Linking social capital and self-rated health: A multilevel analysis of 11,175 men and women in Sweden
Introduction
The main purpose of the present study was to analyse the association between the recently minted concept of linking social capital, measured as voting in national elections, and self-rated health.
Social capital and social trust are broad concepts that are produced in the interactions between members of the society. The social scientist Durkheim stated in 1895: “The group thinks, feels and acts entirely differently from the way its members would if they were isolated. If therefore we begin by studying these members separately, we will understand nothing about what is taking place in the group” (Durkheim, 1982). A century after this statement, researchers such as Putnam, 1993, Putnam, 2000, Coleman (1990) and Bourdieu (1986) shaped the concept of social capital. Social capital could be regarded as a societal characteristic while age, gender, ethnicity, educational attainment, and social position are truly individual characteristics. Putnam has claimed that social participation in informal as well as formal social networks is the core of the social capital concept or “what it is” (Putnam, 2001). Social trust, on the other hand, is a consequence of social capital or “what it does”. In contrast to Putnam, Fukuyama has claimed that social trust is the core of social capital, or “what it is” (Fukuyama, 1996).
Social capital and social trust have stimulated considerable interest in health sciences (Green et al., 2000; Hyyppa and Maki, 2001a; Kawachi et al., 1997; Smith and Polanyi, 2003), economic research (Holzmann and Jorgensen, 1999; Woolcock and Narayan, 2000), and political sciences (Rothstein, 2000, Rothstein, 2003). However, major criticism has been raised against the conceptualisation of exactly what social capital refers to (Davey Smith and Lynch, 2004). Economic researchers have endeavoured to conceptualise social capital, which is associated with economic growth (Holzmann and Jorgensen, 1999; Woolcock and Narayan, 2000). Theories have been developed in which bridging social capital operates at the horizontal level (trust in other people) and governance at the vertical level (trust in government institutions) (Narayan, 2002; Narayan and Cassidy, 2001). Szreter and Woolcock developed these theories further and presented three concepts: bonding, bridging, and linking social capital (Szreter and Woolcock, 2004). Bonding social capital refers to “trusting and co-operative relations between members of a network who see themselves as being similar in terms of their shared social identity” and bridging social capital comprises “relations of respect and mutuality between people who know they are not alike in some socio-demographic (or social identity) sense (differing by age, ethnic group, class, etc.)”. Linking social capital includes “norms of respect and networks of trusting relationships between people who are interacting across explicit, formal or institutionalised power or authority gradients in society” (Szreter and Woolcock, 2004). Vertical trust, which Rothstein defines as “trust in political and societal institutions” (Rothstein, 2000), is thus closely related to linking social capital. Linking social capital connects people across vertical power gradients and creates trusting ties to formal institutions (Szreter and Woolcock, 2004). Surveys of people from Scandinavia, the United States, and Australia have found associations between political action, political interest and measures of trust (Woolcock and Narayan, 2000). We therefore argue that voting is an important component of people's trust in formal political institutions and a good indicator of linking social capital.
Several previous studies have found a positive association between self-rated health and various measures of social capital (Blakely et al., 2001; Carlson, 2004; Hyyppa and Maki, 2001b; Jun et al., 2004; Kawachi et al., 1999; Lindstrom, 2004; Pollack and von dem Knesebeck, 2004; Rose, 2000; Subramanian et al., 2001, Subramanian et al., 2002). One of these studies found that inequalities in political participation at state-level were associated with poor self-rated health (Blakely et al., 2001). The authors of a study that investigated various dimensions of women's status and self-rated health found that women living in states scoring in the lowest quintile on women's political participation were more likely to rate their health as poor (Jun et al., 2004). The present study increases previous knowledge on this topic because we used small area neighbourhood units instead of larger geographic units, such as states, to define linking social capital as proportions of individuals voting in national elections at neighbourhood level. The rationale for using small geographic units instead of larger geographic units is that social capital to a high extent is created in the social interactions between individuals, i.e. social participation in informal as well as formal social networks. We believe that larger geographic units, such as states, do not capture the important social interactions and social networks that are the core of the social capital, according to Putnam (2001). In addition, we analysed our sample, representative of the entire population in Sweden, within a multilevel framework in order to investigate both fixed and random effects at two levels, i.e. individuals nested in neighbourhoods. The outcome self-rated health was selected because of its usefulness as a predictor of mortality (Idler and Angel, 1990; McGee et al., 1999; Miilunpalo et al., 1997; Mossey and Shapiro, 1982; Sundquist and Johansson, 1997), which indicates that a subjective health assessment is a valid health status indicator that can be used in population health monitoring.
The first aim of this study was to analyse the association between neighbourhood linking social capital and self-rated health in a sample of 11,175 individuals aged 25–64 years. The second aim was to investigate whether the hypothesised association between linking social capital and self-rated health remains after accounting for potential confounding factors related to individual power in the society, i.e. age, gender, country of birth, marital status, housing tenure, attained level of education, and individual voting.
Section snippets
Methods
We used the Swedish Annual Level of Living Survey (SALLS) for the analyses. SALLS has been conducted annually for the last 30 years by Statistics Sweden (the Swedish Government-owned statistics bureau) on the basis of a simple, random sample, representative of the entire non-institutionalised adult population in Sweden. The participants in this study consisted of 5710 women and 5465 men, aged 25–64, interviewed face-to-face by well-trained interviewers. The study was based on the survey in
Results
The description of the sample is shown in Table 1. In neighbourhoods with low levels of linking social capital there was a high proportion of foreign-born people, people who lived alone, people who rented their home and people with low and middle educational status. An opposite pattern was observed in neighbourhoods with high levels of linking social capital, where a high proportion of Swedish-born people, married/cohabiting people, people who owned their home and people with high educational
Discussion
The main finding of this study was that individuals living in neighbourhoods with the lowest levels of linking social capital, defined as voting in national elections at neighbourhood-level, had a significantly higher risk of poor self-rated health than individuals living in neighbourhoods with the highest levels of linking social capital. This increased risk remained significant after adjustment for age, gender, country of birth, marital status, housing tenure, attained level of education, and
Conclusions
Our findings demonstrate the need to increase the levels of linking social capital in powerless neighbourhoods. Since there is a connection between education and community participation (Hall, 1999; Wilson, 1997), investments in education could be an important component in order to increase both human and social capital (Fukuyama, 2001). Such investments could help individuals to build social skills, establish strong social partnerships and engage in shared norms and values (Hall, 1999). In
Acknowledgements
This work was supported by grants from the National Institutes of Health (1R01 HL71084-01), and the Swedish Research Council to Dr. Kristina Sundquist (K2005-27X-15428-01A).
References (60)
Stress and physical health: the role of neighborhoods as mediating and moderating mechanisms
Social Science & Medicine
(2004)The European health divide: a matter of financial or social capital?
Social Science & Medicine
(2004)- et al.
Individual-level relationships between social capital and self-rated health in a bilingual community
Preventive Medicine
(2001) Social capital, the miniaturisation of community and self-reported global and psychological health
Social Science & Medicine
(2004)- et al.
Self-rated health status as a health measure: the predicitive value of self-reported health status on the use of physician services and on mortality in the working-age population
Journal of Clinical Epidemiology
(1997) - et al.
Social capital and health among the aged: comparisons between the United States and Germany
Health Place
(2004) How much does social capital add to individual health? A survey study of Russians
Social Science & Medicine
(2000)- et al.
Social norms, social behaviours and health: an empirical examination of a model of social capital
Australian and New Zealand Journal of Public Health
(2003) - et al.
Does the state you live in make a difference? Multilevel analysis of self-rated health in the US
Social Science & Medicine
(2001) Cluster Analysis for Applications
(1973)