Linking the Micro- and Macrolevel Dimensions of Community Social Organization

Abstract

This study evaluates a multilevel theory of community social organization and its structural antecedents. The main hypothesis derived from a systemic conceptualization of community is that effect of residential stability on social cohesion is mediated by density offriendship and acquaintanceship networks. This hypothesis is tested in two stages using data bases constructed from a 1984 national sample of 11,030 residents of over 500 localities in Great Britain. First, community-level results show that positive effect of residential stability on a measure of social cohesion is accounted for by increased level of friendship/acquaintanceship and decreased level of anonymity among residents, regardless of urbanization and other sociodemographic controls. Second, individual-level analysis indicates that residential stability has both individual and contextual effects on local social ties, which in turn promote individual attachment to community. Social cohesion is also directly related to community satisfaction at both levels of analysis. Overall, results support and extend a systemic theory that bridges microand macrolevel dimensions of community social bonds. There has been an increasing recognition in recent years of need for research that links microand macrolevel dimensions of social behavior (Alexander et al. 1987; Huber 1990). This need appears especially true in area of community studies. As Sampson (1988) has argued, contemporary research on local community bonds has concentrated primarily on individuals as causal unit of empirical analysis and theoretical inference, particularly with regard to psychological adjustment (see also Freudenburg 1986). Furthermore, even competing explanations to major ecological constructs in classical tradition such as size and density (e.g., Wirth 1938) have relied on individual-level demographic characteristics. Thus, a common research issue deals with relative effects of urbanization, life cycle, social status, and age on individual* 7his is a revised version of an article presented at conference Micro-Macro Link in Sociological Theory at Inter-University Centre for Post Graduate Studies in Dubrovnik, Yugoslavia, April 16-27, 1990. 1 thank participants of that conference for helpful comments, as well as referees of Social Forces. Direct all correspondence to Robert J. Sampson, Department of Sociology, University of Chicago, Chicago, IL 60637. ? The University of North Carolina Press Social Forces, September 1991, 70(1):43-64 This content downloaded from 207.46.13.111 on Tue, 09 Aug 2016 04:59:16 UTC All use subject to http://about.jstor.org/terms 44 / Social Forces 70:1, September 1991 level social participation, alienation, psychological adjustment, and neighboring in local community (for more detailed reviews see Baldassare 1979; Fischer 1977, 1982; Freudenburg 1986; Sampson 1988; Tittle 1989; and Wilson 1985). As a consequence, research on communities has tended to neglect two crucial issues: macrosocial determinants of community social organization and contextual effects of community structure on individual behavior (Moksony 1990; Sampson 1988). This negligence is surprising because a fundamental assumption of ecological approach is that social systems exhibit structural properties that can be examined apart from personal characteristics of their individual members (Berry & Kasarda 1977; Hawley 1950). In particular, social cohesion of a collectivity is a central concept in communitylevel theory that is not easily reduced, if at all, to an individual-level attribute (Buckner 1988; Sampson & Groves 1989). Despite macrosocial and contextual grounding of human ecological and community theory, typical design of sample survey in modern sociology seems to have channeled research to an individual level of analysis and to a focus on causally proximate (i.e., individual) factors (Sampson 1988; Coleman 1986). As Buckner (1988) has also argued, the development of psychological instruments designed to measure attributes of individuals has vastly outpaced formulation of instruments to assess attributes of collectives (772). Even if researchers do adopt a macrolevel or contextual perspective, it is especially difficult to obtain a survey sample of residents that is sufficient to construct macrolevel measures across a large number of communities that vary along urban-rural dimension (Sampson 1988). Accordingly, when macrolevel factors are examined, sociologists must rely on census data that usually provide inadequate measures of theoretical interest. And while ethnographies continue to provide rich substantive accounts of community processes (e.g., Rieder 1985), they tend to preclude analysis of quantitative data on macrolevel variations. To address these limitations, I bridge microand macrolevels of community analysis through a two-stage, multilevel approach to study of local social bonds and community social cohesion. More specifically, I replicate and extend theoretically Sampson's recent model (1988) of multilevel dimensions of community social organization. At macro (betweencommunity) level, I examine structural antecedents of friendship/acquaintanceship and neighborhood social cohesion. At micro (i.e., individual) level, I employ a contextual framework by examining simultaneous effects of both individual and community factors on individual-level dimensions of community attachment. This research design is made possible by analysis of a national sample of approximately 11,000 residents of over 500 localities in England and Wales in 1984. Prior Research on Community Social Organization The systemic model of community social organization conceptualizes local community as a complex system of friendship and kinship networks and formal and informal associational rooted in family life and ongoing socialization processes (Kasarda & Janowitz 1974). The term systemic highlights This content downloaded from 207.46.13.111 on Tue, 09 Aug 2016 04:59:16 UTC All use subject to http://about.jstor.org/terms Community Social Organization / 45 theoretical focus on system of social embedded within ecological, institutional, and normative community structures. The basic hypothesis derived from this conceptualization is that length of residence is key exogenous factor that influences attitudes and behavior toward community. As Kasarda and Janowitz (1974) argue: Since assimilation of newcomers into social fabric of local communities is necessarily a temporal process, residential mobility operates as a barrier to development of extensive friendship and kinship bonds and widespread local associational ties (330). Using survey data from a national sample of England, Kasarda and Janowitz (1974) provide empirical support for systemic model. Independent of urbanization, density, and other factors (e.g., socioeconomic status [SES], life cycle), length of residence was positively related to an individual's local friendships, community sentiment, and participation in local affairs. In fact, neither large population size nor high density significantly weakened local social bonds, leading Kasarda and Janowitz to question major contrasting perspective on communities that they labeled linear-development model. Stemming from classical tradition best exemplified by Wirth (1938), linear-development model focuses on increased size and density as primary exogenous factors hypothesized to weaken community kinship and friendship bonds, social participation in local affairs, and attachment to community (Wirth 1938; for more extensive discussion, see Fischer 1982 and Tittle 1989). Kasarda and Janowitz's (1974) rejection of linear-development model rested on explaining individual-level variations in community attachment as a function of an individual's length of residence. Sampson (1988) recently tested an extended version of basic systemic model by examining length of residence at both individual and community levels. In terms of latter, community residential instability was posited as a major structural impediment to community-level social organization (see also Kasarda & Janowitz 1974; Kornhauser 1978). The focus on community as an important unit of analysis in its own right is consistent with a core idea of human ecological paradigm: locality-based social networks and collective attachment undergird social organization of human communities (Hawley 1950; Hunter 1974). The systemic model considered by Sampson (1988) also points to role of community-level residential stability in promoting an individual's social integration into community regardless of his/her own length of residence. Residents of areas characterized by frequent population turnover face formidable constraints compared to residents of stable areas. In all likelihood, an individual has fewer opportunities to form friendships and participate in local affairs in areas of high residential turnover. And if residential mobility increases institutional instability (Kornhauser 1978), then individuals in unstable communities will find fewer opportunities for organizational contact. The motivation to form local friendships may also be reduced in areas of high population turnover since residents know they will not last (Freudenburg 1986). Moreover, neighborhood instability and population change may reduce individual satisfaction with community, both for long-term residents and