Social capital and sense of insecurity in the neighbourhood: a population-based multilevel analysis in Malmö, Sweden

Abstract

The aim of this study was to investigate the influence of social capital on self-reported sense of insecurity in the neighbourhood. The public health survey in Malmö, Sweden in 1994 was a cross-sectional study. A total of 5600 individuals aged 20–80 years were asked to answer a postal questionnaire. The participation rate was 71%. A multilevel logistic regression model, with individuals at the first level and neighbourhoods at the second, was performed. We analysed the effect (intra-area correlation, cross-level modification and odds ratios) of individual (social participation) and neighbourhood social capital (electoral participation in the 1994 municipal election) on sense of insecurity after adjustment for compositional factors. Neighbourhood factors accounted for 7.2% of the total variance in individual insecurity. This effect was marginally reduced when the individual factors were included in the model. In contrast, it was reduced by 70% by the introduction of the contextual variable. This study suggests that social capital, measured as electoral participation, may partly explain the individual's sense of insecurity in the neighbourhood.

Introduction

Social capital has been shown to be of great importance for population health. Social capital is a feature of the social context, not of the individual actors within the social context. Neighbourhood social capital may be associated with population health by providing affective support and by being the source of self-esteem and mutual respect (Wilkinson, 1996; Kawachi, Kennedy, & Glass, 1999), by increasing access to local services and amenities (Kawachi et al., 1999), by promoting more rapid transmission of health information, adaptation of health behaviour norms and social control over deviant health-related behaviours (Lindström, Hanson, & Östergren, 2001; Lindström, Isacsson, & The Malmö Shoulder Neck Study Group, 2002), and by facilitating the prevention of crime (Kennedy, Kawachi, Prothrow-Stith, Lochner, & Gupta, 1998; Kawachi, Kennedy, & Wilkinson, 1999; Putnam, 2000).

Shaw and McKay argued already in 1942 that crime could be linked to broad social forces (Shaw & McKay, 1942). The level of social capital in the states of the USA has been shown to be negatively correlated with the rates of violent crime (homicide, aggravated assault and robbery in descending order). The high incidence of crime in some urban areas has been explained by the loss of social buffers that normally exist in middle-class neighbourhoods. These social buffers consist of formal and informal networks of organisations and social norms concerning labour force participation and educational attainment (Wilson, 1987; Anderson, 1990). A study of British crime data found that in areas where people are connected through tight bonds of friendship, and people are active in local committees and clubs, there are fewer muggings, assaults, burgleries and so forth (Sampson & Groves, 1989). Another study found that participation in community organisations helped to lessen the effects of socioeconomic disadvantage on juvenile delinquency (Simcha-Fagan & Schwartz, 1986). The social networks in older urban neighbourhoods often had a high ability to produce public safety. This kind of neighbourhoods had a plentiful stock of social capital embedded in the relationships among families, shopkeepers and other business owners in the neighbourhood. Crime control was largely a matter of adult supervision, and the adults kept track of young people who were likely to get into trouble (Fukuyama, 1999; Jacobs, 1992). In the 1950s and the 1960s, many such neighbourhoods were pulled down to make way for new housing projects in both the USA (Scott, 1998) and in Europe. Neighbourhoods that combined dwelling, shops and other activities within the same geographic areas were replaced with single-use tracts that kept working people out of residential areas during the day. As a result crime rates began to rise in all Western countries in the 1960s and the 1970s (Fukuyama, 1999). It has been demonstrated that the strong associations of concentrated disadvantage and residential instability with violent crime are largely mediated by collective efficacy, i.e. the linkage of mutual trust and the willingness to intervene for the common good (Sampson, Raudenbush, & Earls, 1997). Neighbourhoods differentially impose informal social control. The willingness of local residents to intervene for the common good depends in large part on conditions of mutual trust and solidarity among neighbours (Coleman, 1990; Putnam, 1993). The relationship between social capital and crime appears to be reciprocal. High crime rates may themselves be a forceful contributor to the deterioration of the social capital of a society. The relationship between social capital and crime may induce a vicious circle of feedback processes. These feedback processes, by which declining social capital contributes to crime, and crime contributes to declining social capital, include fear of crime leading to physical and psychological withdrawal from community life, deteriorating conditions leading to the exit of businesses, accompanying job losses and change in the social composition of the population within the neighbourhood (Kawachi et al., 1999; Skogan, 1991).

Although several studies have shown that police reporting corresponds fairly accurately with public perceptions of crime when those crimes are serious (Huang, 1993), self-reported fear of crime or subjective sense of insecurity in public places seems to represent an independent psychological consequence of crime at the population level apart from the direct physical and psychological consequences for the individual victims of crime. First, fear of crime is theoretically and conceptually defined as an emotional response of dread or anxiety to crime or symbols that a person associates with crime (Ferraro & LaGrange (1987), Ferraro & LaGrange (1992); Ferraro, 1995). To produce a fear reaction in humans, a recognition of a situation as possessing at least potential danger, real or imagined, is necessary (Ferraro, 1995). Second, there are substantial and consistent empirically observable age and sex differences in fear of crime. Elderly people (Clarke, Ekblom, Hough, & Mayhew (1980), Yin (1982); Clarke, Ekblom, Hough, & Mayhew, 1985) and women (Ferraro & LaGrange, 1992) are more likely to express self-reported fear of crime or sense of insecurity in public places, despite the fact that they are not necessarily victimised to any higher extent than others. Sense of insecurity in public places/fear of crime is related to health independent of the causal relationship between crime and health. People are continually making decisions about the relative risks from personal and ecological forces. Ecological concerns might include air and water quality, congestion, and crime. Personal factors would include genetic and lifestyle influences, where lifestyle would include both health-destructive and health-protective behaviours ranging from dietary to occupational activities (Langlie, 1977; Becker & Rosenstock, 1989; Ferraro, 1995). Factors that may produce fear reactions in humans include (1) personal experience of victimisation, i.e. having experienced crime, (2) sense of vulnerability, i.e. both a sense of physical and social vulnerability, (3) rumours and experiences communicated by others, i.e. the transmission of values, rumours and facts concerning crime through media or important others, (4) neighbourhood characteristics, i.e. integration, social ties and neighbourhood relations, and (5) fear of disorder (Skogan & Maxfield, 1981).

Social capital has mostly been defined and operationalised as social participation in civic matters and other social activities, and generalised trust to other people. Generalised trust to other people, i.e. “thin trust”, concerns the general trust for people that you are not necessarily acquainted with (Putnam, 1993). Sense of insecurity in public places or fear of crime might theoretically be considered as one of the aspects or parts of this form of generalised trust. However, generalised trust to other people strongly differ from self-reported fear of crime/sense of insecurity in public places both theoretically/conceptually and empirically. Conceptually, the difference is obvious. Furthermore, in both the USA and Sweden the empirical age and gender patterns of lack of generalised trust to other people show no differences between men and women, and the prevalence of lack of generalised trust to other people decreases with increasing age (Putnam, 2000; unpublished data from Scania, southern Sweden 2000). These statistical patterns strongly contrast with the age and gender patterns of fear of crime/sense of insecurity in public places, which is often more prevalent among women and the elderly (Clarke, Ekblom, Hough, & Mayhew (1980), Yin (1982), 1982; Clarke et al., 1985; Ferraro & LaGrange, 1992). Sense of insecurity in public places/fear of crime is thus both conceptually and empirically independent in relation to generalised trust and social capital. This conceptual independence does not exclude a causal relationship between social capital and sense of insecurity/fear of crime in public places. In fact, there are strong arguments that suggest such a causal relationship.

Neighbourhoods with strong social ties, good neighbour relations and high levels of civic activities also have much lower prevalence of sense of insecurity/fear of crime than other neighbourhoods (Skogan & Maxfield, 1981; Ferraro, 1995). Lewis and Salem considered fear of crime/sense of insecurity as an expression of problematic social control. Individuals living in neighbourhoods with high levels of fear of crime regarded themselves as incapable of controlling the deviant behaviours of others (Lewis & Salem, 1986). To our knowledge, no multilevel analysis has been performed to disentangle the complex relationship by which social capital creates sense of security or insecurity among the citizens. This problem will be dealt with in this study.

Social capital is a multifaceted and at present ill-defined concept with a number of competing definitions (Lomas, 1998; Wall, Farrazzi, & Schryer, 1998). Social capital is defined by most theorists as the formal and informal social networks and associations that connect individuals and groups of individuals to each other (Gilles, 1998; Putnam, 2000). Pierre Bourdieu defines social capital as the product of significant relationships with social others (Bourdieu, 1998). It is thus, at present, almost impossible to construct a complete empirical measure of social capital that completely covers the concept (Forbes & Wainwright, 2001). In this study social capital was measured as social participation reported by the individuals participating in the postal public health survey questionnaire in 1994 (individual level data) used in this study, and by electoral participation in different electoral districts in the municipal election in Malmö in 1994 (contextual level data). Social participation is a central part of the definition of social capital. Participation/voting in elections is regarded by Putnam as an important aspect of political participation that, although not a direct measure of social capital, is the most common form of political activity. Voting is an instructive proxy measure of broader social change. Compared to demographically matched non-voters, voters are more likely to be interested in politics, to give to charity, to volunteer, to attend community school board meetings, to participate in public demonstrations, and to cooperate with their fellow citizens in community affairs. The decrease in electoral participation and voting in recent decades in both Sweden and the USA reflect a growing distrust in government, declining party mobilisation, fraying social bonds, and political dealignment (Putnam (1993), Putnam (2000), Putnam, 1993 (2001)). A recent study has also demonstrated that socio-economic inequality in political participation (measured by voter turnout) is associated with poor self-rated health, independently of both income inequality and state median household income (Blakely, Kennedy, & Kawachi, 2001).

Age, sex (Clarke, Ekblom, Hough, & Mayhew (1980), Yin (1982); Clarke et al., 1985; Ferraro & LaGrange, 1992), country of origin, and education might be confounders of the associations between social capital (electoral participation and social participation) and sense of insecurity in public places.

The aim of this study is to test the influence of social capital measured both as social participation and electoral participation in the 1994 municipal election on the individual's sense of insecurity in the neighbourhood after darkness (expectation and fear of being assaulted, robbed, etc.) in the city of Malmö, Sweden, using a multilevel model.