Perceptions of social capital and the built environment and mental health

https://doi.org/10.1016/j.socscimed.2005.11.037Get rights and content

Abstract

There has been much speculation about a possible association between the social and built environment and health, but the empirical evidence is still elusive. The social and built environments are best seen as contextual concepts but they are usually estimated as an aggregation of individual compositional measures, such as perceptions on trust or the desirability to live in an area. If these aggregated compositional measures were valid measures, one would expect that they would evince correlations at higher levels of data collection (e.g., neighbourhood). The aims of this paper are: (1) to investigate the factor structure of a self-administered questionnaire measuring individual perceptions of trust, social participation, social cohesion, social control, and the built environment; (2) to investigate variation in these factors at higher than the individual level (households and postcodes) in order to assess if these constructs reflect some contextual effect; and (3) to study the association between mental health, as measured by the General Health Questionnaire-12 (GHQ-12), and these derived factors. A cross-sectional household survey was undertaken during May–August 2001 in a district of South Wales with a population of 140,000. We found that factor analysis grouped our questions in factors similar to the theoretical ones we had previously envisaged. We also found that approximately one-third of the variance for neighbourhood quality and 10% for social control was explained at postcode (neighbourhood) level after adjusting for individual variables, thus suggesting that some of our compositional measures capture contextual characteristics of the built and social environment. After adjusting for individual variables, trust and social cohesion, two key social capital components were the only factors to show statistically significant associations with GHQ-12 scores. However, these factors also showed little variation at postcode levels, suggesting a stronger individual determination. We conclude that our results provide some evidence in support of an association between mental health (GHQ-12 scores) and perceptions of social capital, but less support for the contextual nature of social capital.

Introduction

There has been an ongoing debate on the importance of people and places for health. This interest has been stimulated by the debate on social capital and health (Kawachi & Berkman, 2000; Muntaner & Lynch, 2002; O’Brien Caughy, O’Campo, & Muntaner, 2003; Pearce & Davey-Smith, 2003; Sampson, 2003; Tunstall, Shaw, & Dorling, 2004) and studies examining health variations across geographical areas (Ellaway, Macintyre, & Kearns, 2001; Macintyre, Ellaway, & Cummins, 2002; Macintyre, Maciver, & Sooman, 1993; Skapinakis, Lewis, Araya, Jones, & Williams, 2005).

Interest in finding better ways of measuring and capturing the contextual nature of places, neighbourhoods, and communities has also grown. Some of the difficulties in the measurement of contextual variables can be appreciated when examining how social capital has been measured so far.

Although there is no consensually agreed definition of social capital, one commonly used one is that it refers to how social relations and networks influence collective action for mutual benefit (Kawachi, Kennedy, Lochner, & Prothrow-Stith, 1997; Putnam, 1993). The concept of social capital can be disaggregated into at least two important components, structural and cognitive (Bain & Hicks, 1998). Whilst the former refers to the extent and intensity of associational links, the latter has more to do with qualitative aspects of these links, such as levels of trust or reciprocity.

Much of the research in social capital and health has used these so-called cognitive features assessed at an individual level to estimate indirectly the amount of social capital in an area (Kawachi, Kennedy, & Glass, 1999; Kawachi et al., 1997; Lochner, Kawachi, Brennan, & Buka, 2003; Lochner, Kawachi, & Kennedy, 1999; McCulloch, 2003; Subramanian, Lochner, & Kawachi, 2003). This is problematic because these cognitive aspects of social capital are meant to represent a contextual construct, rather than a compositional one obtained through aggregating individual data. Two different, but not mutually exclusive, approaches to pursuing this challenge have been used. Firstly, these features can be measured through direct observation of certain collective behaviours in an area; e.g., people not respecting zebra crossings or arguing in public spaces might indicate lower levels of social capital. Although this approach might be more contextually valid, it can be resource intensive, some features are not directly observable, and some observations require subjective inferences, such as deciding if two people engaged in a discussion can be regarded as ‘arguing’. Secondly, some of these cognitive components can also be measured indirectly by investigating individuals’ perceptions of, e.g., how much people trust each other in the neighbourhood. These individual perceptions are then aggregated or analysed at higher levels of aggregation using multilevel models to obtain an estimate of the level of trust in the area. Several self-reported questionnaires have been developed to assess these perceptions on various aspects of social capital with differences depending on the specific aims for which they were designed (Lochner et al., 1999; Sampson, Raudenbush, & Earls, 1997; The World Bank Social Capital Thematic Group, 2002).

Although this methodology is simple and feasible, it is still questionable if these perceptions are valid contextual measures or just simply the sum of individuals’ perceptions (compositional). Furthermore, even if these perceptions reflected a truly contextual characteristic it is still possible that this estimate could be somehow confounded by the characteristics of the individuals living in that place. One way of indirectly attempting to clarify if these perceptions represent some contextual construct is by trying to ascertain what proportion of the variance on any of these constructs (e.g., perceptions on trust or reciprocity) is explained at higher levels, such as neighbourhoods, after accounting for individual factors. In a recent study, Subramanian et al. (2003) used individual data on the perception of trust by individuals in Chicago, USA, to examine whether there were true differences in trust between neighbourhoods after accounting for individual variation. Their results suggested that, even after accounting for individual socio-demographic variables, significant neighbourhood variation remained in the individual perception of trust (Subramanian et al., 2003).

The built environment can be assessed using direct observations of the characteristics of geographical areas (Perkins, Meeks, & Taylor, 1992; Weich, Holt, Twigg, Jones, & Lewis, 2003) or through perceptions of residents on the quality of their built environment (Dalgard & Tambs, 1997). Perceptions on the built environment, such as graffiti on walls or dirtiness, are hypothetically subject to similar respondent bias as perceptions on social capital, e.g., trust or social cohesion.

It could be argued that the quality of the built environment is a consequence of different levels of social capital or vice versa. For instance, a high proportion of houses with broken windows is probably the consequence of low levels of social capital in the area. However, it is also plausible that unfavourable changes in the physical environment might lead to deterioration in trust or social cohesion.

A few studies have been published in peer-reviewed journals reporting on the association between social capital and common mental disorders among adults. Most of these studies have only measured and analysed social capital data at the individual level (Ellaway et al., 2001; Harpham, Grant, & Rodriguez, 2004; Ross, 2000; Silver, Mulvey, & Swanson, 2002; Steptoe & Feldman, 2001). Others have either aggregated individual data to create compositional variables representing higher levels (e.g., neighbourhood) or analysed individual data using hierarchical multilevel models to estimate the level of variation at different levels (Cutrona, Russell, & Hessling, 2000; Skapinakis et al., 2005).

The underlying, but yet unproven, assumption has been that social capital is good for mental health. Although some studies analysing data at individual level have found inverse associations between mental illness and social capital, most studies using aggregated data or multilevel models have failed to find statistically significant associations between social capital and common mental illness at higher levels, such as neighbourhoods. It is worth emphasising that most of these studies have used different sampling designs, measures of social capital and mental health, and different hierarchical data structures, making comparisons rather problematic.

Studies on the built environment and mental health have focused on residents’ perceptions of their environment (Dalgard & Tambs, 1997; Ross, 2000) and geographical area variations or contextual assessments of the quality of the built environment (Duncan, Jones, & Moon, 1995; Pickett & Pearl, 2001; Reijneveld & Schene, 1998; Wainwright & Surtes, 2003; Weich, Blanchard, & Prince, 2002; Weich et al., 2003). Most of these studies have also failed to find statistically significant area effects on mental health after accounting for individual factors. Amongst studies that have used multilevel modelling with positive findings, Skapinakis et al. (2005) found a small but significant association between mental health and geographical areas in a nationally representative sample in Wales but no specific factor explained these findings (Skapinakis et al., 2005). Ross (2000) found that neighbourhood disorder and residential instability were associated with depression (Ross, 2000). Other studies using multivariate analysis of individual data have shown that characteristics of the built environment can be associated with psychiatric symptomatology (Dalgard & Tambs, 1997; Sampson, 2003; Weich et al., 2002).

The complex way in which the social and built environment might interact to affect mental health is unknown but there is no shortage of speculation on the potential mechanisms. It is possible that the social and built environment can effect changes in each other and eventually impact on mental health. For instance, a poorly maintained built environment with derelict buildings and covered in rubbish might affect the sense of social cohesion in the neighbourhood, the combination of both leading to poorer mental health among its residents. But it is also possible that poor social cohesion might lead to a poorer built environment as residents might have little interest to look after their common areas. Poorer mental health among residents might also lead to less interest to keep the neighbourhood tidy and to engage in social interactions. All these combinations are possible and thus it is important to conduct studies in which both the perceptions of the social and built environment are simultaneously assessed and analysed at different levels of data aggregation. Different aspects of the built environment may affect residents’ perceptions of their neighbourhood and lead to behaviours congruent with these beliefs. For instance, empty and boarded houses may facilitate criminal activity and lead to perceptions of lack of safety and unwillingness to interact with other people. Social withdrawal, isolation, and fear are likely to lead to the emergence of psychiatric symptoms among vulnerable people. However, there is as yet no empirical evidence showing unequivocally these or many other potential associations between the social and built environment and mental health.

This study was part of a comprehensive research programme [housing and neighbourhood and health (HANAH)] investigating the relationships between the built and social environment and health. In this paper, we present the results of the self-administered questionnaire developed to measure the residents’ perceptions of aspects of the social and built environment. The aims of this paper were: (1) to look for common factors grouping items of a self-administered questionnaire to measure perceptions of the social and built environment; (2) to investigate variation in factors derived from the questionnaire at area level, after accounting for individual variables, in order to estimate if these compositional constructs might reflect some contextual effect; and (3) to study the association between mental health and these factors.

Section snippets

Sampling strategy

A cross-sectional household survey was undertaken during May–August 2001 in a district of South Wales with a population of 140,000. The area's economy used to be dominated by heavy industry, including coal, steel, and petrochemicals, but these have seen a significant decline since the 1970s. This has left a legacy of environmental pollution from heavy industry, poor standards of housing and amenities, high levels of poverty and economic inactivity. Some large-scale post-war housing developments

Sample characteristics

In total, 1058 adults from 647 households completed the questionnaires. This represents an individual response rate of 66% and a household response rate of 73%. The number of individuals per postcode varied from 1 to 47 with a mean of 20.7. There was a range between 1 and 22 households per postcode. The mean age of residents was 46 years (SD 16.1 years), 55% were females, and 32% of the households had children under the age 16. Two-thirds were married or co-habiting, 19% were single, and the

Discussion

The design of our questionnaire was a comprehensive attempt to assemble a wide array of questions relating to important factors included in social capital definitions and studies on the quality of the built environment. Principal component analysis grouped questions in factors similar to the theoretical constructs we had envisaged. Our neighbourhood quality factor captured almost a third of the variance at higher levels. Although a lower proportion of the variance was explained at higher levels

Acknowledgements

We would like to thank N. Weaver, J. Patterson, T. Bell, and P. Jones for their contribution to this study. We thank Gareth Williams and Ben Rolfe for their help in developing the questionnaire. The research was supported by a joint grant from the UK Medical Research Council and the Engineering and Physical Sciences Research Council (grant reference number G9900679).

References (46)

  • T.A. Blakely et al.

    Ecological effects in multi-level studies

    Journal of Epidemiology & Community Health

    (2000)
  • S. Curtis et al.

    Is there a place for geography in the analysis of health inequality?

    Sociology of Health & Illness

    (1998)
  • C.E. Cutrona et al.

    Direct and moderating effects of community context on the psychological well-being of African American women

    Journal of Personality and Social Psychology

    (2000)
  • O.S. Dalgard et al.

    Urban environment and mental health: A longitudinal study

    British Journal of Psychiatry

    (1997)
  • C. Duncan et al.

    Psychiatric morbidity: A multilevel approach to regional variations in the UK

    Journal of Epidemiology & Community Health

    (1995)
  • A. Ellaway et al.

    Perceptions of place and health in socially contrasting neighbourhoods

    Urban Studies

    (2001)
  • D. Goldberg et al.

    The user's guide to the general health questionnaire

    (1988)
  • Goldstein, H. (1995). Multilevel statistical models....
  • T. Harpham et al.

    Mental health and social capital in Cali, Colombia

    Social Science & Medicine

    (2004)
  • I. Kawachi et al.

    Social cohesion, social capital, and health

  • I. Kawachi et al.

    Neighborhoods and health

    (2003)
  • I. Kawachi et al.

    Social capital and self-rated health: A contextual analysis

    American Journal of Public Health

    (1999)
  • I. Kawachi et al.

    Social capital, income inequality and mortality

    American Journal of Public Health

    (1997)
  • Cited by (0)

    View full text