis analyses
different aspects of health in relation to migration, and includes different
levels and scales in the analysis.
Social and critical perspectives on health
Both the biopsychosocial approach to health and mind/body medicine
acknowledge the importance of the social fabric for health (Dreher 2004); yet
it is within social medicine, public health and epidemiology that the most
extensive research has been conducted on the so-called social determinants of
health and illness – for example, the role material deprivation plays in shaping
health outcomes (see e.g. Turner 2004; Wainright 2008). The literature
shows, perhaps not surprisingly, that conditions such as poverty,
unemployment, low wages, inadequate housing, dangerous working
conditions, poor diet, insufficient sanitation, and poor environments have
negative effects on health22. Health geography also shares this social
understanding of health and thus acknowledges the importance of, for
example, economic, political, cultural and social factors for health. Moreover,
health geography also looks at health from a critical standpoint, stressing the
importance of power relations in producing and reproducing
inequalities/inequities in health (e.g. Rosenberg & Wilson 2005; Moon 2009).
The sociology of health and illness also emphasizes social and cultural factors
for health, and is tightly connected to the social determinant perspective on
health. The field also acknowledges that power relations, as inherent to all
social activities, also influences health outcomes; power is, thus, regarded as
“a key social-structural factor in health” (Freund, McGuire & Podhurst 2003:
x). Health, disease and illness are thus understood as not only related to
biophysical changes but also influenced and shaped by the wider socioeconomic context, power relations and social inequalities (Nettleton &
Gustafsson 2002). The emphasis the perspective of social determinants of
health places on the social fabric is very useful in this study, since for example
unemployment, low wages, and poverty in the study context of Nicaragua are
widespread and problematic (see Chapter 4). The critical perspective on
health, proposed by health geographers and sociologists, is also of importance
in the present study due to the inequalities in migration and health processes.
22 Consequently, the field stresses the necessity of social development and preventive medicine for
improvements in population health to be realized. This focus has been criticized, primarily for placing too much
emphasis on the structural forces of society and hence forgetting the individual’s role and responsibility in
shaping health outcomes.
The intersectional perspective emphasizes – in line with the above – the
importance of social differences, or stratifications, for understanding societal
processes (e.g. health). Gerry Veenstra (2013: 16) explains intersectionality as
a “framework for conceptualizing the nature of relations of power pertaining
to racism, sexism, classism and heterosexism in modern societies”.
Intersectionality is thus useful in understanding the complex nature of
inequalities – for example in health – in which power relations along the lines
of race, gender, class and sexuality, for instance, are intrinsically entwined.
Similarly, Nira Yuval-Davis (2006: 199) describes intersectionality as a
perspective on “differential positionings in terms of class, race and ethnicity,
gender and sexuality, ability, stage in the life cycle and other social divisions”,
which, when interlinked, “tend to create, in specific historical situations,
hierarchies of differential access to a variety of resources – economic, political
and cultural” (thus creating a “matrix of domination”, to use the words of Hill
Collins 2000; as referred in Veenstra 2013). Hence, when analysing health,
for example, according to the intersectional perspective it is necessary to look
at how different axes of inequality (in terms of, e.g., class, gender and
ethnicity/“race”) interact and create hierarchies that influence the individual’s
specific situation23. Immigration status (legal/irregular) may additionally be
added to the social differences of importance when analysing health in
relation to migration. The intersectional perspective provides a useful
analytical lens for examining inequalities in health in relation to migration in
the Nicaraguan case.
At the core of the perspective of social determinants of health also lies the
intrinsic dynamic between health and development (see Footnote 12 and
below for definition). Health is integral to development processes, and much
research has been conducted on the connections between the two over the
years (see, e.g., Ashtana 2009; Ruger 2003; Phillips & Verhasselt 1994). One
study area scrutinizes the health effects of the structural adjustment
programmes (SAPs) of the 1980s and 90s, which were imposed on many
developing countries (including Nicaragua) by the World Bank (WB) and the
International Monetary Fund (IMF) with the aim to stabilize the economies of
highly indebted countries through neo-liberal economic policies (see e.g.
Potter et al. 2008; Chant & McIlwaine 2009). The SAPs entailed massive
restructurings of the economies in many Third World countries, including
Nicaragua, with great implications on people’s living conditions. Studies have
shown that these programmes generally had profound consequences on the
health of the populations in these countries because of the cutbacks in social
23 As, for example, in Khan et al.’s (2007) study on health care experiences of patients from different ethnocultural groups in Canada, in which it was concluded that health and well-being were influenced by “the
complex issues at the intersections of gender, race, class, and culture” (ibid. p. 231), which the authors
furthermore argued were shaped by history.
spending, for example in the health care sector (e.g. Ashtana 2009; Ruger
2003; Phillips & Verhasselt 1994; and Wainright 2008). Even though changes
have been made to improve the programmes, and some countries have
cancelled parts of their debt (e.g. Nicaragua; see Chapter 4), the effects of
structural adjustments still linger. Many Third World countries will, for
example, probably not realize the health-related UN Millennium
Development Goals (MDG) (see Footnote 13), partly as a result of these
cutbacks in public services (Wainright 2008; see also Vargas 2006, in relation
to Nicaragua)24. Another example of relations between health and
development concerns “medical globalization”. This process began during a
period when modernization theories25 flourished. The faith in “modern”
medicine for eradicating diseases (such as smallpox) in developing countries
was great at this point in time, and Western (occidental) medicine and health
care were consequently spread globally, and people were often encouraged to
use the products of modernity. Consequently, in many parts of the world today
Western medicine is dominant over indigenous forms of medicine (traditional
medicine, TMK), and it is also occasionally over-used (Turner 2004;
Wainright 2008). The relation between health and development is at centre
stage in the case of Nicaragua; for example, in relation to the health effects of
the structural adjustment programmes, and the progress towards meeting the
Millennium Development Goals. Nicaragua has also been affected by the
process of medical globalization.
Social capital and social citizenship
The idea of “social capital”, and its relation to health, has received much
attention in research26. One definition understands social capital as “the social
investments of individuals in society in terms of their membership in formal
and informal groups, networks, and institutions” (Turner 2004: 13). Social
capital theory is relevant in this thesis, since it regards social relations as
24 WHO’s work with the Commission on Social Determinants of Health (CSDH), mentioned in Chapter 1, is a
good example of how the social determinant perspective is integral in the work to achieve human development.
As mentioned, the Commission stressed the role of social factors in health (CSDH/WHO 2008; see also Pearce
& Dorling 2009).
25 The perspective usually referred to as Modernization Theory (or Modernism) was largely based on the
European experience of development, and influenced by evolutionary and diffusionist theory, arguing that
development is “a positive and irreversible process through which all societies eventually pass” (Chant &
McIlwaine 2009: 27). Development was thus regarded by some as “all about transforming ‘traditional’ countries
into modern, westernized nations” (Potter et al. 2008: 5; italics in original).
26 Different meanings have been assigned to the concept of social capital over the years, by scholars in various
academic disciplines. It was first used within economics to explain the importance of economic investments in
education, training and welfare, in order to produce human capital (Turner 2004). In sociology, the concept
originates from Émile Durkheim who, at the end of the 19th century, found that rapid social changes and social
disruption caused negative effects on health (i.e. higher levels of suicide) (1966 [1897]). Later, James Coleman
(e.g. 1988) and Robert Putnam (e.g. 1993, 2000) made important contributions to the concept’s further
development (Turner 2004).
fundamental to health, and because it highlights specific aspects of the social
fabric that are especially important for health (social integration and social
inclusion, which according to the theory are created through people’s social
investments, i.e. social ties and social support). Moreover, social capital theory
has also provided new insight into the research on inequality and health.
According to Richard Wilkinson (see e.g. Wilkinson 1996; and Wilkinson &
Pickett, 2009), more egalitarian societies have higher levels of social cohesion,
lower levels of social disruption and lower levels of individual stress, which
result in better health and longer, more satisfying lives for their citizens. Other
scholars similarly argue that income inequality leads to low trust in society,
which negatively affects the social environment (social cohesion), and in turn
also individual health (Turner 2004, with reference to Kawachi et al. 1999;
Antonucci et al. 2003). More critical voices (e.g. Coburn 2000, 2004) argue
that the cause behind income and health inequalities is neo-liberalism, as
social inequality has increased and social cohesion has decreased during the
neo-liberal era27. Coburn’s idea is interesting because it relates people’s
subjective experiences of health and illness to macro changes in the political
environment. In this thesis, the theoretical idea of social capital has provided
an operationalization of particular social aspects (e.g. social ties and support)
that may influence migration-health relations.
Social capital theory thus proposes that social relations are crucial for health.
However, health is also determined by power structures within society, and is
consequently related to the access to and control over society’s resources (such
as material, cultural, and social resources) (Turner 2004). The concept of
“social citizenship” has been developed in relation to this, concerning the
rights and entitlements citizens enjoy – either because they have earned it (for
example through work or other social activities) or because governments
acknowledge that they have needs which could be satisfied collectively (Turner
2004). The social rights of citizenship are important in order to overcome
disparities in health. In welfare states, such as Sweden, health is often
regarded as a social right; this is why individual health standards in such
countries are often high, and the differences in health between individuals
small, according to Turner (ibid.). In Turner’s words (2004: 8), “health can be
27 According to Coburn, the nation-state has lost its authoritarian position due to neo-liberalism, with the result
that social inequality has increased and social cohesion has decreased, leading to lower levels of social trust and
self-esteem, as well as a lack of social support – in short, diminishing social capital – which affects health
negatively. Some scholars, e.g. Wilkinson (2000), have nevertheless criticized Coburn; while others, e.g. Tarlov
(2000), find his argument plausible. In line with this, contemporary social capital theory proposes that rapid
social and economic changes – such as the fall of Communism, civil wars and ethnic conflicts, the globalization
of the world economy, and neo-conservative economic policies – have produced social isolation and social
disorganisation, resulting in poor health (increasing morbidity and mortality) in many societies around the
globe. In this way, labour migration from south to north caused by unregulated global economic markets can
be understood as causing social disintegration, and possibly also affecting health standards negatively (Turner
seen as the unintended consequence of the social rights of citizenship, and not
necessarily an intended outcome of medical interventions”. There is evidence
that higher levels of social cohesion lead to more egalitarian patterns of
political participation, which in turn are associated with a better provision of
health services and distribution of medical information to the public (Turner
2004, with reference to Berkman & Kawachi 2000). The concept of social
citizenship, and its importance for health, is relevant in the context of this
study, where people’s access to resources is highly uneven (see Chapter 4). The
analysis of health issues in the case of Nicaragua must consequently be related
to these inequalities.
Embodiment, emotions and health
In order to fully understand how the relation between the social (society) and
health is played out, theories on the body, “embodiment”, and emotions are
also central in this thesis. As Nettleton and Watson (1998: 1) write:
“Everything we do, we do with our bodies […]. Every aspect of our lives is
therefore embodied”. This field thus argues that people’s experiences are
embedded within the body – basically, “the human being is an embodied
social agent” (ibid. p. 9). Embodiment, thus, also has a spatial dimension. In
the words of philosopher/sociologist Henri Lefebvre (1991: 162), “it is by
means of the body that space is perceived, lived – and produced”. It is thus
through our bodies that we perceive, assess, adapt and design space
(Kenworthy Teather 1999); in other words, it is through our bodies that we
live our lives, have relations with other people, find our way and experience
the world. The concept of embodiment takes the idea of the relation between
the social and health a step further than other social perspectives on health.
The embodiment of people’s experiences (of the social) might then perhaps
explain more profoundly why, for example, social isolation and income
inequality cause ill-health – the experience of it (isolation, inequality) is
embodied, that is, “brought into” the body, and thereby affects health.
The unity of body and mind that is integral to the idea of embodiment is also
stressed by emotion sociologists28 (see e.g. James & Gabe 1996; Williams &
Bendelow 1996; Lupton 1998; and Barbalet 2002). However, through
including emotions (i.e. mental states, strong feelings) it is possible to explore
even more deeply how embodiment “works” in practice, and how social
structures in reality affect people’s health. According to Deborah Lupton
(1998), emotions are essential in how people experience their body,
28 Emotions have made a rather late entry into geographical research, and the field of emotional geographies
has mostly focused on the phenomenology of emotion, i.e. the experiential aspect of emotional phenomena (e.g.
emotional responses and attachments to particular places) (Smith et al. 2009; Davidson, Bondi & Smith 2005;
and Anderson & Smith 2001). Therefore, I rely more on sociological research on emotions in this thesis.
subjectivity, and relationships with others, and are therefore central in
understanding embodiment. Emotions can be conceptualized in different
ways29. Social psychologist Parrott (2001) proposes a complex system of
human emotions, all believed to stem from a few primary emotions (love, joy,
surprise, anger, sadness, and fear). In this thesis, I have used Parrott’s
classification to analyse the emotions the interviewees expressed during our
talks. I believe this to be necessary since our emotions – both those we feel “in
reality” and those we express to others – may influence our health, as well as
our ability to cope with difficulties in life (see below).
Since emotions are naturally embodied, it can be argued that emotions and
emotional experiences are the most powerful factors determining health,
regardless of whether the explanation is seen as biological, cultural or social
(Williams & Bendelow 1996). Due to this, emotion sociologists argue that
emotions should be regarded as the mediator – the “glue” – between micro
and macro relations; between the individual body and the social structure; for,
as Williams and Bendelow (1996: 46) write, “the body is in the mind, society
is in the body, and the body is in society”. Emotions are thus central in
understanding the dynamic between society, the body, and health. Instead of
simply acknowledging that social structures – e.g. social relations, social
support, and social status – affect health (as for instance social psychologists
or health sociologists do), emotion sociologists state that emotions are key in
this process, and thereby more profoundly explain how social structures can
actually affect people’s health. There is in fact a bulk of predominantly
psychological and social psychological research that explores the relations
between emotions and health, for example between emotions, stress and
health (for an overview see Folkman 2011; see also, e.g., Part VII in Lewis,
Haviland-Jones & Feldman Barrett 2008; Pennebaker 1995; Lazarus 2006),
and there is evidence that positive emotions are often better for one’s health
than negative ones (see Chapter 6 in Greco & Stenner 2008). Another relevant
concept in this thesis is “suffering”, as it was commonly expressed by the
interviewees as well as by Nicaraguans in general who I came across during
the fieldwork. Suffering not only embraces “negative” emotional experiences,
for example sadness, anxiety and distress (Wilkinson 2005), but also refers to
many other aspects (physical, psychological, social, economic, political, and
cultural) and is thus a holistic concept that can be said to capture “the
vulnerability of lived experience” (Wilkinson 2005, with reference to Turner
& Rojak 2001). Moreover, entire populations can also experience social
29 There are many theoretical views on human emotions (for an overview, see Lewis, Haviland-Jones &
Feldman Barrett 2008), and throughout the years scholars have also presented several different
conceptualizations of emotions (see, e.g., Turner 2007). Paul Ekman (e.g. 1999, 1992a,b) has proposed a
conceptualization of six basic emotions (anger, disgust, fear, sadness, happiness and surprise), which he argues
are general to all human beings, and which he has characterized as either positive or negative.
suffering, for example when a nation is under collective stress due to extensive
migration (Helman 2007).
Stress, health and coping
The relations between stress, health and coping are other important issues in
this thesis. Stress, that is, demands that require behavioural changes – for
example life events (e.g. migration), chronic strains (e.g. poverty), and daily
hassles (e.g. traffic) (Thoits 1995, 2010) – may have many negative health
effects (see e.g. Williams et al. 2003; Weiss & Lonnquist 2000; Helman 2007).
Stress furthermore stimulates “coping” processes (Thoits 1995); coping thus
takes place when the individual makes the appraisal that something relevant
to him or her is being harmed or threatened, or has been lost, in a situation
that feels difficult to handle (Folkman & Moskowitz 2004; Lazarus & Folkman
1984; Antonovsky 1979). Individuals’ capabilities to cope vary depending on
their access to “coping resources” (for example, self-esteem, social support,
and sense of control or “mastery over life”; see Thoist 1995, 2010; cf.
Antonovsky 1979) – and the employed “coping strategies” (attempts to
manage the situation) (Thoits 1995). The coping resource “social support”
(including, e.g., emotional support) can be important for health, since it
functions as a “social fund” to draw from when handling stressors (ibid.).
Research shows that coping strategies can be more or less beneficial for
health. There is also evidence that religion plays an important role in the stress
process (Folkman & Moskowitz 2004). The understanding of the connections
between coping and psychological, physiological and behavioural outcomes is
still not clear, in part because of the complexities surrounding stress
The framework of stress and coping is relevant in this thesis, as both
underemployment and poverty) and migration patterns indicate that many
may be living under stress, which may set off coping processes and produce
health effects. In the thesis I have also investigated various types of social
support (for example, perceived and received emotional and economic
support) in the analysis of migration-health relations, which is also why this
theoretical foundation is important.
Health care
The model of medical care applied in a country generally has a large impact
on the provision and utilization of health care services. Two common models
are the collectivist/public model, which aims for universal access to health
care, and is structured around the principle of need rather than the ability to
pay, and the anti-collectivist/private model, which gives precedence to the
market and is funded by private health insurance, or by fees paid by the user
at the point of use (Gatrell & Elliott 2009). The localization of health care
services, and people’s distance to them, are important issues in relation to the
access to and use of health services. The question of distance as a constraint
to people’s utilization patterns has indeed received a great deal of attention in
research; there is evidence of a clear “distance decay” relationship between
physical distance to health services and health-seeking behaviour, meaning
that people living farther from health services seek health care more seldom
(Gatrell & Elliott 2009). The reasons behind this may be time-space
constraints, and costs in terms of time and travel, that deter people from
seeking care. However, it is not only physical distance that is important for
people’s use of health care; social and cultural factors are also important – for
example, affordability (that is, being able to afford the costs, not only of travel
but also of the health care itself), cultural barriers (e.g. appropriateness,
language), and social marginalization (e.g. of the homeless) (ibid; see also
Curtis 2004). A distance decay effect also seems to exist in developing
countries (e.g. Muller et al. 1998, referred to by Gatrell & Elliott 2009; and
Feikin et al. 2009, referred to by Anthamatten & Hazen 2011). Yet, a more
important question in these regions is perhaps whether there is an adequate
overall level of health care delivery at all, particularly primary care, and that
this care is not only geographically concentrated to urban areas (Gatrell &
Elliott 2009).
However, health care is not only about provision, access and use; it is an
integral part of society, and thus a political, economic and cultural concern
(Conradson 2005). The question of who cares for whom therefore largely
reflects power relations in society. The analysis of health care thus also
concerns issues of gender, ethnicity/“race” and class (e.g. Lawson 2007). For
example, the greatest burden of caring is often shouldered by women and by
immigrant groups (particularly immigrant women), and care work is often
devalued. Furthermore, as health care is often characterized by inequalities
(as mentioned above), it is also a question of social justice. Neo-liberal
economic policies and restructuring (e.g. in the form of structural adjustment
programmes) have led to a more privatized health care sector in many
countries (as mentioned), causing greater social inequalities in the access to
care (e.g. Lawson 2007).
The issue of health care is central in this thesis, since it has great implications
on people’s possibilities to reach and afford care. The health care situation in
Nicaragua will therefore be described in Chapter 4.
A social transformation and relational perspective on
As mentioned in the introduction, this study analyses health in relation to
both moves within the borders of Nicaragua, and moves between Nicaragua
and other countries. The thesis consequently requires a theoretical framework
that includes theories on both internal and international migration – of which
there is a wide range (see overview in Boyle, Halfacree & Robinson 1998; and
for international migration see Castles, de Haas & Miller 2013). However, as
several scholars point out the need to integrate migration theory with social
theory, I understand migration as part of social transformations and
development processes, and therefore apply the concept of “mobile
livelihoods”31 (Olwig & Sørensen 2002) in the thesis, as a way to conceptualize
migration in relation to people’s possibilities and efforts to make a living.
Moreover, as recent migration theory points out that contemporary migration
patterns often entail many different kinds of moves, encompass different
phases, and link different places and peoples in the process, I make use of
theoretical ideas and concepts that highlight the processual and relational
nature of migration; most importantly, “translocal geographies” (Brickell &
Datta 2011), “transnationalism” (e.g. Vertovec 2009), and “transnational
social space” (Faist 2000).
Migration, social transformations and development processes
This thesis applies a social transformation perspective on migration, which
integrates migration theory and social theory in the analysis (see e.g. Castles
2010; Davies 2007; Portes 2009; and de Haas 2010). Migration is seen in this
thesis as interlinked to global socio-economic transformations (see e.g.
Castles 2010; Davies 2007; Portes 2009) – or, “globalizing” processes (see e.g.
30 Following Boyle, Halfacree and Robinson (1998: 34), migration is defined in the thesis as “the movement of
a person (a migrant) between two places for a certain period of time” (italics in original). Migration is often also
defined spatially (as movement across internal or international boundaries), temporally (as more or less
permanent moves), and according to motivation (as more or less voluntary moves) (ibid.). These distinctions
between, for example, temporary and permanent migration are not so clear-cut in reality. They may, however,
be useful for analytical purposes, as King (2002, 2012) points out. In this thesis, both internal and international
moves, of both permanent and temporary character, are regarded as “migration”.
31 Since this thesis focuses on “migration” but uses a concept including the term “mobility” (i.e. mobile
livelihoods), the terminology needs some further explanation. Migration, according to Boyle, Halfacree and
Robinson (1998), should be distinguished from mobility, which “embraces all forms of geographical
movement” (ibid.; my emphasis), ranging from e.g. international moves, changes of domicile within the same
residential area, and shopping trips (see Åkerlund 2013 for a detailed discussion of the concept of mobility).
Scholars within the “mobilities paradigm” (e.g. Urry 2000, 2007) sometimes argue that migration should also
be seen as a subset of spatial mobility, but as King (2012) points out, this has its disadvantages. For example,
attention can be diverted from the embodied experience and “groundedness” of migration, as well as from the
unequal power relations that influence border control regimes. Based on this, I generally use the term
“migration” in this study, and mostly make use of literature within migration studies. The concept of mobile
livelihoods is an exception to this, since I find it useful for integrating migration and livelihood studies.
Jensen & Tollefsen 2012; Eriksson 2007; Potter et al. 2008; and Bauman
2000) – which are characterized by large social differences (see e.g. UNDP
2009, 2013). More concretely, the transformationalist perspective emphasizes
“how a variety of conditions and parallel processes combine to bring about
large-scale patterns of transformation”, which produce “broader – indeed,
global – enduring, structural shifts in social, political and economic
organization” (Vertovec 2009: 21, 23; following Held et al. 1999). Thieme
(2008: 55) argues that migration – for many, but certainly not for all – should
be regarded as “a necessary and enforced strategy to adapt to economic
globalisation”. One pertinent example of the embeddedness of migration in
socio-economic transformations concerns the structural adjustment
programmes (SAPs) (see above), which have indirectly shaped migration
patterns in many countries (see Davies 2007). Because of their magnitude,
social transformations differ slightly from more general development
processes that usually take place in only one locality or on one level or scale
(Vertovec 2009). Still, social transformations naturally influence local
conditions (e.g. the level of development in a particular place), at the same
time as the local commonly affects the global. Hence, social transformations
may be described as global-local processes, or as a dialectic relationship in
which the local and the global are constantly co-constructing one another32
(Massey 2005). This global-local relationship is, furthermore, framed by
macro- and micro-level power relations; hence, there exists a “power
geometry of time-space compression” (Massey 1994) in which people and
social groups are placed (see also McDowell 1999).
Development processes are nevertheless important in the thesis, especially
with regard to their dynamics with migration and health. Development is a
value-laden concept with many meanings. Long regarded as synonymous to
economic growth, the term has successively come to include social and
political values. In general, development entails changes; preferably – but not
always – in terms of improvements in people’s lives, or in societies at large
(for longer discussions on the concept, see Chant & McIlwaine 2009; and
Potter et al. 2008). In this thesis I follow the United Nations’ (UN) view, which
implies that human development is a dynamic concept that concerns initiating
and sustaining positive outcomes (changes) over time, as well as eradicating
structural injustices that impoverish or oppress people (see, e.g., UNDP
2010). Based on ideas primarily stemming from the work of Amartya Sen (e.g.
Sen 2000), according to the UN development also concerns personal freedom
and choice. In short, human development may thus be defined as “the
expansion of people’s freedoms to live long, healthy and creative lives” (UNDP
32 This dialectic relationship is what the term “glocalization” tries to capture (see e.g. Robertson 1995; Bauman
2010: 2). Based on this view of development, the UN has developed the
Human Development Index (HDI) through which the level of development in
all countries of the world is measured every year (UNDP 2013). The index, a
summary measure of human development ranging from 0-1, measures the
average achievements in a country in three basic, but crucial, dimensions of
human development: (i) a long and healthy life (i.e. life expectancy at birth),
(ii) access to knowledge (i.e. mean years and expected years of schooling), and
(iii) a decent standard of living (i.e. GNI per capita)33. The HDI makes it
possible to follow a country’s development in different indicators over time
(see Chapter 4 for the HDI development in Nicaragua).
Migration and development interrelate in manifold ways; yet, despite
extensive research on the migration-development nexus – for example, the
prospect of “brain drain” in countries with a high number of high-educated
emigrants, or the effect of remittances on local development – the results and
conclusions are varied, and there are still no clear-cut answers to the overall
question of whether migration is beneficial for human development (see e.g.
Faist, Fauser & Kivisto 2011; Glick Schiller & Faist 2010). One reason for this
is that the effects of migration often “[depend] upon who is migrating, where
they are moving from and to, how they move, what they do after they move,
and the political, economic, social and cultural contexts in which the
movement occurs” (Raghuram 2009: 107). Hence, paraphrasing Raghuram
(2009), “context is critical” when analysing the migration-development
nexus. Indeed, migration does not take place in a social, cultural, political or
institutional void (as implied in neo-classicist economic migration theory) (de
Haas 2010), which makes the surrounding spatio-temporal context utterly
important when analysing the potential developmental effects of migration.
Portes (2009: 18) provides a useful typology in which he highlights the
importance of the type of migration for its developmental effects – whether it
is skilled or unskilled, cyclical or permanent migration. I would suggest adding
the factor of immigration status to Portes’ typology, since the fact of having or
not having the legal right to stay in another country may have substantial
implications, both on the migrant’s experience in the new country and on
33 Life expectancy at birth refers to the number of years an infant is expected to live at birth if prevailing
mortality rates continue. Mean years and expected years of schooling are the average number of years of
education received by people older than 25 years, and the number of years of schooling a child can expect to
receive if prevailing enrolment rates continue. GNI per capita refers to Gross National Income (GNI) per capita
(PPP US$) (i.e. the aggregate national income converted to international dollars and divided by midyear
population). The indicators for the second and third dimensions of the HDI have varied since the first Human
Development Report (HDR) was published in 1990 (due to refinements in methodology). Previous reports used
“adult literacy rate” and “combined gross enrolment in school” (except for the years 1991-94 when “adult
literacy” and “mean years of schooling” were used, and 1990 when only “adult literacy” was used), and instead
of GNI per capita, GDP (Gross Domestic Product) per capita was used (UNDP 2011, 2013, 2014). For further
information on how the HDI is calculated, see the Human Development Report 2013, Technical notes, available
online: http://hdr.undp.org/sites/default/files/hdr_2013_en_technotes.pdf.
his/her possibilities to work, send remittances, and have contact with the
origin. Raghuram and Portes thus point to the need to contextualize and
differentiate the migration-development nexus. Even though the results from
the field are inconsistent, one might however argue that for the people
involved in the process, remittances may be an important – and sometimes
crucial – source of income, and also possibly lead to enhancements in life.
Moreover, “brain-drain” effects are possibly less salient today, with migration
patterns increasingly characterized by transnationality and circularity (so that
ties are maintained between origin and destination countries) (de Haas 2005).
Mobile livelihoods
In order to further connect migration to social theory but still relate it to social
transformations and development processes, I make use of the concept of
“mobile livelihoods” (Olwig & Sørensen 2002; see also Briones 2009 for an
application of the concept in a study on Filipina domestic workers in Paris)34.
Olwig & Sørensen (2002:9) conceptualize mobile livelihoods as “the various
practices involved in ‘making a living’, as well as the social relations used to
make a living possible, in the different contexts where they take place” – either
close-by or far away, within the same state or locality, and/or in another
nation35. Using mobile livelihoods is thus a way to conceptualize migration in
relation to the practice of livelihoods. Livelihood approaches generally seek to
empower “poor” people, by viewing them as active agents in trying to improve
their livelihoods within the (often) constraining surrounding conditions (de
Haas 2010). “Livelihoods” may, furthermore, be defined as the “capabilities,
assets (material and social resources), and activities required for a means of
living” (de Haas 2010: 244, with reference to Carney 1998; see also Chambers
& Conway 1991, referred to by Helgesson 2006). The concept thus embraces
three main aspects that are necessary for making a living: (i) capabilities, i.e.
personal characteristics and abilities to cope with stress and shocks, and to
find livelihood opportunities and pursue them; (ii) assets, including various
types of “capital”, i.e. human capital (labour resources within the household),
social capital (social networks, access to institutions), financial capital
(savings, remittances, pensions), natural capital (land, water), and physical
capital (transportation, housing, equipment); and (iii) the activities
undertaken within the household to make a living (Helgesson 2006, based on
34 Susan Thieme (2008) also proposes the linking of (transnational) migration theories and livelihood studies,
as a way to connect migration studies to social theory and show its embeddedness in people’s livelihoods (as
does Jonathan Rigg, see e.g. Rigg 2007). Thieme proposes the concept of “multi-local livelihoods” to capture
the importance of migration in people’s livelihoods (thus similar to the concept of mobile livelihoods).
35 Depending on the context, different theoretical perspectives can be used in the analysis; for example, a local,
translocal or transnational perspective. I apply the translocal and transnational perspectives in this study, as
explained further on in the text.
Rakodi 2002; see also de Haas 2010, with reference to Carney 1998). Taken
together, these capabilities, assets and activities form so-called “livelihood
strategies”, which can be defined as “a strategic or deliberate choice of a
combination of activities by households and their individual members to
maintain, secure, and improve their livelihoods” (de Haas 2010: 244)36.
According to Hein de Haas (2010), migration is increasingly recognized as an
important livelihood activity (yet, he emphasizes that migration is generally
not the only livelihood strategy in a household but is instead often combined
with other multi-local or multi-sectoral household activities). In his words,
migration is “one of the main elements of the strategies households employ to
diversify, secure, and, potentially, durably improve, their livelihoods” (ibid. p.
244) (even though de Haas’ argument is mainly based on evidence from ruralurban migration, he argues that it can also be applied to international
migration). Remittances are seen as part of a household’s livelihood strategies.
The household perspective promoted in livelihood approaches is useful, as it
situates the individual migrant’s actions in relation to the broader social
context, which is especially relevant in studies on migration in Third World
countries, where individuals often act on behalf of the family (Bauer &
Zimmermann 1998, in de Haas 2010)37. Nevertheless, a sole focus on the
household or the family may be problematic, since individuals often also have
social bonds with persons other than household members (de Haas 2010).
Furthermore, it is important to acknowledge that power relations also exist
within families or households, based on for instance age and gender, which
may create inequalities and have implications on the decision-making
concerning livelihood strategies). Susan Thieme (2008) argues, in fact, that
both migration and livelihood studies have tended to be “blind to inequalities
and unequal power relations in the migration process, as well as to the social
and cultural differences between societies and the resulting respective (and
conflicting) networks of migrants” (Thieme 2008: 57). More recent
approaches to migration therefore emphasize the need to analyse power
relations and inequalities in migration processes, especially when the research
focuses on households or families (see e.g. de Haas 2010).
36 Livelihood strategies are also usually shaped or constrained by surrounding factors, such as policies,
institutions, infrastructure and service, vulnerability context, and external environment (Helgesson 2006,
based on Rakodi 2002). Nevertheless, even under constrained circumstances, people – even “poor” people –
have the possibility to develop and pursue strategies. As Helgesson (2006: 18) states “[g]iven the horizon of
possibilities and the space of constraints, one could argue that each actor has a space of action” (with reference
to Long 2001; italics in original).
37 The focus on families and households is shared with the New Economics of Labour Migration (NELM) (cf.
Stark 1978, 1991), which perceives migration as a household risk-spreading strategy, and at the same time a
strategy for overcoming market constraints (thus an extension of earlier, neo-classical theory, which exclusively
interpreted migration as an “optimal allocation of production factors” [de Haas 2010: 230]). Moreover, the
NELM considers remittances to be of central importance in migration processes, even as a fundamental reason
for migrating, and crucial for providing the household with money to possibly improve the household economy
(de Haas 2010, 2012).
Translocal geographies and transnational social spaces
In the thesis I also make use of other conceptualizations of migration that
stress its processual and relational nature. The concept of “translocal
geographies” (Brickell & Datta 2011) builds on previous relational migration
perspectives, most importantly the “transnational” perspective, which sees
migration as a process in which migrants interact and identify with multiple
nations, states and/or communities, rather than as a static act consisting of
moves from one country to another without further contact between origin
and destination (see, e.g., Glick Schiller, Basch & Blanc-Szanton 1992; Portes,
Guarnizo & Landolt 1999; Faist 2000; Levitt & Glick Schiller 2004; Tollefsen
& Lindgren 2006; Vertovec 2009; Glick Schiller & Faist 2010), as well as the
idea of “translocality”, which emphasizes local-local connections as a way to
situate “deterritorialized notions of transnationalism” (Brickell & Datta 2011:
3). The concept of translocal geographies extends the transnational and
translocality frameworks by emphasizing how spaces and places are both
situated in and connected to a variety of locales – e.g. local, regional, national,
and international – without giving preference to any particular situatedness
(such as the nation in previous perspectives). It thus provides a framework for
analysing the geographies of everyday lives across spaces, places and scales;
and hence, for analysing “the overlapping place-time(s) in migrants’ everyday
lives” (ibid. p. 4), including different types of movements – be it internal or
international migration, or other types of mobility and non-mobility.
The transnational perspective is still an important theoretical foundation in
the thesis, particularly in the analysis of health in relation to international
migration. Scholars who use the concept of transnationalism often (but not
always) complicate the acts of migration, through critically asking who
migrates, and under what circumstances (see e.g. Marchand 2009; Grewal &
Kaplan 2001). This critical approach is important to also keep in mind in the
process of trying to understand health in relation to Nicaraguan migration
processes, for instance by asking: who is moving? where to? under what
circumstances? and with what effects? A central idea within the transnational
perspective is that of “transnational social space”, which Faist (2000: 199)
defines as “combinations of sustained social and symbolic ties, their contents,
positions in networks and organizations, and networks of organizations that
can be found in multiple states”. Primarily three forms of transnational social
spaces exist, according to Faist: transnational kinship groups (multi-local
households, e.g. transnational families), transnational circuits (trading
networks, e.g. Lebanese businesspeople), and transnational communities
(diasporic groups, e.g. the Jewish diaspora). Transnational kinship groups will
be discussed further in the empirical part (see Chapter 7).
Even though I use a translocal and transnational perspective on migration in
the thesis – which emphasizes cross-border movements and contact – I would
like to stress that borders, particularly international borders between one
independent nation and another, still have a role to play in today’s “mobile”
world, and that border thinking is still fundamental to the human experience
(see, e.g., Khosravi 2011; Yuval-Davis & Stoetzler 2002; Newman 2003; Silvey
2006). The concept of borders is therefore also important, and will
consequently be discussed in the thesis (see Chapter 6).
The interrelations between migration and health
This section will first point out some basic understandings of migration-health
relations that are fundamental in the thesis, and thereafter discuss certain
themes of migration-health relations of particular importance to the thesis
subject. After this section, the chapter closes with a re-capitulation of the
theoretical framework used in the thesis.
The migration-health nexus as a bi-directional process
The migration-health nexus is commonly viewed in the literature as a bidirectional process in which migration and health may affect one another
(Jatrana, Graham & Boyle 2005, with reference to Hull 1979; Gatrell & Elliott
2009). This thesis therefore analyses the various effects of migration on
health, and the various effects of health on migration. As migration is
understood as a relational process here, health is scrutinized in relation to the
whole process of migration, including the different actors, spaces, places, and
scales it involves. I therefore make use of the frameworks for analysing
migration-health relations developed by Haour-Knipe (2013) and
Zimmerman, Kiss and Hossain (2011). In these frameworks, migration is seen
as encompassing several different places and phases, which the authors argue
should be included in analyses of migration and health: (i) the origin (predeparture phase); (ii) the transit (travel phase); (iii) the destination
(reception, interception and integration phases); and (iv) the origin (return
phase). In the thesis, health is thus “traced” within the migration process.
Consequently, the study analyses migration-health relations from both the
individual and the broader structural perspective (relations at the micro,
meso, and macro levels). On the individual scale the focus has been on, for
example, health problems (physical and mental). On the relational scale,
attention has been paid to social relations and social networks, for example.
Lastly, on the structural scale the focus has been on aspects such as national
politics and global institutions.
The “globalized” body
As mentioned previously, a globalization of the body has taken place in
relation to the “globalizing” world (Turner 2004). Accordingly, much
attention has been given to, for example, the “risk society” and global
regulations. Concern has also been raised in the field of public health, for
instance regarding the challenges population mobility imposes on nationstates’ health systems and health policies (see e.g. MacPherson & Gushulak
2001). In this context, migration has persistently been viewed as a threat to
public health; a view dating back to medieval times when measures to control
the spread of communicable disease were common, and perhaps sometimes
necessary. A more rights-based approach towards migration and health has
gained footing in recent years, in connection with the research and policy field
of global health. In relation to this field, a discussion has arisen concerning
health as a human right –for migrants as well. These views will be discussed
The tendency to regard migrants as potential disease-spreaders is highly
connected to what the anthropologist Mary Douglas (1966) called “fear of
pollution”. Douglas argued that an important mechanism for preserving social
structure is to distinguish “purity” from “impurity” or “polluting” elements (in
the context of this study, for example, to preserve national sovereignty by
distinguishing foreigners from nationals). Following Douglas’ reasoning,
migrants can be seen as “polluted” and “polluting” (Khosravi 2011; Malkii
1995). Similarly, today’s post-colonial theory theorizes the “othering” of
migrants. This theory provides a useful outlook, through “making visible the
process by which concepts such as ethnicity, race and culture have been
constructed and used to create binaries locating non-European peoples as the
essentialized, inferior, subordinate Other” (Khan et al. 2007: 230). For
example, writing about “stranger danger” (i.e. fear of “others”), Sarah Ahmed
(2000) says that “[b]y defining ‘us’ against any-body who is a stranger, what
is concealed is that some-bodies are already recognised as stranger and more
dangerous than other bodies” (pp. 3-4). In the process of self-definition, of
identification, the stranger (the “other”) is consequently ousted as the origin
of danger. Hence, the stranger is “an effect of processes of inclusion and
exclusion, or incorporation and expulsion, that constitute the boundaries of
bodies and communities” (Ahmed 2000: 6). Consequently, the “stranger
danger” discourse discussed by Ahmed may, for example, lead to societies
characterized by hostile relations to immigrants. Sandoval-García (2004)
writes about these processes in relation to Nicaraguan migration to Costa Rica
(see Chapter 6).
In contrast to the view of migrants as a health threat, the rights-based
approach stresses the health implications for individual migrants, and the
challenges health care systems face to provide services to all (Thomas &
Gideon 2013; Zimmerman, Kiss & Hossain 2011). The proceedings from the
Seminar on Health and Migration, organized in 2004 by the International
Organization for Migration (IOM) and WHO (amongst others), clearly
demonstrate that the rights-based approach has largely been adopted by the
international community, but also that the view of migration as a threat to
public health is still viable today38. The idea of a transnational, or flexible,
citizenship has been introduced in relation to this, through which citizen
rights – usually tied to the nation-state – are instead seen as tied to the
individual, wherever he or she should travel or settle down (Turner 2004, with
reference to Bauböck 1994, and Ong 1999). Turner (2004) emphasizes the
need for a “medical citizenship”, which would ensure all humans’ right to
health, regardless of national citizenship or residence. In Turner’s (2004: 269)
words: “the right to health cannot be detached from more general questions
of social equality and security. […] Health is an outcome of a complex web of
rights that attempt to address issues of justice in a world where such questions
can find only global solutions”. Women’s health rights have been particularly
highlighted in this discussion, as women often suffer the most from poverty
and inequality, and are also often exposed to more vulnerability as migrants
(e.g. as sex workers, refugees or labour migrants) (Turner 2004).
This thesis analyses the migration-health nexus in the case of Nicaragua as
part of the discourses and research on both migrant health rights and
migrants as a health treat. Paricularly relevant are the issues of the “othering”
of migrants (including experiences of vulnerability and precariousness) and
migrants’ access to health care, and the effects these have on migrant health.
Migrant health
This section will discuss aspects of migrant health that I have found to be
particularly relevant for analysing the study’s empirical material – migrant
work, vulnerability and precariousness, stress, discrimination and racism, and
access to health care.
38 As seen in the issues discussed at the meeting: national migration policies, pre-departure health screenings,
the mental health of migrants, the migration of health care workers, health care for undocumented migrants,
policy reform, and investments in migration health; as well as in the meeting’s concluding statements, which
state for example that there is a “need for a global approach to public health management, the creation of
comprehensive policies capable of servicing all migrant populations, and partnerships in migration health”
(IOM 2005: 10).
The issue of migrant work is highly relevant in this thesis, since contemporary
Nicaraguan migrations are primarily connected to people’s strategies for
making a living. Consequently, many Nicaraguans have experiences of
migrant work, often characterized by unequal, racialized and gendered work
relations, which may affect health diversely. Through work, the body is
intimately connected to larger socio-economic relations, often characterized
by unequal power relations in today’s complex, “globalized”, and polarized
labour markets (Wolkowitz 2006). Bonacich et al. (2008: 342) argue that the
division of labour within today’s global capitalist system is “a hierarchically
organized, racialised labour system that differentially exploits workers based
upon their gendered and racialised location”. Similarly, María Lugones
(2007), discussing “the coloniality of power” (cf. Quijano 2000), states that
the division of labour under “global, Eurocentered, capitalist power” is
“racialised as well as geographically differentiated” (Lugones 2007: 191).
Furthermore, the social position of migrant workers is also determined by
these processes (Weiss 2005). Care work provides a good example of the
gendered, international division of labour. Scholars argue that global
inequalities are a precondition for functioning “body work”39 in the North,
since a large part of the labour power it involves consists of immigrants or
migrant workers (Wolkowitz 2006). The concept of “transnationalization of
reproductive labour” (e.g. Ehrenreich & Hochschild 2002; for an overview, see
also Wolkowitz 2006) has been developed in relation to this, illuminating the
process whereby migrant women from poorer countries leave their families
and children to take care of the reproductive work (child-rearing and
household work) in richer families in richer countries (see also Lutz 2002).
“Global care chains” also evolve in this process, through the
commercialization of care and love (e.g. Hochschild 2000; for an overview see
also Yeates 2012). These care chains also exist on a societal level, since
immigrants and migrant workers represent a large part of the labour force
within public health care in many richer societies40. The work performed by
migrants in other sectors of the labour markets in richer societies, such as
agriculture and industry, can also be seen in this light (Wolkowitz 2006).
Thus, health may be profoundly affected by work relations; in terms of both
what type of work is performed and the social relations and social structures
surrounding the work process. The large-scale changes in the structure of
labour markets41 indicate that work should also be analysed in connection to
39 “Body work” concerns work involving the body, such as beauty care, physical exercise, health care, child
care and housekeeping (Wolkowitz 2006).
40 This is said to potentially lead to the “brain drain” of the Third-World countries from which the care
workers originate.
41 For example, the global shift in manufactural production from developed to developing countries, and the
post-industrial, service-oriented economy, including new forms of work and employment.
the social aspects of society. The theoretization on work, the understanding of
today’s global labour markets as increasingly unequal, racialized and
gendered, as well as its relation to health, is highly relevant in this thesis, since
many study participants had migrated for economic reasons. Furthermore,
the concepts of transnationalization of reproductive labour and global care
chains are applicable, since many Nicaraguan women work as nannies and
maids in Costa Rica.
The concepts of “vulnerability” and “precariousness” are important in relation
to migrant work and health. Vulnerability is a broad concept referring to, for
example, people’s vulnerability in the face of large-scale changes (e.g. climate
change, natural disasters, economic decline), and social aspects (e.g.
vulnerability within households). At the core of the concept lies the exposure
to shocks or strains, and experiences of suffering (Pendall et al. 2012); thus
one definition relates it to humans’ exposure to psychological damage, and the
propensity to suffer morally and spiritually (rather than physically) (Turner
2004). Wolkowitz (2006) argues that a “structure of vulnerability” (cf. Nichol
1997) has arisen globally, due to the more insecure and difficult working
conditions in the world today. Moreover, illegal border crossings often put
migrants in a particularly vulnerable situation, especially females, who are
often subject to sexualized violence during the border crossing (Khosravi
2011; Falcón 2007; Ruiz Marrujo 2009). Besides difficult strains and
sufferings, migrants also face death during the journey (see e.g. Slack &
Whiteford 2011). Especially relevant in this thesis is the Mexico-US border,
where many migrants – including Nicaraguans – die every year (Eschbach et
al. 1999, 2001; Sapkota et al. 2006; Holmes 2013; Guerette 2007). The
literature points to a relation between increased border security and migrant
fatalities (e.g. Eschbach et al. 2003; Cornelius 2001; Orraca Romano & Corona
Villavicencio 2014), partly due to a redistribution of migratory flows (a “funnel
effect”) into more remote and dangerous areas. Moreover, the enforced border
politics also lead to more difficulty travelling back and forth, thus resulting in
the migrant’s “entrapment” inside the US.
Precariousness refers above all to the insecure working conditions we see
throughout the world today. Guy Standing (2014) argues that the large-scale
changes in labour relations, patterns of work, and systems of social protection,
regulation and redistribution that have taken place under neo-liberal global
capitalism have created a new class structure in which the “precariat” is a new,
lower class. Migrant workers are in a particular vulnerable and precarious
situation, for they are over-represented in precarious jobs (“dirty” and lowpaid jobs, such as cleaning and agriculture) (e.g. Wolkowitz 2006). Moreover,
it can be argued that migrant workers suffer a “double” precariousness,
particularly if they have irregular status42. Goldring and Landolt (2011)
discuss how precarious work and precarious legal status (e.g. irregularity)
intersect, and argue that migrant workers’ insecurity and vulnerability stem
not only from irregularity, but also from the migrants’ social positions
(ethnicity/“race”, gender and class). There is research that points to the
negative health effects of precariousness (e.g. Tompa et al. 2007; Brabant &
Raynault 2012).
Both concepts, vulnerability and precariousness, are central in this thesis.
Through vulnerability, the exposure to risks and suffering in relation to the
migration process can be highlighted and better understood; and, through
precariousness, the dimension of insecurity in relation to work (“precarious
work”) can be added to the analysis.
In light of the above, it is evident that migration can be a stressful experience.
The act of moving to a new country perhaps entails the most stress,
particularly if the move is conducted without documents. Still, closer moves
can also have stressing effects. In fact, all kinds of migration generally entail
dislocation and disruption – not only spatially, but also socially, culturally,
and environmentally – which may be experienced as stressful (e.g. Gatrell and
Elliott 2009). The processes of “estrangement” (Ahmed 2000) and “cultural
bereavement” (Helman 2007; with reference to Eisenbruch 1988) may cause
stress in the individual migrant, in relation to being in a new place. Another
potential source of stress among migrants is the reception at the destination,
often characterized by xenophobia, racism and discrimination (Gatrell &
Elliott 2009), which may cause negative health effects (e.g. Williams et al.
2003; Paradies 2006). However, there is still uncertainty regarding the
relation between racism and stress (for example, whether racism is distinct
from other types of stressors), which points to the need for further
investigation (Paradies 2006). The issue of migration and stress is important
in this thesis, for analysing the health impacts of migration. The concepts of
estrangement and bereavement are used to illustrate the emotional side of
migration stress, and discrimination and racism are used to discuss both the
direct and indirect implications it may have on the health of migrants.
In the literature, the issue of migrants’ access to health care is discussed
primarily in relation to international migrants. Hargreaves and Friedland
(2013) (with certain focus on European conditions) state that existing studies
show that immigrants – even though they may be entitled to health care –
often face certain barriers in accessing it, which influence service use and may
42 There are certainly other types of migrants besides migrant workers who are in a precarious situation, for
example refugees, sex workers and victims of trafficking or smuggling, whose particular vulnerabilities have
been highlighted in recent literature (see e.g. Van Liempt & Bilger 2009).
explain patterns of ill-health. Socio-economic position and immigration
status, together with discrimination and a lack of understanding of the health
care system, seem to cause the most important constraints to migrant’s access
to health care (Cabieses & Tunstall 2013; Gideon 2013; Gatrell & Elliott 2009).
In relation to internal migration, one may argue that rural-urban migrants, in
general, improve their access to health care due to the urban bias in
localization of services. The issue of migrants’ access to health care, and the
barriers constraining this access, is important in this study, in relation to both
internal and international migration. As regards internal migration, the
question of better access in the cities is especially important; and concerning
international migration, the question of entitlements and discrimination is of
particular importance, especially for the undocumented migrants in the study.
Transnational families and health
In relation to the increasingly transnational and circular patterns of migration
described earlier in the thesis, and the resulting changes in family relations,
there has been an upsurge in literature on “transnational families”. The issue
of transnational families is important in this thesis for understanding the
implications of transnational – as well as translocal – migration on family life,
and the effects these implications may have on health and caregiving.
“Transnational parenthood” (transnational mothering and fathering) and
“transnational caregiving” are particularly important for analysing the effects
of family separation, and the ways in which transnational family life is coped
“Transnational families” are a particular type of families, characterized by the
fact that the family members “continue to feel they ‘belong’ to a family even
though they may not see each other or be physically co-present often or for
extended periods of time” (Baldassar & Merla 2013: 6). Despite being
separated over time and space, transnational families thus maintain a sense
of “family-hood” (Bryceson and Vuorela 2002; quoted in Baldassar & Merla
2013). In order to sustain this family-hood, transnational families employ
different strategies, including, for example, “negotiating a plan for family or
kin to care for children upon the parent or parents’ departure and until their
return, ensuring that economic remittances are sent to support family wellbeing, and using phone calls, letters and video to stay involved in each other’s
lives” (Schmalzbauer 2008: 334).
One central aspect in the research on transnational families is familial
separation. Research shows that separation from family may induce great
stress, affect the emotional well-being, and sometimes even cause depression,
for both migrants and their family members who remain in the country of
origin (e.g. Aguilera-Guzman et al. 2004; Aroian & Norris 2003; Espin 1987,
1999; referred in Silver 2011). In relation to children who are separated from
their parents, Suárez‐Orozco, Todorova and Louie (2002) write that the
literature points to a negative experience for the children, both during the time
when their parents are gone and when/if reunification takes place. According
to Silver (2011), one important reason for this is that migration involves a
strain on the support networks of both migrants and their family members,
sometimes even leading to the breakdown of social support. This might cause
stress in the individual since social support, particularly stemming from close
relationships (e.g. spousal and parent-child relations), is an important buffer
against mental distress, as mentioned earlier (see also Thoits 1995, 2010).
Another, closely aligned reason for the health effects of separation is that the
family undergoes fundamental transformations in relation to transnational
migration. New familiar relations, new roles, and additional responsibilities,
for example, need to be adapted to, and these changes may be an important
source of stress for both the migrant and the family members left behind (e.g.
Silver 2011; Schmalzbauer 2004; Pribilsky 2004). The issue of parenthood has
consequently received much attention in the literature, since parent-child
relations are fundamentally changed in cases of parents migrating and
children being left behind at the origin (see e.g. Carling, Menjívar &
Schmalzbauer 2012). A new form of “transnational parenthood” has arisen, in
which “the parent-child relationship is practised and experienced within the
constraints of physical separation” (ibid. p. 193). As parenting roles are
commonly gendered (performed in different ways by men and women),
transnational parenthood is also “affected in gender-specific ways” (ibid. p.
192), both for the mothers and fathers who stay and for those who migrate.
Even though both migrating mothers and fathers perform similar
transnational parenting activities – for example, sending money and gifts, and
maintaining communication – many studies show that greater expectations
often are placed on mothers who migrate, and that migrating mothers
continue to be responsible for the emotional care of children (see e.g. Parreñas
2000, 2001, 2002, 2005; and Hondagneu-Sotelo & Avila 1997). A
predicament of transnational motherhood is therefore that the act of
“mothering” is performed from afar, which may be a difficult and stressful
experience (ibid.). Transnational fathers and “fathering” have not received as
much attention in the literature as have transnational mothers and mothering
(Parreñas 2008), probably because the father’s absence through migration is
more consistent with traditional gender norms (e.g. male breadwinning) and
therefore does not reconstitute the gender behaviour in the family.
The separation within transnational families can lead to other stressful
experiences as well. Differences in access to resources and decision-making
between family members may arise, for example, which can cause stress for
both the migrant and the family members left behind (Silver 2011). The
difficulties separation entails are perhaps felt even more when migration is
undertaken irregularly, since it might lead to entrapment in the destination
country, as mentioned earlier (see Khosravi 2011; Menjívar 2012). However,
the separation within transnational families naturally does not always cause
emotional pain or stress. Some may experience the separation as providing
new opportunities, independence and empowerment (e.g. Silver 2011).
Transnational families are greatly sustained by the exchange of “transnational
caregiving” (Baldassar & Merla 2013), a particular form of care and support
that is reciprocal yet uneven: “[t]ransnational caregiving, just like caregiving
in all families (whether separated by migration or not), binds members
together in intergenerational networks of reciprocity and obligation, love and
trust, that are simultaneously fraught with tension, context and relations of
unequal power” (Baldassar & Merla 2013: 7). The idea of transnational care
circulation extends previous conceptualizations of global care chains, and
highlights “how care circulates around a wide network of friends and family,
crisscrossing both local and national settings” (ibid. p. 12).
In this thesis the conceptualizations concerning transnational families are
important in the analysis of translocal family life and the effects separation
may have on health.
Recapitulation: a critical framework for analysing the
migration-health nexus
In this thesis I emphasize the need to place Nicaraguan migration processes
in relation to relevant globalizing processes in modern times, for example the
“globalized” labour market, and the period of structural adjustments that also
fundamentally affected Nicaraguan society. The analysis of migration-health
relations also needs to be situated in relation to development processes, and
how they interact with global social transformations and their inherent power
relations. Additionally, the question of whether migration is positive or
negative for development – including that of people’s health – needs
contextualizing and differentiation in order to grasp the interrelations and
effects. In order to analyse migration-health relations in the case of Nicaragua,
I make use of the frameworks elaborated by Haour-Knipe (2013) and
Zimmerman, Kiss and Hossain (2011), which state that migration-health
relations should be analysed within the whole process of migration. I
consequently scrutinize the entire process of migration in order to “track” the
bi-directional connections between migration and health, including the
different actors, spaces, places, and scales this involves.
I use the concept of mobile livelihoods as it highlights the embeddedness of
migration in people’s lives and livelihoods, of which remittances are one
important aspect. Moreover, the concepts of translocal geographies,
transnationalism, and transnational social spaces are used in the thesis as they
highlight the processual and relational nature of migration. The concept of
translocal geographies is especially applicable in the analysis, as it emphasizes
relations across different spaces, places and scales (thus, both internal and
international migrations can be analysed within this framework). By use of the
transnational perspective, cross-border relationships, ties, and networks, as
well as inherent inequalities and mechanisms of in/exclusion from global
migration processes, can also be analysed, as can the connections of migration
to social transformations. Conceptualizations of borders are furthermore
applied in the thesis since they shed light on inherent asymmetries in
migration processes.
An integrative/holistic view of health is applied in the thesis, based on the
biopsychosocial model of health, and the concept of mind/body health. These
perspectives imply that health is composed of a combination of biological,
psychological and social factors, and that physical and mental health are
tightly connected. The social perspective (e.g. the social determinants of
health) further stresses that health is shaped by social factors (e.g. social class,
gender and ethnicity), and also that power relations – which are inherent to
all social activities – also influence health outcomes. Social factors and power
relations have therefore been taken into account in the analysis of migrationhealth relations; for instance, through the use of the concept of “social
citizenship”, which highlights the importance of access to society’s resources
for health, and the entitlements citizens have to, for example, health care
services. The dynamic between health and development is also crucial in the
analysis since, for example, the issues of medical globalization, structural
adjustment programmes, and the Millennium Development Goals are at
centre stage in Nicaragua. The idea of social capital is, furthermore, used in
the thesis in order to operationalize particular aspects of the social fabric that
may influence migration-health relations (e.g. social relations, social support,
and reciprocal exchanges of help/remittances). Moreover, the concept of
embodiement is used in the thesis as a way to more deeply elaborate on the
connections between the social and health. Additionally, emotions are
analysed, based on theories on the relations between emotions and health,
stress and coping.
The findings of the study are put in relation to the issues of migrant health
rights and the “othering” of migrants. The sociological theoretization on work,
and its relation to health, is used in the thesis for analysing migrant work in
the context of unequal, racialized, global work relations, and their effects on
health. The concepts of transnationalization of reproductive labour and global
care chains are used in relation to this for analysing the particular situation
for female migrant workers in care sectors. Moreover, I apply the concepts of
vulnerability and precariousness to discuss the exposure to risks and suffering
during the migration process, as well as insecure work conditions.
Additionally, literature on the stresses of migration – most importantly
discrimination, racism and health – is used for discussing the health impacts
of migration. The issue of migrants’ access to health care is also central in the
thesis, in relation to both internal and international migration. Lastly,
research on transnational families, transnational parenthood, and
transnational caregiving is applied in the analysis of transnational – as well as
translocal – migration, and the effects these processes may have on family life,
health and care. Moreover, it is a firm understanding that social differences,
power relations and inequalities shape the relations between migration and
health under scrutiny in this thesis, in which questions of ethnicity/“race”,
class, gender, and legal status play a crucial part.
The next chapter will discuss the empirical material the study is based on, as
well as the methods used to analyse it. The last chapter in Part I, Chapter 4,
will thereafter provide a background to the context of the study – Nicaragua,
and the two study settings of León and Cuatro Santos.
“Welcome to San Franscisco del Norte”, Cuatro Santos.
Materials and methods
A mixed-methods case study
This thesis builds on a case study of migration-health relations in Nicaragua.
The study combines qualitative and quantitative research strategies in order
to shed light on both the deeper and the general picture of migration-health
relations, and thereby explore the case to the fullest. The empirical material,
including qualitative interview data and survey data, was gathered through
fieldwork in two settings in Nicaragua: the town of León, situated in the Pacific
coast area, and the area of Cuatro Santos, consisting of four predominantly
rural municipalities in the northern part of Chinandega (see map on p. x).
This chapter starts with an introduction to case study methodology, followed
by a critical discussion of mixed-methods research. Then, the fieldwork and
the qualitative and quantitative studies are presented in detail, including the
study’s empirical material and the methods used for analysing it. Lastly, some
reflections on the use of mixed methods as well as its advantages and
disadvantages are provided at the end of the chapter.
Case study methodology
“By whatever methods, we choose to study the case.” (Stake 2003: 134)
As Stake (2003) asserts, case study research is not a choice of a particular data
collection method but rather of what is to be studied. Therefore, case studies
may use qualitative or quantitative research methods, or a mix of both – as I
have done in this study. Common to all case studies is that they focus on one,
or a few, specific and unique case(s) in depth in the study of complex
phenomena (Tollefsen Altamirano 2000). Therefore, when applying case
study methodology in research, the definition (selection) of the case is of
major importance. In this thesis, the main case under investigation is
migration-health relations in Nicaragua. The two study settings, León and
Cuatro Santos, can be regarded as “cases within the case” (Stake 2003: 153),
as can the interviewees and survey respondents who participated in the study.
Another important aspect when using case study methodology regards the
type of case study being conducted. While some aim only at describing and
analysing the particularity of a case – intrinsic case studies – others also aim
at portraying what the particular case may say about other cases (i.e. to
generalize) – instrumental case studies. A number of cases may also be
studied with the ambition to say something about other situations – collective
case studies (Stake 2003). This thesis shares many similarities with collective
case studies, since the decision was made to conduct the study of migrationhealth relations with a large number of participants (the survey study), aiming
at generalizing the findings to the greater Nicaraguan population. However,
there are also elements of the intrinsic case study, as I made a good deal of
choices concerning who would participate based on, for example, where they
lived. Moreover, I also found the particular case under study interesting and
important in its own right, which also means that my approach shares
elements with the intrinsic case study.
Case study research shares similarities with the holistic tradition, which
emphasizes the importance of taking into consideration “the whole”43 in
research (Bubandt & Otto 2010). As a methodological tool, the holistic
tradition places great weight on open-mindedness (including as many factors
of importance as possible) and comprehensiveness (including the
surrounding context of the phenomenon under study in the analysis) in order
to grasp the totality of the whole (Bubandt & Otto 2010)44. Although the task
of grasping the totality of a complex whole may be impossible (if a
phenomenon can be regarded as a “whole” to begin with), I find the idea
inspiring. I have therefore taken a comprehensive grip on migration-health
relations in this study and, with an open mind, have tried to include as many
relevant aspects as possible. The study therefore scrutinizes both physical and
43 A “whole” culture, society, region or system, for example. The idea of “wholeness” was foundational in the
“classic” era of geography – primarily in the works of von Humbolt and Ritter, who in different ways tried to
understand the whole complex system of the universe, and emphasized the unity of nature and humanity; and
subsequently in regional geography (e.g. Vidal de la Blache), which regarded regions as complexes of natural
and cultural phenomena in unity (Holt-Jensen 2009). Holism as a concept is used more seldom in human
geography today – as a result of the specialization of geography, and perhaps also due to the shift to relative
and relational views of space – yet, the idea of holistic inquiry is still up for discussion (see e.g. Ley & Samuels
2014 on humanistic geography, and Trudgill & Roy 2014 on physical geography). Sui and DeLyser (2012) even
argue that a “holistic turn” has taken place in geography in recent decades, which they exemplify by the calls for
a unified geography, and the mixing of qualitative and quantitative methods. In anthropology, holism has a
more prominent position (Otto & Bubandt 2010) and is primarily derived from structural functionalism, which
assumed social phenomena to create a whole. Even after post-modern and post-structural criticism the concept
still lingers in anthropology, albeit in new forms. Today, holism can be regarded as “a comprehensive approach
to the human condition” (Bubandt & Otto 2010: 3).
44 Complexity theory also builds partly on the holistic tradition focusing on system’s complex behaviour – for
example, the behaviour of ecosystems or transport networks in terms of relations and networks (e.g.
interconnectedness and non-linear interactions) (Gatrell 2005). Complexity theory thus shares similarities with
system-based approaches in general, and with network theories specifically, but according to Gatrell (2005) its
notions of emergence and hybridity add value, as does the focus on exploratory research. Scholars in health
geography have recently advocated the use of complexity theory in order to enhance the understanding of the
complex relations that influence health, which can seldom be reduced to, for example, linear models of
individual behaviour such as logistic regressions (Gatrell 2005; see also Curtis & Riva 2010).
mental health in relation to the whole process of migration, including the
effects of health on migration and the effects of migration on health, for both
migrants and family members of migrants (left-behinds). Moreover, I analyse
a variety of aspects, on different spatial scales, that may be of importance for
migration-health interactions. Additionally, I provide a rather “thick”
description of the contextual landscape for the case of Nicaragua (Chapter 4),
since I believe – in accordance with case study and holistic methodology – that
it is necessary to recognize that the migration-health relations under scrutiny
do not take place in a vacuum but rather in economic, political, social and
cultural environments (with inherent relations of power), which act as
prerequisites for these interrelations.
Mixed-methods research
The mixed-methods research approach (MM), which I have employed in this
study, “combines elements of qualitative and quantitative research
approaches (e.g. use of qualitative and quantitative viewpoints, data
collection, analysis, inference techniques) for the broad purposes of breadth
and depth of understanding and corroboration” (Johnson, Onwuegbuzie &
Turner 2007: 123). By use of MM, information in both narrative and
numerical forms concerning a research issue can be provided; for example,
qualitative data can be used to add meaning to quantitative data, and
quantitative data can contribute to the generalization of qualitative data. The
research problem is often more at centre stage in MM than the methods per
se, which instead are often seen as mere tools for answering the research
questions. The integration of approaches usually takes place throughout the
research process: in study design, data collection, analysis and presentation
(Hesse-Biber 2010).
The use of mixed methods as a research approach has gradually become more
common in the social sciences. The first study to explicitly talk of mixing
methods stems from the mid-1950s (Campbell & Fiske 1959; in Teddlie &
Tashakkori 2009), but MM as a research approach in its own right –
distinguished from qualitative and quantitative research methods – did not
develop until the 1990s/2000s (for more details on the history of mixedmethods research, see Johnson, Onwuegbuzie & Turner 2007; and Teddlie &
Tashakkori 2009). The health and nursing sciences have produced many
mixed-methods studies since the 1990s (e.g. Cohen et al. 1994, and Bryant et
al. 2000, in Teddlie & Tashakkori 2009; see also e.g. Stewart et al. 2008).
Within geography, at the end of the 1990s population geographers and
migration scholars pointed out the need to employ mixed methods in order to
move migration research forward (see e.g. Findlay & Li 1999; McKendrick
1999; and Lawson & Silvey 1999; for a recent geographic study combining
qualitative and quantitative methods, see Hjälm 2011)45. The mixed-methods
research approach did not enter health geography until the 2000s (see e.g.
Crooks et al. 2011); however, according to Kearns and Collins (2010) it is
gradually becoming more common (but there were still very few MM studies
in 2012, at least at the annual meeting of geographers in Canada; see
Giesbrecht et al. 2014). Still, scholars continue to make further calls for it, for
instance in relation to the advocacy of complexity theory (see Footnote 44).
Curtis and Riva (2010: 220), for example, argue that “the counterproductive
dualism between ‘quantitative’ and ‘qualitative’ methods will have to be
abandoned to integrate both approaches in the understanding of the complex
processes influencing population health”.
Rationales for conducting mixed-methods research
A common reason for using mixed methods as a research strategy is for its
ability to create a synergistic effect and thereby aid in the development of a
research project (for example, using findings from qualitative interviews for
formulating survey questions). Two other major reasons for conducting
mixed-methods research are triangulation and complementarity (HesseBiber 2010). Triangulation refers to the use of more than one method in
studying the same research question in order to find convergence of the data;
it is, in other words, “the display of multiple, refracted realities
simultaneously” (Denzin & Lincoln 2003: 8). Complementarity involves the
use of both qualitative and quantitative data in order to investigate the
research problem to the fullest. Both triangulation and complementarity are
thus useful for cross-validation, but while the former aims to enhance the
credibility of the findings, the latter can offer a more thorough comprehension
of the research problem (Hesse-Biber 2010). As Teddlie and Tashakkori
(2009: 33) write, an advantage of using mixed methods is that “it enables the
researcher to simultaneously ask confirmatory and exploratory questions and
therefore verify and generate theory in the same study”.
The term bricolage has also been used to describe and validate mixedmethods research. According to this idea, the researcher is proposed to be
seen as a quilt maker (bricoleur) who deploys different strategies, methods or
empirical materials depending on the research questions. The researcher as
bricoleur thus works “between and within competing and overlapping
perspectives and paradigms” (Denzin & Lincoln 2003: 9), depending on the
context of the research questions. Similarly, pragmatism asserts that the
45 Furthermore, in 2004, Kwan made a call for “hybrid geographies” that would transcend traditional divides,
for example between qualitative and quantitative geographical studies.
researcher should focus on the values, or outcomes, of research. The most
fundamental issue for pragmatist researchers is consequently how to choose
the methods and theories/approaches that will most likely provide answers to
the research question. I will return to the issue of pragmatism below, and at
the end of the chapter I will discuss it together with the other strategies
mentioned here (see “Reflections on conducting mixed-methods research”).
Mixed methods – mixed paradigms? Some notes on methodology
The concepts of paradigm, methodology, ontology and epistemology46 require
some attention when discussing mixed-methods research, because criticism
has been raised (see e.g. Sale et al. 2002) regarding the potential paradigmatic
difficulties involved with combining qualitative and quantitative methods in
the same study. The criticism of MM is very much related to the historical ties
between qualitative and quantitative methods, on the one hand, and different
scientific paradigms and their respective methodologies on the other. Very
briefly, quantitative methods developed within sciences grounded in
positivistic realism and objectivism, while qualitative methods arose from
research with interpretivist, constructivist and subjectivist understandings47.
Decades of paradigmatic “wars” have consolidated the associations between
the two types of methods and ontological and epistemological assumptions,
which has led to a common perception of immense differences between
qualitative and quantitative methods (Denzin & Lincoln 2003; see also
Lawson 1995). However, as Lawson (1995: 451) states, the “methodological
dualism” that has long dominated human geography, as well as the social
sciences at large, “is historically produced and is not necessary or inevitable”
(italics in original). In fact, according to Lawson, “quantitative methods rely
on considerable subjective interpretation, and qualitative methods necessarily
entail considerable objectification” (ibid.). However, as methodological
understandings “[lead] the researcher to ask certain research questions and
prioritize what questions and issues are most important to study” (HesseBiber 2010: 11), interpretative, transformative and critical methodologies (as
well as, for example, feminist methodologies) have traditionally been regarded
46 Paradigm can be defined as a system of beliefs and practices, or as “a worldview including philosophical and
sociopolitical issues” (Teddlie & Tashakkori 2009: 21), while methodology is more related to the research
process, and embraces the researcher’s understandings of ontology and epistemology (ibid.). Ontology
concerns the understandings of the nature of existence, human beings and reality, while epistemology relates
to the nature of knowledge, and the relationship between the “observer” and “reality” (Hesse-Biber 2010;
Lincoln & Guba 2003).
47 Positivistic realism and objectivism proclaim that reality is the way it is regardless of whether or not we
observe it, and that this reality can be studied and explained in a neutral, objective way; while interpretivist,
constructivist and subjectivist understandings imply that any apprehension of reality is made by means of
interpretation, and that all knowledge about reality – as well as the nature of reality – therefore consists of
subjective interpretations (Lincoln & Guba 2003).
as qualitative, while positivist and post-positivist methodologies have been
seen as quantitative – because of the type of research usually conducted within
these traditions48.
Nevertheless, as Teddlie and Tashakkori (2009: 12) state, “[a] methodological
perspective is not inherently quantitative or qualitative in terms of its use of
method… […] In fact, qualitative and quantitative methods are carried out
within a range of methodologies”; for example, a critical theorist can conduct
quantitative studies, just as a post-positivist can use qualitative methods.
Similarly, Lawson (1995) argues that it is necessary to distinguish techniques
(methods) from methodological positions. For, in contrast to research
methodology, research methods are “specific strategies for conducting
research” (ibid. p. 21). Important to keep in mind, however, is that “the
method is but the tool” while “the methodology determines the way in which
the tool will be utilized” (Teddlie & Tashakkori 2009: 17). In relation to this,
Lawson (1995) states that we as researchers must acknowledge and take
seriously the fact that we “mark the knowledge that we produce” (ibid. p. 452),
and that the use of one method or the other (i.e. qualitative or quantitative)
does not absolve us from the fact that our understandings are partial and
Mixed-methods research can also be approached differently depending on the
researcher’s methodological standpoints – from a qualitative or a quantitative
standpoint, or somewhere between the two (Hesse-Biber 2010; Johnson,
Onwuegbuzie & Turner 2007). This is important to recognize, since it
influences the way methods are mixed and utilized (how much the research
aims to confirm or explore issues, for example), and since it – if unnoticed –
may introduce difficulties of a paradigmatic kind into the research process.
For example, if, as a researcher, one were positioned in either of the abovementioned paradigmatic “extremes” (e.g. positivism), and in a mixedmethods study used contradictory ontological and epistemological
understandings from another “extreme” (e.g. interpretivism), problems could
obviously arise in the research process. These potential difficulties in a mixedmethods study are clearly something that should be acknowledged. However,
there are paradigms whose underlying elements can be blended more easily
48 For example, researchers following interpretative methodologies have often premiered the study of humans’
lived experiences, while those following transformative and critical methodologies have emphasized, for
example, power relations and questions of social justice in their research. Researchers following positivist and
post-positivist methodologies, on the other hand, have tended to see hypothesis testing and causality as the
main goals of social inquiry (ibid.; see also e.g. Lincoln & Guba 2003).
49 Lawson (1995) provides directions for how a post-structuralist feminist – who commonly uses qualitative
methods – can do quantitative analysis in ways so that it does not “violate” methodological understandings. She
cautions against using inferential statistics, and argues that counting should only be used descriptively for
“carefully contextualized relations” (ibid. p. 454), for instance for describing relations of power, illustrating the
results of the exercise of power, or conducting exploratory analysis to reveal patterns in places.
(e.g. positivism and post-positivism, or realism and subjectivism). Thus, if
one belongs to paradigms that can be blended, the issue of commensurability
– and the mixing of methods – becomes less problematic (Lincoln & Guba
2003). In today’s research practice, neither qualitative nor quantitative
methods can be said to belong to a specific discipline, with empirical material
of a quantitative character occasionally used by qualitative researchers and
vice versa. Additionally, at least within the social sciences, the majority no
longer adheres to conventional positivism, which perhaps makes the task of
combining different methods and approaches less problematic (Johnson &
Onwuegbuzie 2004; Denzin & Lincoln 2003; Teddlie & Tashakkori 2009).
Pragmatism has been advocated in relation to this, and is regarded by several
scholars as the philosophical partner of mixed-methods research (e.g.
Johnson & Onwuegbuzie 2004; Teddlie & Tashakkori 2009; and Morgan
2007). In short, pragmatism is concerned with finding a middle ground
between earlier paradigms (Johnson & Onwuegbuzie 2004), and finding a
workable solution for mixed-methods research. It thereby rejects “dogmatic”
philosophical standpoints in favour of a focus on the search for answers to
research questions (from whatever methodological standpoint, and by
whatever method) (Teddlie & Tashakkori 2009). Pragmatism nevertheless
holds some standpoints of its own. First, it sees the research process as an
inductive-deductive cycle, involving both inductive and deductive
reasoning50. Second, pragmatists question the epistemological dualism of
objectivity and subjectivity (usually stressed by positivists and constructivists,
respectively) and instead emphasize intersubjectivity, which is thus a reflexive
orientation, stressing openness about how knowledge is produced in the
interplay between researcher and research subject, and about the researcher’s
movement between different frames of reference. Third, ontologically,
pragmatists often adhere to the realist view (stating that an external reality
exists independent of our minds) but, at the same time, deny any claims of
truth regarding this reality (hence, they say that there are multiple viewpoints
of social realities). Following this, pragmatists working quantitatively believe
that causal relationships between variables may exist, but that they are
transitory and hard to identify (Teddlie & Tashakkori 2009; Denzin & Lincoln
Besides pragmatism, as mentioned, there are a number of research strategies
today that are used for describing and validating mixed-methods research
(e.g. triangulation, complementarity and bricolage), which makes the mixed50 Qualitative researchers usually employ inductive reasoning, moving from the specific (observations, facts)
to the more general (theory), while quantitative researchers employ deductive reasoning, working from the
general to the particular. Mixed-methods researchers thus employ both these types of reasoning, in an ongoing
process – the inductive-deductive cycle (Teddlie & Tashakkori 2009).
methods process more manageable. Nevertheless, there is still a need for those
conducting mixed-methods research to be conscious of their own stance on
paradigmatic and methodological issues, in order to be able to mix methods
in a research project (Hesse-Biber 2010). At the end of this chapter, I will
return to a discussion on how the mixed-methods approach was implemented
in this thesis, and thereby share some of my own methodological
understandings in relation to the use of mixed methods. The advantages and
disadvantages involved in applying a mixed-methods approach will also be
The fieldwork
The study’s empirical material, consisting of interview and survey data, was
gathered through fieldwork in two settings in Nicaragua. For practical
reasons, the fieldwork was carried out in the form of shorter, repeated visits.
The first four fieldwork periods took place between 2006 and 2008 (one visit
each semester), and the last took place in 2013, seven years after the first visit.
The length of these fieldwork periods varied; three visits lasted for two to three
weeks, and another lasted six weeks. The longest stay, in 2007, lasted three
months. In all, I spent around six months in Nicaragua doing fieldwork for
this study.
During the fieldwork I also acquired knowledge about the research topic and
context through other ways than only the interview and survey studies. The
most important secondary source was naturally academic research and other
literature relevant to the study. I visited Managua to meet with social science
researchers at the university, and with persons involved in a network working
on migration issues (Red Nicaragüense de la sociedad civil para las
migraciones). Furthermore, on one fieldwork trip (in October 2007) I paid a
short visit to Costa Rica, where I met researchers with a great deal of
knowledge about the Nicaraguan-Costa Rican migration process, and who had
also conducted research on the situation of Nicaraguan immigrants in Costa
Rica. I also briefly visited La Carpio, a neighbourhood on the outskirts of San
José where many Nicaraguan immigrants live. In Nicaragua, I also gained
useful insight through doing “nothing”; for instance, talking to people (in
restaurants, parks, offices, hotels, etc.), overhearing conversations, and
watching people in their everyday activities. Other important sources of
information were broadcast news and printed media (mainly the daily
newspapers La Prensa and El Nuevo Diario). This type of fieldwork could be
regarded as similar to “participant observation”, the main ethnographic
research method. However, since I did not explicitly use ethnographic
methodology, I prefer to call the work I did “observant participation”, in line
with Helgesson (2006: 66) (and similar to Tedlock 2003, who speaks of
“observation of participation”). My use of observant participation in this study
entailed being observant of matters of special relevance to my study when
participating in conversations, activities, events, etc.
I acknowledge that fieldwork can be problematic (see e.g. Staeheli & Lawson
1994). First of all, for research purposes the “field” must be delimited (and
thereby constructed). Even though fieldwork “necessarily involves being in a
place and focusing on context and everyday experience” (ibid. p. 98), it is
important to include not only place-based (local) but also nonlocal processes
(i.e. processes on multiple scales) in the conceptualization of the field, so that
it is not portrayed as (for example) fixed and homogeneous, which could
contribute to the perpetuation of power structures and difference (ibid; see
also Katz 1994). Another important question – especially in relation to
fieldwork in which the researcher is an “outsider”, or fieldwork including
marginalized people and groups – is whether it is possible for the researcher
to understand and represent the field correctly. According to Staeheli &
Lawson (1994), this difficulty should not intimidate the researcher from doing
fieldwork, but it necessitates a recognition that what we may know about the
field is partial and situated.
Getting to know the field, and holding test interviews
During my first visit to Nicaragua in October 2006, I focused on getting
acquainted with the people I would be collaborating with during the study, e.g.
researchers and other staff members at CIDS and CHICA (see Chapter 1 for
further details on these organizations). I also spent time getting to know the
two settings where the work would take place. Besides this, I held a couple of
test interviews in León, with assistance from a researcher from CIDS (Wilton
Pérez), as well as one interview in Cuatro Santos51. The purpose behind
conducting these interviews was to see whether the interview questions I had
prepared beforehand in Sweden functioned well in the setting, and served to
give answers to the research questions. Two of the test interviews illuminated
some very interesting aspects, and are therefore part of the material used in
the analysis. However, the major lessons from the test interviews were, first,
that I would choose to do the next interviews without an interview guide
containing pre-formulated questions, as I found it difficult to engage in the
conversation with a piece of paper in my hand; and, second, that I needed to
become much more fluent in Spanish before conducting more interviews,
since I wished to do them alone without an interpreter as I experienced that
this obstructed the “flow” of the conversation.
51 Accompanied by my colleague, Gunnar Malmberg.
The interview study
During my second visit to Nicaragua in March to May 2007, I did most of the
interviews for the qualitative study. These were conducted towards the end of
my stay, when I felt I mastered the language well enough. Still, I must admit
that I sometimes experienced that my skills in Nicaraguan Spanish were
halting, and occasionally caused misunderstanding (this was particularly
noticeable when listening to the recordings). Still, the interviews provided
enough rich empirical material for the qualitative analysis.
Based on the experience from the test interviews, I chose to conduct the
interviews without following a guide. The interviews were therefore more
similar to informal conversations (unstructured interviews), rather than
“question-answer” interviews, and were largely guided by the interviewee’s
wishes or interests. Of course, my ambition was to cover certain themes during
the interviews, but I kept these in my head and tried to direct the interviewees
to these issues as much as possible. Sometimes there was no need for further
direction; in these cases I tried to capture the themes as the interviewee spoke.
Moreover, inspired by the biographical approach (see p. 63), I wanted to
explore the connections between migration and health in the interviewees’ life
histories. All interviews (except for the test interviews) therefore centred
round the biographies of the persons involved. To reach the ambition I had
set, I started the interviews with one broad question, or invitation, which went
something like this:
“I would like you to tell me about your life...where you were born and raised, if you’ve
moved on any occasion. About your educational background and working experience.
About your family: wife/husband/partner, children, parents, brothers, sisters? Where
they were born, if they have moved, about their education and work experience...
About your health situation and the health of your family members... If your/their
health situation has been affected by migration... And so on...”
This inviting question made most respondents quite relaxed, and many told
their story without hesitation. For those who did not start speaking
spontaneously, I further suggested that they begin with when and where they
were born. I let the interviewees speak freely with as few interruptions as
possible, and asked follow-up questions as the interview progressed
(concerning the themes I wanted to cover, and to clarify matters or help the
respondent keep to the subject matter). Even though the interviews were thus
rather unstructured in format, they all covered the same themes (either
spontaneously or through my direction), as listed below:
Background information on the interviewee and his/her family (work,
education, etc.)
Personal experience of migration
Migration of significant others (close family, relatives, friends)
Health – personal and that of significant others (problems, changes
related to migration)
Use of and access to health care
Help between significant others in the person’s social network
The economic and social situation in Nicaragua and in destination
countries (e.g. Costa Rica and the United States)
Recorded interviews were conducted with 15 persons (5 men and 10 women).
Besides these recorded interviews, several other conversations of a more
informal character also took place during the fieldwork; two such talks are
used in the analysis (with Esmeralda and Aleyda) (see Table 1, next page). I
came into contact with the interviewees in many different ways – through
friends, by having met them on the street or at hotels, and occasionally
through selection from the HDSS (see below and Chapter 1, for further
information on the HDSS). The interviews were conducted in several different
places; at workplaces and in homes, at hotels and at offices. The length of the
interviews varied greatly; the shortest was only 30 minutes long, while the
longest lasted over two hours. The number of times I met the interviewees also
varied; some I only met once, whereas others I met on several occasions, over
a long period of time. Hence, I only conducted follow-up interviews with some
of them. On these occasions, I read through the first interview beforehand to
refresh my memory, and also prepared some questions concerning matters I
found especially interesting or confusing. These talks generally added new
information relevant to the study; however, they are not used in the analysis
in the same way as the recorded interviews are.
Upon finalization, the recorded interviews were carefully transcribed. An
assistant, Yamileth Gutiérrez, did the major part of the transcription work,
because I believed a Nicaraguan would be able to do a more accurate
transcription (as I myself was not completely fluent in the language). When
Yamileth had finished her work, I listened through the interviews and
transcribed the parts that were missing in the first transcription (which made
up quite substantial parts of some of the interviews, but in others only
occasional words). The transcriptions can thus be seen as a product of this
collaboration. Even though my colleague did a large part of the transcriptions,
through my part of the work I developed a thorough familiarity with the
material, which in the end amounted to around 230 pages of written text.
The interviewees
Background information and the migration experience of each interviewee are
presented in Table 1 (next page) (all names are pseudonyms and locations are
not exact, in order to ensure the informants’ privacy). As a summary, I can say
that the interviewees were aged between 22 and 60 years, and had various
educational backgrounds and working experience. They originated in both
rural and urban areas, and were presently living either in their place of
origin/birth, in another town, or abroad. Their experiences of internal and
international migration varied a great deal; they were either migrants
themselves, or family members of a migrant (labelled “Left-behind” in the
table). Those who had migrated internationally had experience of both legal
and irregular/undocumented migration (some interviewees had experience of
both types). More thorough descriptions of each interviewee are given on the
following pages.
Table 1: The interviewees
60 years, married, 6 children; no schooling
Farmer; C. Santos
28 years, married, 2 children; secondary*
International migrant
Shop attendant; León
31 years, married, 2 children; secondary
International migrant
Taxi driver; León
24 years, married, 1 child; secondary
International migrant
Painter; Miami, USA/León
24 years, married, 1 child; secondary
Housewife; León
39 years, single, 2 children; university
International migrant
Nurse; León
55 years, married, 5 children; primary
International migrant
Town councillor and farmer; C. Santos
Internal migrant
50 years, single, 5 children, primary
House keeper; León
33 years, single, no children; primary
International migrant
Shop attendant/security guard; León
27 years, single, 3 children; primary
Int. and intern. migrant
Hotel cleaner; León
28 years, married, 2 children; secondary
International migrant
Housewife; C. Santos
22 years, single, no children; no schooling
Internal migrant
Housekeeper; León
33 years, married, 2 children; no schooling
Farmer and small enterpriser; C. Santos
34 years, married, 6 children; primary
Internal migrant
Small enterpriser; León
47 years, married, 6 children, primary
Internal migrant
Small enterpriser; León
24 years, single, 1 child, secondary
International migrant
Unemployed; C. Santos (not recorded)
28 years, single, 1 child, university
Factory supervisor; León (not recorded)
International migrant
Notes: * highest level of education (either completed or not completed).
Gloria. The interview with Gloria, conducted in October 2006 on the
outskirts of a village in Cuatro Santos, is one of the test interviews mentioned
above. Gloria was a 60-year-old woman I met through a research colleague,
who was a leading figure in a development project in the area, as well as
Gloria’s neighbour. Although he stayed in the background during the
interview, his presence might have influenced Gloria to talk about the benefits
of the development project rather than the focus of our interview.
Nevertheless, the interview was interesting and is therefore included in the
study. Gloria and her husband made their living as farmers. They had six
children together; at the time of the interview three of their sons were living
elsewhere. The interview took place on the porch outside Gloria’s house and
lasted about half an hour. During the interview there was a thunderstorm with
heavy rain, and a great deal of chickens, pigs and dogs were running around
our feet. I met Gloria once more in 2008, a year and a half after our first
interview. By that time two of her sons had returned home.
Sandra. The interview with Sandra is the other test interview included in the
study’s qualitative material. It was conducted in León in 2006 with assistance
from Wilton Pérez, who helped during the interview with translation when
necessary. Sandra was included in the HDSS, and we looked her up after
having selected her for an interview since she had lived in Guatemala for
several years. Sandra was born and raised in Chinandega, but had been living
in León for a while. The interview took place in the house where she worked
as a shop attendant, and where she also lived with her two children. Her eldest
son had lived with Sandra’s mother in Chinandega while Sandra was in
Guatemala. Sandra’s husband presently lived in the US, but she herself had no
plans to leave León.
Cesar. The interview with Cesar took place in May 2007 (as did the rest of
the interviews). Cesar was a married man in his 30s, and the father of two
small children. He worked as a taxi driver in León, and when he drove me
across town one day we started chatting in the car. I learnt that he had worked
abroad for several years, mostly in Costa Rica, so I asked if he wouldn’t mind
sharing his story with me and my supervisor, Aina Tollefsen, who was there
on a short visit. The interview, which lasted an hour and a half, took place in
the hallway of a hotel where we all knew some of the staff and felt comfortable.
I met Cesar on all my visits to León. In 2013, the last time we met, Cesar was
working at a retail company just outside León. In 2014, I heard that he went
to Panama to work.
Juliano. I met Juliano through my local Spanish teacher. He was 24 years
old and married to Cindy (see below), with whom he had one child. The
interview took place at his sister’s house in León, where he was spending time
during a visit from the US, where he had lived for four years at that point.
Juliano was a US resident, thanks to a petition by his father (who had earlier
been granted residency thanks to his mother, who had come to the US in the
1980s). To avoid the afternoon heat during the interview, we sat in rocking
chairs just outside the entrance to the house; a fan was on to blow away the
persistent flies. Just across the street some kids were playing basketball, and
the next-door neighbours did their best to drown out the twitter of birds by
playing reggaeton at a high volume. We talked for a little less than an hour.
Cindy. I interviewed Juliano’s wife Cindy two days after my talk with Juliano,
on their last day together during his visit, as Juliano was leaving for the US
the next day. Cindy was 24 years old, and had a six-year-old son with Juliano.
Cindy had been a stay-at-home mom and a part-time architecture student ever
since Juliano left for the US. The interview took place at Juliano’s sister’s
house, and lasted about 45 minutes.
Maribel. Maribel was a 39-year-old nurse who lived in León. She had
separated from her husband ten years earlier, and since then had supported
herself and her two children by working in both Nicaragua and Costa Rica. At
the time of the interview, she was working for a non-governmental
organization in León. I met Maribel through a mutual friend who worked at
CIDS, which is also where the interview took place, in a conference room. It
lasted about an hour and a half. In 2013, I learnt from our mutual friend that
Maribel was still living and working in León.
Fernando. I interviewed Fernando when I was out with the CHICA survey
team on a follow-up trip around the municipalities in Cuatro Santos. I selected
Fernando from the HDSS because he seemed to have interesting migration
experience. Fernando was in his 50s and worked as a member of the town
council, as well as a farmer. He lived in a small town, the city centre of the
municipality, with his two youngest children; his three eldest children lived in
other parts of the country and supported themselves. For many years,
Fernando and his wife had run a small shop in their house. After they had been
forced to close the shop due to low revenue, Fernando had gone to the US for
six months to work. At the time of the interview, his wife was working in Spain.
During our talk, we sat on white plastic chairs in the parlour, facing the open
front door. A great deal of noise welled in from the street – from cars, horse
carriages, construction work, roosters, and music. The interview lasted about
45 minutes.
Marta. During my longest stay in León I shared a kitchen with a family who
had a housekeeper by the name of Marta. As time went by, we got to know
each other through our chats by the stove. Marta was willing to share her life
story with me, so late one night when everyone else was out we sat down in
the living room and talked for about an hour. Marta was 50 years old, and was
born in the countryside outside León. Upon separating from her first husband
when she was in her 20s, she had gone to León with her two small children to
look for work. After some time she met a new man, with whom she had three
more children. Twenty-three years later, Marta’s husband decided to go to
Costa Rica to look for work and she started working again, as a housekeeper.
I visited Marta on each of my visits to Nicaragua, and talked about how life
had evolved since my last visit. In 2008, as well as 2013, she was living and
working in the same house as before. Life was pretty much the same, and on
both occasions she said she was rather pleased with life.
Santos. I met Santos by chance on a visit to a shop in León. When I told him
about my work, he immediately mentioned that he had many experiences of
migration that he would gladly share with me. The interview took place at my
house late one night, a couple days after our first meeting, and lasted almost
two hours. Santos was 33 years old and had made a living mainly from lowskilled, short-term, low-paid jobs both in León and in the countryside. He had
made three attempts to go abroad (to the US and to Costa Rica) in order to
look for work. Santos was a great narrator, speaking with sincerity and
expressing a great deal of feeling. As I will come back to later, the interview
seemed to have a therapeutic meaning for him since it gave him the chance to
put into words some very difficult experiences in his life. I saw Santos from
time to time during my stays in León. In 2008, he was still working in the same
shop, although with other tasks.
Rosa. Rosa was working as a cleaner at a hotel I frequently visited during my
stays in León. Her working schedule was quite tough, and because of this the
interview had to take place in a hotel room after a long day’s work. Despite
this, Rosa talked unceasingly for over two hours about her life, which had been
very eventful even though she was only 27 years old. Rosa had moved many
times since her childhood, both within Nicaragua and abroad. At the time of
the interview, her three children were living with her mother in another town
quite far away, while Rosa stayed at the hotel where she worked. When I saw
Rosa a year later, in 2008, her life situation was practically the same. In 2013
she was still working and living at the same hotel, and her children were still
living with her mother, although a bit closer to León.
Joanna. I interviewed Joanna in her home, in connection with a follow-up
of the survey in Cuatro Santos. She was 28 years old, and lived with her two
children in the outer parts of a small town that lay surrounded by a stunningly
beautiful, lush and mountainous landscape. She had recently returned to her
birthplace after spending seven years working in Guatemala. Her husband
was working as a truck driver, travelling around Central America. The
interview was rather short, just over half an hour, because of Joanna’s
toddler’s need for attention.
Ana. Twenty-two-year-old Ana came from a small rural community “in the
mountains” – as she said – far from León, where she was residing at the time
of the interview. She had left home at 15, and since then had made a living on
her own, mainly through working as a live-in maid. I met Ana through her
current employers, whom I knew from work. Our talk took place in their living
room on a hot afternoon, and lasted about an hour and a half. The children
who lived in the house sometimes passed by, on their way out or to the kitchen,
but Ana didn’t seem intimidated by this, except that she lowered her voice
when talking about delicate matters. In 2013, I learnt from our mutual friend
that Ana had moved back to the “mountains” where she came from.
Carmen. I met Carmen in connection with a survey follow-up in Cuatro
Santos. Carmen was 33 years old, and lived with her two children at her
stepmother’s house, in the outer parts of a small town in Cuatro Santos. She
and the family made their living from farming and animal-keeping, as well as
from making and selling handicrafts made of pine. At the time of the
interview, Carmen’s husband had been working in the US for a year. Carmen’s
stepmother, “Aurora”, was also present during the interview and sometimes
participated in our discussion. This didn’t seem to affect Carmen negatively;
on the contrary, it seemed as if her presence made Carmen a bit more
comfortable with the interview situation. During the interview we sat outside
the house; occasionally, a neighbour, a dog, or a pig passed by on the trail next
to where we were sitting. The coffee and cake I’d been served were invaded by
small, red ants when I put them down on the bench next to me. The interview
lasted well over an hour.
Mercedes and Orlando. Thirty-four-year-old Mercedes was a beautician I
got in touch with through a mutual friend who was one of her clients.
Mercedes’ husband Orlando, aged 47, also took part in our discussions since
he was sitting just beside us during the interview, working in his artisan
workshop, which took up half of the house. They both openly shared their life
histories with me during the two-hour interview. Mercedes and Orlando were
born in two small neighbouring villages north of León, but had in different
ways and at different ages left their places of origin and ended up in León.
When Mercedes and Orlando met, he had an artisanry business and she soon
moved in and started working with him. Over the years they had six children.
Due to unfortunate circumstances, a couple of years prior to our interview they
had moved to a less well-off neighbourhood in León. Orlando still worked with
his artisanry, and Mercedes received clients in their home. I visited Mercedes
and Orlando once more in 2008. At that time they said life had become worse,
and that they were thinking of going abroad to look for work. In 2013, I learnt
from our mutual friend that Mercedes and Orlando had separated when
Mercedes was expecting their seventh child. She and her children were now
living with a new man, and she was still working with her beauty business.
Esmeralda. In a rather small house with clay brick walls and an earthen
floor in a rural part of Cuatro Santos, lived Esmeralda, 24 years old, with her
year-old son and 11 other family members. I got in touch with her during a
follow-up tour of the survey in the area, and asked if I could sit down for a talk,
which lasted about 45 minutes. Because of the family’s poverty, Esmeralda
had been obliged to go abroad and look for work after graduating from
secondary school. She had therefore followed her brother, who was in El
Salvador, and worked there for three years as a maid and a clothing vendor.
She had returned to her birthplace about a year before our interview, when it
was time to give birth to her child. At the time of the interview, Esmeralda did
not mention any plans to return to El Salvador, or to go anywhere else.
Aleyda. In a restaurant in León one day I started talking to Aleyda and her
three-year-old daughter, who were sitting beside me. As it turned out, they
lived next to the hotel where I was staying, and later that evening when I was
passing by they invited me in for a longer chat with the family. Aleyda, who
was a pharmacist, had been commuting on a weekly basis for the last three
years to her job in Chinandega, while her daughter stayed in León with
Aleyda’s mother and the rest of the family. Two of her brothers lived abroad
(in Costa Rica and the Netherlands). Aleyda was also thinking of going abroad,
to the US where one of her closest friends lived. I learnt in 2008 that she had
gone through with this plan. She had been approved for a tourist visa to the
US, thanks to her Nicaraguan friend who lived there, which she overstayed in
order to look for work. I saw her when she returned to León for a visit;
thereafter, she returned once more to the US.
The interview situation
In my experience, interviewing is a work that evolves over time, as a process.
The first interviews are often rather tentative and almost stumbling, whereas
later ones more take on the form of informal conversations. Each interview is
unique since the interviewees are different people with different experiences,
and the information provided by each interview is consequently also always
unique. I therefore believe it is fruitless to ask exactly the same questions, and
to try to cover exactly the same issues in all interviews (for instance by
following an interview guide). Instead, I find it necessary to adapt the
questions, and the way they are posed, to each new person, and to follow his
or her personality and interests. Like in similar research situations (other
types of interviews, as well as in surveys), the interviewees in this study
constructed their answers, narratives and life stories so as to fit to the
interview situation (Arvidsson 1998; Riessman 2008). Therefore, there were
almost surely issues that were omitted, as well as emphasized, and the
interviewees most probably made an effort to picture themselves as “good”
people. Although there thus are variances between the interviews regarding
the information they provide, each interview offered new, different and
specific insights that added to the analytical process. To give just one example,
the most important insight from the interview with Gloria was how
emotionally affected she was by her sons’ absence from home due to their
work abroad and in other parts of Nicaragua.
I tried to make the interview situation as comfortable as possible so that the
interviewees would feel they could trust and confide in me. In accordance with
other qualitative health researchers (e.g. Hewitt 2007; De Haene et al. 2010;
Orb et al. 2001; Paavilainen et al. 2014; and Siriwardhana et al. 2013), I believe
it is particularly important to pay attention to ethical aspects in research on
sensitive, potentially delicate issues (e.g. health), as well as in studies
including so-called vulnerable groups (e.g. migrants). Besides acquiring
ethical approval from the medical faculty at León University (UNAN-León,
Comité de Ética para Investigaciones Biomédicas, ACTA no. 15, March 28
2008), I also took several precautions to ensure ethical conduct, as well as to
instil a feeling of informality and comfort during the interviews. First of all, I
informed the interviewees of the purpose of my study, and what themes I
wished to discuss with them. I also emphasized that their participation was
completely voluntary and that their privacy would not be adventured. The
interviews were conducted at places and times according to the interviewees’
wishes, and before and during the interview I stressed that they could decline
to answer my questions or end the interview whenever they wanted.
Moreover, I listened carefully, with empathy and respect, throughout the
interviewee’s story. Instead of taking notes I used a small, almost unnoticeable
MP3 player to record the interview, which I of course asked for permission to
do before starting the interview.
Even though I did my very best to ensure a sense of openness, respect, trust,
and confidence, there might have been moments when my position as a welleducated, well-off, young, white European woman affected the interview and
the interviewee in ways I did not intend. Power relations are inherent in all
social relations, even in interview situations. By letting empathy and respect
characterize these situations and relations, however, there is a greater chance
that such power relations will not become intimidating (Charmaz, 2003;
Schwandt, 2003). Just as my social position may have affected the interviews,
I also believe my personality may have had some impact. For instance, my
rather emotional way of acting might have influenced the interviewees to also
express emotion. Even though I mostly felt very fortunate that the
interviewees trusted me enough to confide in me and talk to me about
sometimes very delicate matters, this also led to a closeness that at times
became difficult to handle. Particularly the interview with one male
interviewee, Santos, was very “heavy”, as his narrative was filled with grim
events, and as he also gave a depressed impression and even talked about
being suicidal (see McGarry 2010, on the emotional effect of qualitative
research on sensitive issues). As Santos pointed out afterwards, the interview
had been a way for him to talk about his hardships for the first time, and,
luckily, he felt that our talk had made him feel a bit better, as he had had the
chance to put his experiences into words. During as well as after the interview,
I also experienced that it might have had a therapeutic meaning for him.
“Storytelling”, in both therapeutic and research contexts, may in fact have a
healing effect, as I will return to in Chapter 6 (see e.g. Pennebaker 1995;
Pennebaker & Seagal 1999; and Rosenthal 2003). My relationship with Santos
became somewhat strained after the interview; he was very persistent when
we occasionally met, almost demanding that we should meet and that I should
help him in various ways (for example financially). This could have affected
the analysis of this interview – though I don’t believe it has – which is why I
discuss it here. I also find it very interesting from a methodological point of
view; it serves as an example of the importance of the relationship between
researcher and informant in qualitative research and how this relationship, in
itself, may provide empirical material that can be used in the analysis and for
answering the research question.
Qualitative research approaches and methods of analysis: the
biographical approach and constructivist grounded theory
The interview study includes a combination of two different qualitative
research approaches. The interviews were carried out based on the
biographical approach, as described below. The analysis of the interview
material was then performed in two ways – by use of constructivist grounded
theory, and by means of the biographical approach, through which the
constructivist grounded theory analysis could be situated in time, as part of
the individual’s life course.
“[M]igration exists as a part of our past, our present and our future;
as part of our biography” (Halfacree and Boyle, 1993: 337)
Biographical research seeks to understand individuals’ life experiences in
their daily lives, as placed within the contemporary cultural and structural
context (Roberts 2002). The empirical material analysed within biographical
research – for example biographies, narratives, life histories, and life stories
– focuses on the stories of individuals, and on seeking to understand the
individual’s life within its social context. The approach is thereby a way to
place and understand societal changes, through the individual’s experiences
and interpretations. According to Arvidsson (1998), biographical interviews –
in their strictest form – start with the interviewee’s birth or origin, and move
along in time until the present. They may, however, also take on other forms,
for example that of interviews starting with a biographical description, which
is then used as a background when focusing on a certain societal phenomenon
or historical epoch. Another form of biographical/life history interviews are
those that start with a short chronological biographical description, and
thereafter focus on specific periods in time or themes, though still as part of
the life history. In migration research, the biographical approach has been
used for many decades (see Skeldon 1995 for an overview). A starting point in
this research is that migrants (just as other human beings) are socially
embedded, and that they influence, and are influenced by, the social world
that surrounds them throughout the life course (Findlay & Li 1997). In
biographical migration research, migration is firstly understood as an “action
in time” (Halfacree & Boyle 1993: 337) rather than a discrete act at a particular
point in time. The time aspect is thus crucial, and an individual’s decision to
migrate is believed to be situated in his/her entire life history. Moreover, the
approach also emphasizes the “rootedness of the migration in everyday life”
(Halfacree & Boyle 1993: 339). Hence, migration events are believed to be
connected to many different dimensions of life, and the decision to move is
seen, for example, as related to a variety of factors (e.g. the economic situation,
family relations and health concerns). Additionally, migration is regarded as
“a manifestation of an individual’s identity” (Findlay & Li 1997: 34) and is thus
understood as a highly cultural event (Fielding 1992) that reconfigures social
identity. For example, migration concerns and affects the worldviews, values
and attachments of both migrants and their families, as well as of the societies
of origin and of destination. By means of the biographical approach, the
complexity surrounding migration can thus be highlighted, making it possible
to gain “insights into the dynamics that shape the actions of individual
migrants during their life course” (Tollefsen Altamirano 2000: 17). Hence, it
is a way to get “glimpses into the lived interior of migration processes”
(Benmayor & Skotnes 1994: 14). I was highly inspired by the biographical
approach in the interview study, and the interviews can be described as a mix
of the biographical interviews outlined by Arvidsson (1998). They all started
with biographical descriptions, but then focused not only on the themes under
investigation (migration and health, social relations, etc.) but also on specific
periods in time, historical epochs, and societal phenomena. The interviews
always returned to the person’s life history when it was sidestepped; it thus
served as a background for the interviewee’s descriptions of the themes,
periods, or phenomena. Not all interviews captured the whole life history of
the interviewee; in these cases, only specific parts of the person’s life course
were covered in the interview (see Tollefsen Altamirano 2000 for a similar use
of the biographical approach).
The other qualitative strategy used in the thesis is constructivist grounded
theory, which stems from the grounded theory (GT) method advanced by
Glaser and Strauss during the 1960s and further developed by Strauss and
Corbin during the 1980s and 90s (Charmaz 2003). The ambition of GT is often
to create or develop a new formal theory from empirical material that is
“grounded” in a natural setting, or, in more modest phrasing, to “generat[e]
new ideas, categories or perspectives that can shed new light on a
phenomenon” (Tollefsen Altamirano 2000: 15). GT findings are furthermore
meant to generalize across cases, in contrast to narrative approaches
(Riessman 2008). There are several guidelines for “doing” grounded theory.
Some of the most common strategies are: (i) a simultaneous collection and
analysis of data (mostly interviews), (ii) data coding processes, (iii) memo
writing for constructing conceptual analyses, (iv) sampling for the refinement
of emerging theoretical ideas, and (v) integration of the theoretical framework
(Charmaz 2003). The coding process usually consists of coding the data in, for
instance, actions, events, processes, experiences, and meanings (initially lineby-line or using larger pieces of the text, and later selective/focused coding,
which is more conceptual). After this foundational work the process of
synthesizing and explaining the data follows, through categorizing the codes
(categories often subsume several codes). According to Kathy Charmaz
(2003), one of the purposes behind Glaser and Strauss’ work was to
systematize qualitative research in order to make it legitimate in the eyes of
quantitative researchers who dominated the social sciences at the time. After
criticism from post-modernist and post-structuralist qualitative researchers,
Strauss and Corbin therefore (in a post-positivist spirit) started emphasizing
the importance of “giving voice” to the respondent, and of recognizing the
respondent’s own view of reality (and that this reality can differ from that of
the researcher). Charmaz (e.g. 2003) has taken the method further from its
positivist/objectivist connotations, and argues for a constructivist grounded
theory (henceforth also called CGT), since she, in contrast to original
grounded theorists, sees data (e.g. interview data) as “narrative constructions”
or “reconstructions of experience”, and “not the original experience itself”
(Charmaz 2003: 258). Consequently, says Charmaz, a constructivist working
with GT sees the study findings as a product of the interactions between
researcher and research subject. Since the interviews, and the results
stemming from the analysis, are created in that specific context, they should
therefore be regarded as “negotiated accomplishments” (Fontana & Frey
2003). Furthermore, constructivist grounded theorists “aim to include
multiple voices, views, and visions in their rendering of lived experience”
(Charmaz 2003: 275). The researcher should thus include both the
researcher’s and the research subject’s meanings of lived experience in the
analysis, instead of merely delivering the researcher’s own version of what has
been said or has happened. In relation to the research process, constructivist
grounded theorists moreover try to go “inside” the experience of the research
subject and thereby find active codes – describing what the subject is doing
and/or what is happening – that can later form categories that approximate
the images of experience. In contrast to the original GT method, which
commonly focuses on the researcher’s interpretations of what informants say
or do, CGT thus tries to come closer to the meaning of lived experience, and
leads to a text in which “[t]heory remains embedded in the narrative, in its
many stories”, and “readers might sense and situate the feeling of the research
subject” (ibid. pp. 278, 280).
Qualitative analysis in practice
I would say that the analysis of the interviews started already while I was
conducting them. During the interview, I got a sense of what the most
significant themes concerning migration and health were. I could also sense
the interviewee’s feelings about what was said. I kept these impressions with
me during the continued research process, and they later informed the
analysis. Directly after an interview, I wrote in a field diary about the interview
and the information it had provided. I also continued taking notes while I was
working with the transcriptions, and when I later read the transcribed
interviews while listening to them. During the second reading of the
interviews I initiated the CGT-coding process first coding the interviews word
by word, then line by line, sentence by sentence, and paragraph by paragraph.
I used both in vivo codes, i.e. the person’s own word(s), and analytical codes
based on my own interpretations and ideas. These codes were then grouped
under more abstract codes (axial codes) based on the patterns or themes that
emerged from comparing the shared characteristics and meanings of the
initial codes. The abstract codes were thereafter grouped into categories, or
themes, that integrated a substantial number of the codes. Most of the coding
was done by means of paper and pen, but part of it was also done in a computer
programme for the analysis of qualitative material (MAXQDA10). The
themes/categories were changed and refined many times during the analytical
work, even though their content (the codes) remained the same. The first,
rather tentative, categories included for example “moving abroad or to town”,
“life in a new country or town”, “health problems or improvements”, and
“social relations and support”. Some time later, I read the interviews once
more and went through the themes and their connected codes. I then added
the category “making a living”, and split the theme about health into two
themes – “mental health/emotions” and “physical health”. At this point I also
performed a selective coding of feelings/emotions since this emerged as an
important theme in all interviews. When I began writing the empirical
chapters, after more months, I looked through the material again and decided
to change the labels of the categories once more, having found a new and more
appropriate/convenient division of the material. My closest colleagues also
took part in the analytical process, through reading and discussing the
interviews. During the whole process I also studied different theoretical
perspectives and previous studies on the research topic. The analysis can
therefore be seen as an abductive process whereby I switched between theory
and data. The final results of the analytical process can be seen in the themes,
categories and narratives presented in the empirical chapters.
In the analysis, I used CGT to analyse the “whole picture” of migration-health
relations in the interview material. CGT was indeed a fruitful way to get an
overview of these relations; however, even though I made use of the
constructivist version of GT, I felt that the complexity in people’s lives was
lost. I thus experienced a need to complement the GT analysis with a more
narrative analytical strategy, and therefore decided to analyse and present the
results in light of the interviewees’ biographies, by use of the biographical
approach. In this way, the interview narratives could be seen as part of the
interviewees’ past, present and future.
The two-step survey study
At the same time as I was conducting the interviews, I was working with the
research teams at CIDS and CHICA in preparation for the survey, which would
be carried out in two steps. Before going into detail about the survey study, I
will first provide more background information about the Health and
Demographic Surveillance System (HDSS) in the two study settings.
The HDSS in León and Cuatro Santos
The HDSS in León dates back to the early 1990s, when two cross-sectional
household surveys were conducted (in 1993 and 1996) as part of a child and
reproductive health project. The study population was selected by use of a
cluster sampling framework, in which 50 urban and rural residential areas
were selected52, covering a total of 43,765 individuals residing in 7,789
households (which represented 22% of the total population of León). An
epidemiological database was constructed, and in 2002-2003 this database
was updated; this time also including new sample clusters in the rural zone.
By then, the HDSS monitored 54,647 persons living in 10,994 households,
which equalled 30% of the total population in León. Two more follow-ups of
52 A cluster was defined as a geographical area with a population of 700-1,000 inhabitants. At the time, 208
clusters were identified in León, of which 50 clusters were randomly selected for the study (Peña et al. 2005,
2008; Pérez 2012).
the baseline in León have been conducted: the first in 2005, and the second at
the same time as our survey, in 2008 (Peña et al. 2005, 2008; Pérez 2012).
The HDSS population at the time of our survey consisted of around 56,000
individuals, residing in over 13,000 households (31% of the total population
in León municipality) (see Figures 2 and 3; design: Margarita Chévez, CIDS).
Figure 2: Study areas in León municipality, 2006.
Rural areas in grey, urban in black.
Figure 3: Study areas in urban León, 2006. In grey.
In Cuatro Santos, the work with the HDSS was initiated in 2004. Since the
beginning it has covered all residents in the area, which in 2007 amounted to
24,568 persons residing in 4,828 households53. Figure 4 shows the dispersal
of houses in the study area’s four municipalities (map design: CIDS).
Figure 4: The HDSS in Cuatro Santos, 2005.
Houses marked as black dots.
The major bulk of data collected since the beginning within the frame of both
surveillance systems mainly concerns three categories: a) socio-economic
background information, e.g. age, sex, education, occupation, household
composition, housing conditions, poverty levels; b) vital/demographic
events, i.e. births, deaths, in- and out-migration; and c) reproductive health
and child health, e.g. obstetric histories of women of reproductive age,
indicating child survival. The HDSS data are commonly collected on the
household level, by means of interviews with the head of household, or if
he/she is unattainable, anyone at home over the age of 16. The information on
reproductive health is, by contrast, collected directly from the woman herself
(Peña et al. 2008; Zelaya Blandón et al. 2008; Pérez 2012). (A more detailed
description of the survey procedure and the HDSS data will be presented
53 According to government statistics the area’s population was somewhat larger in 2007, with 26,765
persons (INIDE 2007a).
Survey step 1: singling out individuals
We had thus been given the opportunity to conduct our survey within the
frames of the HDSS in both areas. However, since the HDSS is based on
household data, and as we wished to direct our questions to individuals for the
purposes of this study, it was necessary to design the survey in two steps – the
first to single out individuals who could constitute a sample, and the second
to gather information from the individuals in this sample (see Figure 5, p. 75,
for an overview of the two-step survey).
The first step of the survey was carried out during my fieldwork in spring
2007. Since the HDSS in Cuatro Santos was about to be updated, and a substudy would take place in León, we could include a set of additional questions
along with those from the original HDSS questionnaires (concerning health
status, use of health services and medicine, and reception of remittances),
which would give us the necessary information to make a sample for the
second step of the survey. The specific questions posed in the first step were:
Has anyone in the household been sick in the past three months?
In what way (what disease/illness)?
Did he/she visit any health care services?
Did he/she use any type of medicine (occidental and/or traditional)?
Has anyone in the household received any remittances in the past
three months?
In Cuatro Santos, all households – approximately 4,800 – were asked these
additional questions. In León, we selected every fifth household in the urban
sector – amounting to 2,500 households, of the total of 13,000 households
included in the HDSS – primarily due to limited resources, but also because
the material would have been too extensive otherwise. Following the same
working procedure as that of the original HDSS surveys, the additional
questions were asked of one person in each household (most often the head of
household), but covered information about all members of the household.
Through the additional questions, we retrieved individual information from a
total of 40,313 men and women – 13,171 in León (24% of the HDSS
population) and 27,142 in Cuatro Santos (all inhabitants) – which thus made
up the study population for our survey (see Table 2, p. 72). Based on the
information from Step 1, we could distinguish between individuals who had
been sick or healthy, and what illness those with health problems had
experienced (categorized as acute, chronic or other). We could also say
something about the use of health care services and medicines, and of the inflow of remittances in the study areas.
Survey step 2: construction of sample and questionnaire
On my third visit to Nicaragua in October 2007, the work of constructing a
sample for the second step of the survey continued. At this moment in the
process, we connected the data gathered during the first step to other HDSS
data available in the databases at CIDS and CHICA; most importantly
migration data (place of residence, and in- and out-migration events) but also
household characteristics (family constitution, poverty, education,
employment). The sample drawn for the second step of the survey was
nevertheless based only on migration data (besides age, and the data
concerning health problems from our initial questions).
The study population and the sample frame
Before making the sample, we first excluded those who were younger than 17
years (since the study was to focus on adults). The study population was
successively categorized according to migration experience (Non-mover/Leftbehind/In-migrant54) and individual health status (Healthy/Chronic
ill/Other ill55). The categories were strictly defined, in the sense that a person
could not share characteristics with someone in another category of the same
type (e.g. both Non-mover and Left-behind). Through combining these
migration and health categories, nine different sample groups56 were created.
The sample frame that was ultimately created consisted of 19,058 women and
men (47% of the respondents of our 2007 survey) (Table 2, next page). A total
of 5,350 individuals resided in León (equalling 10% of the HDSS population,
and 41% of the 2007 respondents), and 13,708 lived in Cuatro Santos (51% of
the HDSS population). As seen in the table below, most individuals in the
sample frame were Non-movers (sample groups 1 a-c) or Left-behinds
(sample groups 2 a-c). A small share were In-migrants (sample groups 3 a-c),
of which the majority lived in Cuatro Santos. Moreover, those classified as
54 Non-mover: person who still lived in his/her place of birth; no migration history in the family; Left-behind:
family member of out-migrant; no personal migration history; In-migrant: person who had moved into the
household from another place, no record of out-migration.
55 Healthy: person who reported no health problems; Chronic ill: person who reported at least one chronic
health problem; Other ill: person who reported at least one acute/other health problem (not chronic).
56 The nine sample groups were thus the following:
1a. Non-mover, healthy; 1b. Non-mover, chronic ill; 1c. Non-mover, other ill.
2a. Left-behind, healthy; 2b. Left-behind, chronic ill; 2c. Left-behind, other ill.
3a. In-migrant, healthy; 3b. In-migrant, chronic ill; 3c. In-migrant, other ill.
Healthy far outnumbered those who were categoriezed as Chronic ill or Other
ill; especially the Chronic ill were few.
Table 2: The study population and sample frame
Survey 2007 (Step 1)b
Sample framec
Cuatro Santos
13,171 (24)
5,350 (41)
27,142 (100)
13,708 (51)
40,313 (50)
19,058 (47)
3,527 (66)
177 (3)
585 (11)
811 (15)
27 (0.5)
61 (1)
145 (3)
5 (0)
12 (0)
5,948 (43)
420 (3)
1,139 (8)
4,696 (34)
248 (2)
466 (3)
698 (5)
15 (0)
78 (0)
9,475 (50)
597 (3)
1,724 (9)
5,507 (29)
275 (1)
527 (3)
843 (4)
20 (0)
90 (0)
Sample groupd
1a. Non-mover, healthy
1b. Non-mover, chronic ill
1c. Non-mover, other ill
2a. Left-behind, healthy
2b. Left-behind, chronic ill
2c. Left-behind, other ill
3a. In-migrant, healthy
3b. In-migrant, chronic ill
3c. In-migrant, other ill
Notes: a) individuals monitored in the Health and Demographic Surveillance Systems (HDSS);
b) participants in Step 1 of our survey (2007); c) individuals in the sample frame; d) sampled
individuals according to sample group; 1) year 2008; 2) year 2007 (the number of inhabitants
was, as seen in the table, larger when we conducted the survey). In absolute numbers and
percentages (in parentheses).
The final sample
Due to time and budget restrictions, we limited our sample to 250 individuals
per sample group (125 per study setting). However, since some sample groups
did not include such a high number of individuals (e.g. group 2b in León, and
3b in both study settings; see Table 2), we could not select the sample in these
groups randomly. Consequently, all individuals in these groups were selected
so as to make the groups as large as possible, and as similar in size as the
others (and, large enough to provide sound/significant results in the
analyses). Furthermore, in order for the group of In-migrants to be large
enough in León, we had to over-sample those in-migrants categorized as
Healthy (sample group 3a; see Table 2). In the forthcoming analyses we have
weighted the sample groups according to their original sizes, in order not to
make the results skewed (see Table 6, p. 80, for the calculations of weights).
Our sample ultimately amounted to 1,718 individuals57 (Table 3). The
distribution, in terms of responses/non-responses, place of residence, sex and
sample group, is presented in the table below.
Table 3: The sample.
Total sample
Sample size
Sample group
1. Non-mover
a) healthy
b) chronic ill
c) other ill
2. Left-behind
a) healthy
b) chronic ill
c) other ill
3. In-migrant
a) healthy
b) chronic ill
c) other ill
“Health status” group
Healthy (1a, 2a, 3a)
Chronic ill (1b, 2b, 3b)
Other ill (1c, 2c, 3c)
Cuatro Santos
Notes: In absolute numbers and percentages.
As seen, 335 of the sampled individuals did not participate in the survey; the
respondents thus amounted to 1,383 persons, and the non-response rate
equalled 19%. The sizes of the sub-groups in the sample varied, ranging from
22% to 45%58. There were more individuals in the categories Non-mover
(44%) and Healthy (45%), and fewer in the categories In-migrant and
Chronic ill (22% and 24%, respectively). The shares of Left-behind and Other
ill were between 31% and 34% (thus close to the expected value of 33%). There
were, furthermore, differences in the sample regarding the two study areas. A
larger share of the sampled individuals lived in Cuatro Santos (56%). The León
sample included more individuals in the categories Non-mover and Healthy,
whereas the sample in Cuatro Santos included more in Left-behind, Chronic
57 The total number of household members whose background data were included in the sample construction,
and who therefore serve as a base in the forthcoming analyses, amounted to 7,188.
58 For equal distribution, each group should be 33.3%.
ill and Other ill. There was also a slight gender bias in the sample (58-59%
were women).
The questionnaire
The work of constructing the questionnaire for the survey in Step 2 began
during my third visit to Nicaragua, with a first draft based on findings from
the qualitative interviews and on other previous research, and then continued
at home in Sweden with discussions with research colleagues at the Division
of Epidemiology and Public Health at Umeå University. During the fourth
period of fieldwork, in March-April 2008, much time was initially devoted to
discussing the content of the questionnaire with researchers and other staff
members at CIDS and CHICA, who then tested it on family members and
friends. I also personally tested it on a couple of people in my surroundings.
Various changes were made before the questionnaire was finally piloted in the
HDSS population, using approximately 15-20 people in León. In the end, the
questionnaire included a mix of questions stemming from theories, previous
studies, and results from the interview study, concerning the following issues
(see Appendix for the full version):
Migration (personal experience of migration, and migration of
significant others)
Self-rated and self-reported health
Use of health services and medicine
Access to social security
Social support (material and emotional help from/to significant
others within the person’s social network)
Social integration and participation
Life situation and legal status of significant others living abroad
After the pilot was finished, along with the fieldwork teams at CIDS and
CHICA, respectively, I drafted lists of the names and addresses of the selected
individuals, which would be used during the fieldwork. With help from CIDS’
GIS technician, Margarita Chévez, maps of the León area were designed,
marking the selected houses. The fieldworkers were also recruited at this
point, based on their earlier experience working with the demographic and
health surveys. Four women in León and 12 women in Cuatro Santos were
employed, and divided into teams with appointed work areas. Just before the
fieldwork started I held a one- or two-day workshop with the fieldworkers at
both locations, along with those in charge of the fieldwork (Aleyda Fuentes at
CIDS and Francisca Trujillo at CHICA), so that the fieldworkers would
understand the purpose of the study, and could practice how to ask the
questions and fill out the questionnaire. When this was done, I had at long last
received ethical clearance from León University (see p. 62); the survey round
could finally start.
The survey procedure
Our survey followed the same, well-developed procedure as all HDSS surveys
generally do. The questionnaires were conducted face-to-face between the
fieldworkers and the respondents. In the first step of our study, just as in other
HDSS rounds, the respondent was typically the head of household, or, if
he/she was absent, anyone in the household aged over 16 years. In the second
part of our study, the respondent was the individual selected in the sampling
process. The fieldworkers were led, and coached by, a team supervisor
responsible for the data collection. Shortly after the questionnaires had been
finalized they were handed over to those in charge at both offices, who then
controlled the quality. Every week, one or two questionnaires per fieldworker
were randomly chosen for an additional quality control, in which the office
staff returned to the selected household to re-do the questionnaire and
thereafter compare the answers to check whether any mistakes had been
made. The information from the questionnaires was then entered in databases
by the data units at both offices. Lastly, the questionnaires were filed and
stored at both offices (see Peña et al. 2008, for further information about the
working procedure). Of the 1,718 sampled individuals, 1,383 completed the
questionnaire: 572 in León and 811 in Cuatro Santos. The whole survey
procedure is visualized in Figure 5, below.
HDSS, 2007
L: 13,000 h
56,000 i
CS: 4,828 h
24,568 i
Step 1, 2007
Step 2, 2008
L: 2,500 h
13,171 i
CS: 5,000 h
27,142 i
L: 750 i
CS: 968 i
L: 572 i
CS: 811 i
Figure 5: The two-step survey.
HDSS = Health and Demographic Surveillance System,
L = León, CS = Cuatro Santos, h = households, i = individuals.
An HDSS round normally takes about three months to complete. Our survey
in León took about this long, but in Cuatro Santos the work was delayed
because of heavy rains. Unfortunately, I only had the possibility to participate
during the first days of the fieldwork, in May 2008. Upon my return to
Sweden, I stayed in close contact with the staff at CIDS and CHICA over the
Internet, in order to be available if problems arose and to follow their progress.
A couple of months after my return to Sweden I received the results of the
survey study in the form of Access databases, sent via e-mail from the data
units at CIDS and CHICA, respectively.
Fieldwork, Cuatro Santos. Photos: Mariela Contreras.
The final result
Thus, 1,383 individuals responded to the survey, which gives a response rate
of 81% on average (76% in León and 84% in Cuatro Santos). A higher number
of respondents resided in Cuatro Santos than León (811/572), yet about the
same number lived in urban and rural areas, respectively (697/686) (there
were no rural respondents in León due to the selection) (Table 4).
Table 4: The respondents.
Mean age (years)
Sample group
Chronic ill
Acute/other ill
Relation to HoH (missing)
Head of household (HoH)
Child (adult child)
Other family
No/very low education
Low education
Medium education
High education
Not appl. (education)
Non-skilled worker
Skilled worker
Highly skilled worker
Informal worker
Not poor
Extremely poor
All (n 1,383)
León (n 572)
CS (n 811)
Notes: All data except on sample groups are derived from the HDSS. In absolute numbers
(unweighted values).
Analysis of non-responses
The reason for non-response was only noted in Cuatro Santos, where the main
reason was that the fieldworkers were unable to contact the respondent, most
commonly due to emigration and, to a lesser extent, absence from home (13
of the sampled individuals had died). The non-response rate was higher in
León than Cuatro Santos (24 and 16% of the respective samples). Most nonresponses were found in sample groups 3a and 2a (i.e. In-migrant, healthy
and Left-behind, healthy), together accounting for more than half of all nonresponses. Nearly half (48%) of the sampled individuals who did not
participate in the survey were heads of household or spouses of heads of
household (28 and 20%, respectively). A third were categorized as children
(i.e. in relation to head of household), and 16% were counted as “other family”
(e.g. grandparents and aunts). There was no great gender difference
concerning the non-response-rate (51% females, 49% males).
The survey data and statistical analysis
As mentioned earlier in the text, the HDSS data collected at both study sites
since the first survey rounds began (in 2002 in León and in 2004 in Cuatro
Santos) comprise information on socio-economic conditions, demographic
events, and reproductive and child health. The HDSS data we have used in our
study are from the years 2002-2007 (2004-2007 for Cuatro Santos); and we
have, furthermore, exclusively used data on socio-economic conditions and
vital events. These data have served two purposes in the present study: the
data on migration events were used as background information in the sample
process; and both the data on socio-economic conditions and on vital events
(particularly migration events) were used as background information in the
analysis of the 2008 survey – more specifically, for examining the
characteristics of the respondents and of their respective household members.
The HDSS data thus served as background information in both the sample
process and the analysis of the 2008 survey. The additional information we
retrieved through the first step of our survey in 2007 also served as
background in the sample process, but it has also been used as analytical
material. Together, these two sources have been used in the analysis of the
data from the 2008 survey. In Table 5 (next page), all data we have used in the
forthcoming analysis are presented: the HDSS data, as well as the data from
the first and second steps of our survey.
Table 5: Variables in the data
-Education (i)
-Occupation (i)
-Poverty index (h)
-Births (i)
-Deaths (i)
-Age (i)
-Sex (i)
-Family (i,h)
-Place of residence (i)
-Event history (i)
-Migration: type &
reason (i)
-Reception of
remittances (h)
-Health problems in the
past three months (i)
-Health service use (i)
-Health problems in the
past three months
-Use of health services
-Use of medicine
-Self-rated health
(physical and mental)
Step 1,
-Migration history
-Family network
-Legal status of
emigrated family
Step 2,
-Social insurance
-Help from others
-Help from others
during sick period
-Help, provision to
-Contact with family
members in other
-Social participation
-Social integration
with health
-Perceived social
-Perceived life
situation for
Notes: The unit for data collection is marked “i” for individual and “h” for household for the HDSS data and
Step 1 of our survey, though not for the 2008 survey since those data are individual.
It is perhaps necessary to point out once again that the HDSS surveys are
conducted with the head of household, or if he/she is unavailable, anyone at
home aged 16 years or over. Consequently, a high degree of the HDSS data is
“second-hand” information, provided by one (or a few) members of the
household. This implies that the information about out-migration is always
provided by the family members who are left behind rather than by the
migrant him/herself. In the present study, this applies to the HDSS data we
have used as background information, and to the data concerning health
problems and remittances we retrieved through our additional questions in
the first step of our survey in 2007. The data from our 2008 survey, in
contrast, were collected directly from the sampled individual.
Definitions in the HDSS data
“Household” is defined in the HDSS as all persons sleeping under the same
roof for at least half of the past month. “Housing conditions” are defined
according the house’s structure of walls, floors and roofs, and the type of water
supply, kitchen, and toilet/latrine. The level of poverty, calculated according
to the Unsatisfied Basic Needs Assessment, is based on three factors: housing
conditions (see above), school enrolment, and dependency ratio and
educational level of the head of household. If a household experiences twothree unmet basic needs in these three areas it is categorized as poor, and if
four needs are unmet it is considered extremely poor. In the HDSS data, a
“resident” is someone who has lived in the same household since baseline.
“Out-migration” refers to a person having moved out of the household at least
six months prior to the visit, whereas “in-migration” refers to the opposite – a
person having moved into the household within the past six months (Peña et
al. 2008).
Analysis of the survey data
As mentioned in the above description of the survey design, some sample
groups were over-sampled in the sample process. Due to this, weights had to
be applied to each sample group before the analyses could be performed
(Table 6).
Table 6: Weights
Sample frame (n)
Sample group
1a. Non-mover, healthy
1b. Non-mover, chronic ill
1c. Non-mover, other ill
2a. Lef- behind, healthy
2b. Left-behind, chronic ill
2c. Left-behind, other ill
3a. In-migrant, healthy
3b. In-migrant, chronic ill
3c. In-migrant, other ill
Sample (n)
Notes: The calculation of weights for each sample group is based on the number of individuals
(n) in the sample frame and in the sample, for each study setting separately (CS=Cuatro Santos).
In order to take into account the appropriate weights of the different sample
groups in the analysis, I used the programme IBM SPSS Complex Samples59.
The main methods for analysing the survey data thereafter consisted of
descriptive statistics (frequencies and cross tables) and binary logistic
regression analysis. The aim of the regressions was, first, to take the
descriptive analysis a step further by looking at what characterized certain
groups in the material (e.g. remittance-receivers), and, second, to explore
associations between migration categories (Non-movers, Left-behinds, and
In-migrants) and socio-economic characteristics and health indicators (e.g.
self-reported illnesses and self-rated health). The regression analyses are
explained more thoroughly in the empirical chapters (Chapters 5 & 7).
59 See: http://public.dhe.ibm.com/common/ssi/ecm/en/ytd03116usen/YTD03116USEN.PDF, for further
information about this programme.
The last fieldtrip: feedback and follow-up
In November 2013 I made a final fieldtrip to Nicaragua, with the ambition to
present and discuss the findings of the qualitative and quantitative studies. I
also aimed to follow up the interviewees who had participated in the
qualitative part of the study. During the trip, I held three presentations for
different audiences in León and in Matagalpa60, which all ended with a
discussion session that brought up many relevant questions and comments. I
unfortunately did not manage to present the study in Cuatro Santos, but our
collaborators from the area (those who currently or in the past had worked at
CHICA) nevertheless attended our presentations in León. Besides the more
formal presentations, we also discussed the study findings with researchers at
CIDS and CHICA as well as those more practically involved with the survey
(the responsible fieldworkers and database technicians), who also
participated in our seminars in León. Moreover, I also followed up how life
now was for some of my interviewees. I personally talked to three of them
(Cesar, Marta and Rosa), and through acquaintances (those who had
introduced me to the respective person) I heard how three more interviewees
were doing (Ana, Maribel and Mercedes).
Reflections on conducting mixed-methods research
In this study I have used both qualitative and quantitative methods because I
wished to investigate both the in-depth and the general picture of migrationhealth relations in Nicaragua. The mixing of methods has taken place
throughout the research project; from the initial data collection to the
analysis, and in the presentation and discussion of the study findings. I have
not followed a particular strategy for conducting mixed-methods research
wholly or exclusively, but triangulation, complementarity, bricolage and
pragmatism have all been inspirational. I have perhaps seen the qualitative
and quantitative data more as complementing than validating each other, but
I have also looked at convergencies between the two types of data (for
example, whether or not a certain issue could be found in the other type of
data). I have had a rather pragmatic outlook throughout the research process,
and focused more on conducting the research and finding answers to the
research questions than on thinking about ontological and epistemological
matters (for example, I have not used only one type of literature to
60 I made the presentations together with Gunnar Malmberg, as part of a programme established by a
Nicaraguan-Swedish research team consisting of Mariano Salazar and Ann Öhman from the Division of
Epidemiology and Public Health, Umeå University. Two presentations were held in León for students,
researchers, fieldworkers, and university staff (e.g. the Dean of the Medical Faculty). One presentation was held
in Matagalpa for people involved in the NGOs Grupo Venancia (a feminist network) and Médicos del Mundo
(Doctors without Borders, Matagalpa section). Approximately 150 people in all listened to the presentations.
contextualize my findings, and I have switched between posing qualitative and
quantitative research questions). In this sense I have perhaps served as some
sort of quilt maker (bricoleur).
Personally, I have not experienced any major problems of a paradigmatic kind
in the mixed-methods research process; largely since I do not regard myself
as belonging to a paradigmatic “extreme”. Instead, I share the realist
ontological view (meaning that reality has an existence independent of human
apprehension), and the subjectivist epistemological view (meaning that the
knowledge we have about reality is always “coloured”, or subjectively
influenced). Understanding is interpretation, as the hermeneutic tradition
proclaims (Schwandt 2003). I have also been inspired by critical
hermeneutics, which “brings the concrete, the parts, the particular into focus,
but in a manner that grounds them contextually in a larger understanding of
the social forces, the whole, the abstract (the general)” (Kincheloe & McLaren
2003: 445). Therefore, I have found it crucial to place the observations I made
in the field – by means of the interviews and the survey – in their social,
cultural and historical contexts. Moreover, I believe that this context is shaped
by power relations; hence, in line with some critical theorists I “locate the
foundations of truth in specific historical, economic, racial and social
infrastructures of oppression, injustice, and marginalization” (Lincoln & Guba
2003: 272-273).
The mixing of methods has been conducted from somewhere between a
pragmatic “middle stance” and a qualitative perspective. However, my belief
that knowledge about reality is produced/created in the interplay between
myself as an academic and my interviewees/respondents, and that this
knowledge is always subjectively “coloured” (situated) (Riessman 2008),
makes a high degree of reflexivity necessary during the whole research
process. I have consequently paid attention to both my own and my research
participants’ embeddedness in various value systems, and reflected upon
identity formations, biases and prejudices, and how these may have
influenced the research process. During the writing process, I have also tried
to be as self-reflexive as possible, and it is for this reason I write in the first
person; emphasizing that these are my words and my interpretations. The way
I have looked upon myself as a researcher, and upon the whole research
process (the collection, analysis, and interpretation of empirical material) has
been the same regardless of which method I have employed (see Lawson
1995). Hence, in this line of reasoning, the findings stemming from the
empirical material should be regarded as providing only partial and
incomplete pictures from reality, and the thesis as a whole should be regarded
as an interpretative account of the information and the stories collected
through the qualitative and quantitative research methodologies.
The use of a mixed-methods approach proved to be of great value for a
comprehensive understanding of migration-health relations in the study
setting. The interviews allowed the connections between migration and health
to be explored with an open mind, and provided a personal and contextualized
understanding of the complex ways migration and health are connected in the
case of Nicaragua. The survey made it possible to investigate the size and
scope of certain aspects of migration-health relations, and this data thus
provided general descriptions as well as information on certain associations
between migration and health. Hence, the two methods have complemented
each other in important ways through generating different kinds of
knowledge, aiming at depth and breadth, understanding and generalization.
Even though I did not experience any major paradigmatic difficulties, I did
have some problems related to my ability to conduct quantitative research,
since I had mainly worked with qualitative studies before this. The
quantitative survey study proved to be time-consuming, both in the collection
phase and in the analytical work. I also experienced some difficulties
concerning how to best present the results from the interview and survey data
without placing emphasis on one over the other.
The next chapter provides a background of the study context – Nicaragua – in
both past and present times. It also presents the two study settings, León and
Cuatro Santos, in more detail.
Mural, central León.
Church of Subtiava, León.
Street view, León.
Nicaraguan landscapes:
“La vida es dura”
A common saying in Nicaragua is “la vida es dura”, which in English
translates to “life is hard”. This phrase captures a great deal of how life is
experienced by the Nicaraguan people61. The men and women I interviewed
for this study all mentioned the difficult living conditions in Nicaragua, in
relation to various areas such as the economy, the labour market, politics,
migration, the educational and health care systems, and the unpredictable
nature conditions. I believe it is essential to acknowledge this context for a
proper analysis and understanding of the migration-health nexus in the case
of Nicaragua. The main objective of this chapter is therefore to provide an
account of relevant aspects of this context, mainly through the use of
secondary sources. Some quotations from the interviews are also included to
illustrate the events portrayed.
The chapter starts off with a historical exposé of crucial moments in the
history of Nicaragua in relation to Central America as a whole. It begins in the
colonial era and the period after independence, and continues with the years
of dictatorship, revolution, the Contra war and structural adjustments.
Thereafter follows a description of present-day Nicaragua, with focus on the
political, social and economic situation at the time of the fieldwork, when
former revolutionary leader Daniel Ortega came back into power. The central
issues of migration and health in Nicaraguan society are portrayed throughout
the chapter. An overview of key events and selected socio-economic indicators
is presented in Table 7, p. 122.
Nicaragua has been damaged by many things…
Earthquakes… and bad governments, above all.
(Fernando, 50 years, Cuatro Santos)
61 Robert Lancaster portrays this saying in his book “Life is hard: Machismo, Danger and the Intimacy of Power
in Nicaragua” (1992).
transformations 1520-2006
“From the violence […] emerged Nicaragua”, wrote the American historian
Bradford Burns in 1991 (Burns 1991: 1). This quote highlights the conflictridden nature of Nicaraguan society after it gained independence from
colonial rule. The social, political and economic imbalances that began during
the colonial era came to characterize Nicaragua in the post-independence
period as well, and have continued to distinguish the country ever since, as we
shall see in this chapter.
Because of a “shared geopolitical destiny” (Torres Rivas 1993: xvii) with the
rest of Central America, I will portray Nicaragua’s history in relation to the
whole isthmus (although important differences between the five countries
there do exist, which have caused them to develop differently over the years)
(Walker & Wade 2009; Cardoso 1985). Nevertheless, Nicaragua and all
Central American countries except Costa Rica share the historical similarities
of externally oriented agricultural economies, patterns of dependency