PAPER CONTRIBUTION I
Public Health and Safety in Context: Lessons from Community-Level Theory on Social Capital*
A long-standing body of research has underscored the association of multiple health-related outcomes with the social and economic characteristics of community contexts. As reviewed below, for example, child and adolescent outcomes correlated with the ecological concentration of poverty include infant mortality, low birthweight, child maltreatment, and adolescent violence; health risks for adults include depression, homicide victimization, cardiovascular disease, and all-cause mortality. At the other end of the spectrum, communities with high socioeconomic status appear to promote the health of children and adults.
Why are so many health-related outcomes concentrated ecologically? A new generation of neighborhood-level research has emerged in recent years to tackle this question. This research has extended across the disciplines of sociology, public health, psychology, economics, and political science. A visible example is provided by the recent publication of two research volumes supported by the Social Science Research Council on “neighborhood effects” (Brooks-Gunn et al., 1997a,b). In public health, renewed interest in neighborhood envi-
Dr. Sampson is Lucy Flower Professor of Sociology at the University of Chicago and Dr. Morenoff is assistant professor of sociology and faculty associate at the Population Studies Center at the University of Michigan. This paper was prepared for the symposium “Capitalizing on Social Science and Behavioral Research to Improve the Public' s Health,” the Institute of Medicine and the Commission on Behavioral and Social Sciences and Education of the National Research Council, Atlanta, Georgia, February 2–3, 2000.
We thank the Institute of Medicine panel along with discussants at the Atlanta symposium for helpful comments. Portions of this paper draw from Sampson et al. (1999) and Sampson (1999a,b,c).
ronments has resulted in the publication of two recent review articles (Roberts, 1999; Yen and Syme, 1999) and a collection of classic and contemporary works on neighborhoods and health (Kawachi et al., 2000).
The goal of this paper is to make sense of current knowledge on health and public safety in local communities. We do so by drawing on the concept of “social capital” to unpack what it is about neighborhoods, above and beyond the status and attributes of the individuals who live there, that might lead to various health outcomes. Indeed, despite the wide range of evidence documenting relationships between community socioeconomic status and health, we still know very little about what these associations mean. As we shall elaborate, the basic idea of social capital provides a possible clue. According to James Coleman, social capital is a resource stemming from the structure of social relationships, which in turn facilitates the achievement of specific goals (Coleman 1990:300). Such resources, be they actual or potential, are often linked to durable social networks (Bourdieu 1986:249). Putnam defines social capital more expansively to include not only the networks themselves, but also shared norms and mutual trust, which “facilitate coordination and cooperation for mutual benefit” (1993:36).
In short, social capital is a resource that is realized through relationships (Coleman 1990:304), as compared to physical capital, which takes observable material form, and human capital, which rests in the skills and knowledge acquired by an individual. Whatever the specific formulation, social capital is not an attribute of individuals but rather a property of a social structure, such as an organization, an extended family, or a community (Coleman 1990; Bourdieu 1986). This paper employs the concept of social capital as a mode of inquiry into why community environments may be consequential for health. We review theoretical formulations on what social capital means at the community level, criticisms and weaknesses of the concept, and recent theoretical innovations. We also highlight research frontiers and promising community-level strategies that put social capital to work in promoting health. Before articulating social capital in more detail, we first note the regularities in prior empirical research that motivate its consideration.
A BRIEF HISTORY OF COMMUNITY HEALTH RESEARCH
The study of community environments and health-related outcomes actually has a long lineage in both sociology and public health. In sociology, the urban ecological approach of the “Chicago School” brought neighborhood-centered research to the fore of the discipline during the early 20th Century.3 One influen-
We set aside a detailed discussion of equally longstanding debates on how to define “neighborhood.” The traditional definition of neighborhood, as proposed by Robert Park, referred to an ecological subsection of a larger community—a collection of both people and institutions occupying a spatially defined area that is conditioned by a set of ecological, cultural, and political forces (Park, 1916:147–154). Neighborhoods are ecological units nested within successively larger communities (1916:114). That is, there is no one neighborhood, but many neighborhoods that vary in size and complexity depending on the social phenomenon of interest and the ecological structure of the larger community. In most cities there exist local community or city planning areas that, although relatively large, usually have well-known names and borders such as freeways, parks, and major streets. For example, Chicago has 77 local community areas averaging about 40,000 persons that were designed to correspond to socially meaningful and natural geographic boundaries. Some boundaries have undergone change over time, but these areas are widely recognized by administrative agencies and local institutions. Census tracts refer to smaller and more socially homogeneous areas of roughly 3,000–5,000 residents on average; their boundaries are usually, but not always, drawn to take into account major streets, parks, and other geographical features. A third and even smaller area more likely to approximate common notions of a “neighborhood ” is the block group—a set of blocks averaging approximately 1,000 residents. Although widely used in empirical research, the fact remains that community or planning areas, census tracts, and block groups all offer flawed operational definitions of neighborhood. In addition, social networks are potentially boundless in physical space.
tial framework in this tradition was Shaw and McKay's (1942) theory of neighborhood social disorganization. Although focusing on delinquency, Shaw and McKay constructed a general framework for understanding how community processes relate to a wide range of outcomes, including health. Community social disorganization was conceptualized as the inability of a community structure to realize the common values of its residents and maintain effective social controls. Social control refers to the capacity of a social unit to regulate itself according to desired principles—to realize collective, as opposed to forced, goals (Janowitz 1975:82, 87). Common ends include the desire of residents to live in neighborhoods characterized by economic sufficiency, good schools, adequate housing, freedom from predatory crime, and a generally healthy environment. The capacity to achieve such goals is linked to both formal, purposive efforts to achieve social regulation through institutional means, and to informal role relationships established for other purposes (Kornhauser 1978).
Shaw and McKay argued that delinquency was not an isolated phenomenon, and showed in their research the association between communities and health. In particular, Chicago neighborhoods characterized by poverty, residential instability, and dilapidated housing were found to suffer disproportionately high rates of infant mortality, delinquency, crime, low birth weight, tuberculosis, physical abuse, and other factors detrimental to health (1969:106). In another seminal study, Faris and Dunham (1965, 1939) applied the idea of social disorganization to mental health, arguing that more disorganized areas had higher rates of hospitalization for mental disorders. Interestingly, both Shaw and McKay and Faris and Dunham observed that high rates of adverse outcomes tended to persist in the same communities over time, despite the movement of different population
groups through them.4 From these findings, the Chicago School sociologists concluded that neighborhoods possess relatively enduring features that transcend the idiosyncratic characteristics of particular ethnic groups that inhabit them—so-called “emergent properties.” In any event, what makes the Chicago-School framework relevant to present concerns is its theoretical emphasis on the characteristics of places rather than people.
Community environments were a focus of early epidemiological research as well. An exemplar in this tradition is the research of Goldberger and his colleagues on pellagra in Southern villages. In a classic study of family income and the incidence of pellagra, Goldberger et al. (1916) found that the probability of contracting pellagra was related not only to individual-level socioeconomic status but also to the availability of nutritional foods in villages. They amassed an impressive array of both individual- and village-level data relating to food supply and malnutrition that included village-level measures of the prevalence of retail grocery establishments and home-provided foods, and contrasts in the type of agriculture in farm areas surrounding the villages. Their study also provided some of the earliest evidence of an interaction between individual- and community-level health risks: in villages with fewer supplies of nutritious food, family income was “less efficient as a protective factor than in other similar localities with better conditions of food availability” (Goldberger et al., 1916:2707).
Although Goldberger may have been “rowing against the epidemiologic current” with his focus on the social context of individual risk factors (Schwartz et al., 1999:19), his work was nonetheless widely influential and illustrative of a much broader tradition of epidemiological research on the ecological context of health that dates back to the early 19th Century. As Susser (1998:609) argues, “Epidemiology, a population science, is in its essence ecological in the original biological sense of organisms in a multilevel interactive environment.”
The importance of an ecological perspective has, in our opinion, increased in recent years. Social characteristics continue to vary systematically across communities along dimensions of socioeconomic status (e.g., poverty, wealth, occupational attainment), family structure and life cycle (e.g., female-headed households, child density), residential stability (e.g., home ownership and tenure), and racial/ethnic composition (e.g., racial segregation). In fact, there is evidence that the concentration of poverty and inequality has increased in the 1980s and 1990s (Wilson 1996; Massey and Denton 1993). A prominent explanation for the clustering of multiple forms of social and economic disadvantage
The assumption of ecological stability has been called into question in recent decades. Ironically, many social scientists have returned to basic Chicago School principles to explain ecological transformations such as those brought on by white flight in the 1950s and the increasing concentration of poverty in the 1970s and 1980s (see Morenoff and Sampson, 1997).
is Wilson's (1987) theory of concentration effects arising from living in an impoverished neighborhood. Wilson argues that the social transformation of innercity areas in recent decades (due primarily to out-migration of the middle class and industrial restructuring that depleted the demand for unskilled labor) has increased the concentration of the most disadvantaged segments of the population —especially black, poor, female-headed families with children. Massey and Denton (1993) describe how the interaction between rising economic dislocation and persistent racial residential segregation created a set of structural circumstances that reinforce the effects of economic deprivation. Massey (1996) notes also the growing geographic concentration of affluence, suggesting an increasingly bifurcated society by wealth.
There is an unfortunate parallel in health regarding ecological concentration; the “co-morbidity” or spatial clustering of homicide, infant mortality, low birth weight, accidental injury, and suicide continues to the present day (e.g., Wallace and Wallace 1990; Almgren et al., 1998). In the period 1995–1996, for example, data from the city of Chicago reveal that census tracts with high homicide rates tend to be spatially contiguous to other tracts high in homicide. Perhaps more interesting, more than 75% of such tracts also contain a high level of clustering for low birth weight and infant mortality, and more than half for accidental injuries (Sampson 1999a). Suicide is more distinct although even here the spatial clustering is significant. The ecological concentration of homicide, low birth weight, infant mortality, and injury indicates that there may be geographic “hot spots” for a number of unhealthy outcomes. The range of child and adolescent outcomes correlated with multiple forms of concentrated disadvantage is quite wide, and includes infant mortality, low birth weight, teenage childbearing, low academic achievement and educational failure, child maltreatment, and delinquency (see Brooks-Gunn et al., 1997a,b).
A growing body of research has also examined community characteristics and individual-level health. Although the evidence is mixed and the magnitudes often small, a number of studies have linked health outcomes to neighborhood context even when individual attributes and behaviors are taken into account (Robert 1999), including coronary risk factors and heart disease mortality (DiezRoux et al., 1997; LeClere et al., 1997), low birth weight (Roberts 1997; O'Campo et al., 1997; German 1999), smoking (Duncan et al., 1999; Diehr et al., 1993), morbidity (Robert 1998), all-cause mortality (Haan et al., 1987; Anderson et al., 1997), and self-reported health (Jones and Duncan 1995; Sooman and Macintyre 1995). Recent analyses of the longitudinal Alameda County Health study in Northern California, for example, revealed that self-reported fair/poor health was 70 percent higher for residents of concentrated poverty areas than for residents of nonpoverty areas, independent of age, sex, income, education, smoking status, body mass index, and alcohol consumption (Yen and Kaplan 1999a). In a related study, age and sex-adjusted odds for mortality were more than 50% higher (odds ratio = 1.58) for residents in areas characterized by poverty and deteriorated housing, after adjusting for income, race/ethnicity, smoking, body mass index, alcohol consumption, and perceived health status
(Yen and Kaplan 1999b). Such patterns are not limited to the United States. A multilevel study in Sweden found a similar elevated risk of poor health for residents of lower socioeconomic-status communities, controlling for age, sex, education, body mass index, smoking, and physical activity (Malmstrom et al., 1999).
Correlational and observational studies suffer well-known weaknesses with respect to making causal inferences. Much research on communities and health implicitly assumes that the social characteristics of neighborhood environments cannot be reduced to the compositional characteristics of individual residents, just as the liquidity of water is a property that does not pertain to the individual molecules of hydrogen and oxygen (Schwartz et al., 1999:26–27). Alternatively, however, it may be that individuals with poor health selectively migrate to, or are left behind in, poor neighborhoods. Under this interpretation, the observed correlations between community characteristics and health may simply be a reflection of the unmeasured processes through which individuals sort themselves into different neighborhoods (Manski 1995).
To address this “reflection problem,” researchers have begun to explore community-level effects on health outcomes with experimental and quasiexperimental research designs. One such example is found in the “Moving to Opportunity” (MTO) program, a series of housing experiments in five cities that randomly assigned housing-project residents to one of three groups: an experimental group receiving housing subsidies to move into low-poverty neighborhoods, a group receiving conventional (Section 8) housing assistance, and a control group receiving no special assistance. A study from the Boston MTO site showed that children of mothers in the experimental group had significantly lower prevalence of injuries, asthma attacks, and personal victimization during follow-up. The move to low-poverty neighborhoods also resulted in significant improvements in the general health status and mental health of household heads (Katz et al., 1999). Because the experimental design was used to control individual-level risk factors, a reasonable inference from these studies is that an improvement in community socioeconomic environment leads to better health and behavioral outcomes.
In summary, research in the social and behavioral science has established a reasonably consistent set of findings relevant to the community context of health. Although causality and magnitude are still at issue, there seems to be broad agreement that (a) there is considerable inequality between neighborhoods and local communities along multiple dimensions of socioeconomic status, (b) a number of health problems tend to cluster together in geographically defined ecological units such as neighborhoods or local community areas, (c) these two phenomena are themselves related—community-level predictors common to many health-related outcomes include concentrated poverty and/or affluence, racial segregation, family disruption, residential instability, and poor quality
housing, and (d) the relationship between community context and health outcomes—especially all-cause mortality, depression, and violence —appears to maintain when controls are introduced for individual-level risk factors. An association between the social environment and individual-level health has emerged in experimental studies as well.
Despite this suggestive body of evidence on the association between community social contexts and health, the question remains as to why. Until recently, answers from the dominant perspectives in social science and epidemiology were not forthcoming given their (ironically) individualistic bent in the modern era. In sociology, the Chicago School paradigm, with its emphasis on ecological and social systems, gave way in the 1950s and 1960s to a focus on estimating individual parameters in national surveys (Coleman 1994). Modern epidemiology became dominated by an analogous “risk factor paradigm” that elevated individual-level causes of disease and generally demoted more “upstream” social and economic antecedents of health (Schwartz et al., 1999; Susser and Susser 1996a, b; Susser 1998; Diez-Roux 1998). The theoretical framework that underlies this paradigm has been described as “biomedical individualism” that “considers social determinants of disease to be at best secondary (if not irrelevant), and views populations simply as the sum of individuals and population patterns of disease as simply reflective of individual cases” (Krieger 1994).
In the last several years, notable programmatic essays assailing the risk factor paradigm have appeared in major public health journals, with titles such as “Does Risk Factor Epidemiology Put Epidemiology at Risk?” (Susser 1998); “The Failure of Academic Epidemiology: Witness for the Prosecution ” (Shy 1997); “Bringing Context Back into Epidemiology” (Diez-Roux 1998); “The Fallacy of the Ecological Fallacy” (Schwartz 1994); and “The Logic in Ecological” (Susser 1994a; b). One theme that runs throughout these articles is the need for more research on the social context of health risks. Some prominent epidemiologists—Krieger (1994) and Susser and Susser (1996a, b) —have called for a new paradigm centered around an ecological metaphor that would emphasize the broader context of individual risk factors, both at the macro level, with more attention to social environments, and at the micro level of molecular biology (Schwartz et al., 1999). Within sociology, Abbott (1997) makes a similar plea for a “contextualist ” paradigm, premised on the idea that “no social fact makes any sense abstracted from its context in social (and often geographic) space and social time” (1152). The need for a contextualist paradigm in health is, we would argue, clear. We thus turn to such a paradigm, that of social capital.
SOCIAL CAPITAL THEORY AND BEYOND
The spatial patterning of crime and health provides a potentially important clue in thinking about why it is that communities and larger collectivities might matter for health. Namely, if multiple and seemingly disparate health outcomes are linked together empirically across communities and are predicted by similar structural characteristics, there may be common underlying causes or mediating
mechanisms at the neighborhood level. For example, if “neighborhood effects” of concentrated poverty on health really exist, presumably they stem from social processes that involve collective aspects of neighborhood life (e.g., social cohesion, spatial diffusion, local support networks, informal social control). The theory of social capital addresses such social processes. (So too did the early Chicago School, if less rigorously. The connection between social disorganization and social capital is this: Communities high in social capital are better able to realize common values and maintain the social controls that foster public safety.)
Although he was not the originator of the idea, Coleman's writings on social capital elevated the concept to the forefront of social science research (1988, 1990).5 He argued that social capital takes a variety of different forms, all of which involve relationships among persons that facilitate the realization of goals (Coleman 1990:300). One form of social capital relevant to health is reciprocal obligations among neighbors, which create a source of “credit” that individuals can draw upon when in need of a favor (Coleman 1990:306–310). A second form of social capital occurs when social relations are used to exchange information (1990:310). Neighbors sometimes provide each other with advice, tips, or other types of information that might be costly to acquire elsewhere. A third form is intergenerational closure. When parents know the parents of their children's friends, they have the potential to observe the child's actions in different circumstances, talk to each other about the child, compare notes, and establish rules (Coleman 1988). Such intergenerational closure of local networks provides the parents and children with social capital of a collective nature. Coleman also argued that voluntary associations can generate social capital either intentionally, by generating community action around a specific purpose, or unintentionally, by creating new ties between people that are then used to facilitate other types of action. For example, parents who join school-based organizations like the PTA or a local school council may form new ties with some of their neighbors and decide to employ those ties towards non-educational pursuits, such as establishing an informal system of childcare.
Since Coleman introduced the idea to a wider audience, there has been an explosion of interest in social capital. The concept is currently so popular that the World Bank maintains an entire web site dedicated to social capital (http://www.worldbank.org/poverty/scapital/index.htm). Unfortunately, the growing popularity of social capital has led to a proliferation of its meanings. Even some proponents of social capital theory (e.g., Fortes 1998) worry that the concept has come to represent so many different things to different people that it is in danger of losing its distinctive social meaning. Part of this confusion stems
Earlier essays exploring the meaning of social capital can be found in the works of Jane Jacobs, George Romans, Glen Loury, and Pierre Bourdieu. For a more extensive discussion of the theoretical origins and development of social capital, see Fortes (1998), Sandefur and Laumann (1998), Woolcock (1998), Astone et al. (1999), and Kawachi and Berkman (2000).
from the way Coleman originally formulated the concept. Rather than undertaking a systematic analysis of how social capital is generated and what its consequences are, Coleman chose to illustrate its operation through vignettes. Recent work bearing on health has thus set out to clarify, extend, and operationalize the concept, a trend particularly evident in the fields of epidemiology, criminology, and urban sociology.
In epidemiology, the recent application of social capital to understanding variations in health has focused largely on norms that encourage cooperation and facilitate social cohesion. Drawing on Putnam (1993) perhaps more so than Coleman, Kawachi and Berkman (2000) argue that the core meaning of social capital is tied to the broader notion of social cohesion, which refers to the absence of social conflict, coupled with the presence of strong social bonds and mutual trust. There is a small but intriguing body of research linking cohesion to health-related outcomes, although usually at higher levels of aggregation than the neighborhood. For example, measures of social cohesion and trust have been found to predict mortality rates at the state level. Kawachi et al. (1997) reported that the level of distrust (the proportion of residents in each state agreeing that most people can't be trusted) was strongly correlated with the age-adjusted mortality rate (r = .79, p < .001). Lower levels of trust were associated with higher rates of most major causes of death, including coronary heart disease, unintentional injury, and cerebrovascular disease. A one standard deviation increase in trust was associated with about a 9% lower level of overall mortality. A negative association between levels of trust/cohesion and homicide rates was found at the state level as well (Kawachi et al., 1998). Kawachi et al. (1999) also found a relationship between state-level social capital and individual self-rated health. Controlling for individual risk factors (e.g., smoking, obesity, income, lack of access to health care), individuals living in states with low levels of social capital (low trust, reciprocity, and voluntary associations) exhibited a significantly greater risk of poor self-rated health.
A rich line of research on social networks in the fields of both health and urban sociology also shares a natural affinity with the surge of interest in neighborhood social capital. In particular, much evidence reveals that friendship ties and family social support networks promote individual health (see House et al., 1988; Berkman and Syme 1979). And contrary to the popular belief that metropolitan life has led inexorably to the decline of personal ties, sociological research has shown that while urbanites may be exposed to more unconventionality and diversity, they retain a set of personal support networks just like their suburban and rural counterparts (see e.g., Fischer 1982). The difference is that the social ties of city dwellers tend to be more dispersed spatially. In other words, social ties are alive and well in American cities, even if not concentrated geographically, and at the individual level they appear to be a protective factor in health.
At the same time, however, we would maintain that local social ties (whether urban or rural in nature) do not necessarily translate into high social capital at the neighborhood level. Wilson (1996), for example, argued that many
poor neighborhoods with strong social ties do not produce collective resources such as the social control of disorderly behavior. His research suggests that disadvantaged urban neighborhoods are places where dense webs of social ties among neighbors actually impede social organization: “(I)t appears that what many impoverished and dangerous neighborhoods have in common is a relatively high degree of social integration (high levels of local neighboring while being relatively isolated from contacts in broader mainstream society) and low levels of informal social control (feelings that they have little control over their immediate environment, including the environment's negative influences on their children)” (Wilson 1996:63–64). Although members of such communities share strong social linkages with one another, they remain socially isolated, in Wilson's terms, because their network ties do not extend beyond immediate social environs to include non-community members and institutions. The limits of tight-knit social bonds were recognized earlier in Granovetter's (1973) seminal essay on the strength of “weak ties” in obtaining job referrals. Whether or not weak ties promote health is unknown.
We believe that the research on cohesion and social ties reveals somewhat of a paradox. Namely, many urbanites have an abundance of friends and social support networks that are no longer organized in a parochial, local fashion. Moreover, urbanites whose “strong ties” are tightly restricted geographically may actually produce an environment that discourages collective responses to local problems. To address this issue, Sampson et al. (1997, 1999) have proposed a focus on mechanisms that facilitate social control without requiring strong ties or associations. As Warren (1975) noted, the common belief that neighborhoods have declined in importance as social units “is predicated on the assumption that neighborhood is exclusively a primary group and therefore should possess the ‘face-to-face,' intimate, affective relations which characterize all primary groups” (p. 50). Sampson et al. (1997) reject this outmoded assumption and highlight instead working trust and the shared willingness of local residents to intervene in support of public order. Personal ties notwithstanding, it is the linkage of mutual trust and shared expectations for intervening on behalf of the common good that defines the neighborhood context of what Sampson et al. (1997) term collective efficacy. Just as individuals vary in their capacity for efficacious action, they argue that so too do neighborhoods vary in their capacity to achieve common goals. Moreover, just as self-efficacy is situated rather than global (one has self-efficacy relative to a particular task or type of task), neighborhood efficacy exists relative to collective tasks such as maintaining public order.
Sampson et al. (1999) thus view social capital as referring to the resources or potential inherent in social networks, whereas collective efficacy is a task-specific construct that refers to shared expectations and mutual engagement by residents in local social control. Moving away from a focus on private ties, the term collective efficacy is meant to signify an emphasis on shared beliefs in a
neighborhood's conjoint capability for action to achieve an intended effect, and hence an active sense of engagement on the part of residents. As Bandura (1997) argues, the meaning of efficacy is captured in expectations about the exercise of control, elevating the “agentic” aspect of social life over a perspective centered on the accumulation of “stocks” of social resources. This conception of collective efficacy is consistent with the redefinition of social capital by Portes and Sensenbrenner in terms of “expectations for action within a collectivity” (1993:1323).
The theory of collective efficacy was tested in a survey of 8,782 residents of 343 Chicago neighborhoods in 1995. A five-item Likert-type scale measured shared expectations about “informal social control. ” Residents were asked about the likelihood that their neighbors could be counted on to take action if: (1) children were skipping school and hanging out on a street corner, (2) children were spray-painting graffiti on a local building, (3) children were showing disrespect to an adult, (4) a fight broke out in front of their house, and (5) the fire station closest to home was threatened with budget cuts. “Social cohesion/trust” was measured by asking respondents how strongly they agreed that “People around here are willing to help their neighbors”; “This is a close-knit neighborhood”; “People in this neighborhood can be trusted”; and (reverse coded): “People in this neighborhood generally don't get along with each other”; “People in this neighborhood do not share the same values.” Social cohesion and informal social control were strongly related across neighborhoods (r = .80), and were combined into a summary measure of “collective efficacy.”
Collective efficacy was associated with lower rates of violence, controlling for concentrated disadvantage, residential stability, immigrant concentration, and a set of individual-level characteristics (e.g., age, sex, SES, race/ethnicity, home ownership). Whether measured by official homicide events or violent victimization as reported by residents, neighborhoods high in collective efficacy had significantly lower rates of violence. This finding held up controlling for prior neighborhood violence that may have depressed later collective efficacy (e.g., because of fear); a two standard-deviation elevation in collective efficacy was associated with a 26 percent reduction in the expected homicide rate (1997: 922). Concentrated disadvantage and residential instability were also linked to lower collective efficacy, and the association of disadvantage and stability with violence was significantly reduced when collective efficacy was controlled. In particular, collective efficacy appears to be undermined by the concentration of disadvantage, racial segregation, family disruption, and residential instability (Sampson et al., 1997, 1999). The cross sectional nature of the data and the likelihood of reciprocality (crime may reduce collective efficacy) means that causal effects could not be reliably determined. Nonetheless, these patterns are consistent with the inference that neighborhood characteristics influence violence in part through the construct of neighborhood collective efficacy (see also Elliott et al., 1996).
Institutions and Public Control
A theory of social capital and collective efficacy should not ignore institutions, or the wider political environment in which local communities are embedded. Communities can exhibit intense private ties (e.g., among friends, kin) and perhaps even shared expectations yet still lack the institutional capacity to achieve social control (Hunter 1985; Woolcock 1998). The institutional component of social capital is the resource stock of neighborhood organizations and their linkages with other organizations, both within and outside the community (Sampson 1999b). Kornhauser (1978:79) argues that when the horizontal links among institutions within a community are weak, the capacity to defend local interests is weakened. Vertical integration is potentially more important. Similar to the idea of “bridging” social capital, Bursik and Grasmick (1993) highlight the importance of public control, defined as the capacity of local community organizations to obtain extralocal resources (e.g., police, fire protection; block grants; health services) that help sustain neighborhood stability and control. Hunter (1985) identifies the dilemma of public control in a civil society. The problem is that public control is provided mainly by institutions of the State, and we have seen a secular decline in public (citizenship) obligations in society accompanied by an increase in civil (individual) rights. This imbalance of collective obligations and individual rights undermines social control, and by implication, social capital. According to Hunter (1985), local communities must thus work together with forces of public control to achieve social order, principally through interdependence among private (family), parochial (neighborhood) and public (State) institutions such as the police and schools.
Any discussion of social capital needs to acknowledge its potential downside. After all, social networks can be drawn upon for negative as well as positive goals—the same strong social ties that benefit members of one particular group can be used to exclude others outside the group from sharing in those resources (Portes 1998:15). As but one example, racial exclusion is not desirable yet dense social networks have been used to facilitate it. As Sugrue's (1996) research on Detroit circa 1940–1970 revealed, strong neighborhood associations were exploited by whites to keep blacks from moving to white working-class areas.
Social capital (and by implication, collective efficacy) may thus be said to have a valence depending on the goal in question (Sandefur and Laumann 1998: 493). Recognizing the valence of social capital, Sampson et al. (1999) apply the nonexclusivity requirement of a social good (Coleman 1990:315–316) to judge whether neighborhood structures serve the collective needs of residents. We believe that resources such as neighborhood safety and healthy environments produce positive externalities (see Coleman 1990:250–251) that are consensually desired but problematically achieved, owing in large part to variabilities in structural constraints. Moreover, even if health and safety are consensually desired,
there may be conflicts over the setting of priorities when resources are limited. One can imagine, for example, that in some contexts a disagreement over whether a violence reduction program or a toxic waste cleanup deserves priority would lead to paralysis and not collective action. As with the existence of strong ties, consensus on ultimate goals does not automatically lead to enhanced outcomes. Like Sampson et al. (1999), then, we view social capital and collective efficacy not as some all-purpose elixir but as normatively situated and endogenous to specific structural contexts (Portes 1998).6
RECENT DEVELOPMENTS IN NEIGHBORHOOD RESEARCH
Consideration of neighborhoods and health should not be confined strictly to social capital, of course. Two developments in particular bear noting—research on spatial inequality and routine activities. Although distinct conceptually from social capital, we believe there are sufficient grounds for integrating these developments to better understand the ecology of health.
Metropolitan-Wide and Spatial Inequality
Research on the political economy of American cities has shown that the stratification of places is shaped, both directly and indirectly, by the extralocal decisions of public officials and businesses. For example, the decline of many central-city neighborhoods has been facilitated not only by individual preferences, as manifested in voluntary migration patterns, but by government decisions on public housing that concentrate the poor, incentives for suburban sprawl in the form of tax breaks for developers and private mortgage assistance, highway construction, economic disinvestment in central cities, and haphazard zoning on land use (Logan and Molotch 1987).
The embeddedness of neighborhoods within the larger system of citywide spatial dynamics is equally relevant (Sampson et al., 1999). Recent research on population change shows that population abandonment is driven as much by spatial diffusion processes (e.g., changes in proximity to violent crime) as by the internal characteristics of neighborhoods (Morenoff and Sampson 1997). In particular, housing decisions are often made by assessing the quality of neighborhoods relative to what is happening in surrounding areas. Parents with young children appear quite sensitive to the relative location of neighborhoods and schools in addition to their internal characteristics. Spatial diffusion processes for dimensions of social capital are even more likely, mainly because social
Although we are critical of the undisciplined appropriation of social capital rhetoric by many health researchers, the renewed attention to what are in fact old ideas about the social nature of life does have an upside. Namely, the excessively individualistic paradigm of “risk factors” has been challenged in an unprecedented way, opening a window of opportunity for consideration of macrolevel strategies for community intervention. We return to this issue in the final section of this paper.
networks and exchange processes unfold across the artificial boundaries of analytically defined neighborhoods. A neighborhood-level perspective on health cannot afford to ignore the relative geographic position of neighborhoods and how that bears on internal dimensions of social capital. The importance of spatial externalities is shown by the finding that ecological proximity to areas high in collective efficacy bestows an advantage above and beyond the structural characteristics of a given neighborhood (Sampson et al., 1999).
A concern with ecology suggests another often-overlooked mechanism in discussions of neighborhood effects—how land use patterns and the ecological distributions of daily routine activities bear on well being. The location of schools, the mix of residential with commercial land use (e.g., strip malls, bars), public transportation nodes, and large flows of nighttime visitors, for example, are relevant to organizing how and when children come into contact with other peers, adults, and nonresident activity. The “routine activities” perspective in criminology provides the important insight that predatory violence requires the intersection in time and space of motivated offenders, suitable targets, and the absence of capable guardians (Cohen and Felson 1979). Rooted in social control theory, the routine activity approach assumes a steady supply of motivated “offenders,” and focuses on how targets of opportunity and guardianship combine to explain criminal events. This strategy has appeal in thinking about a range of adolescent health behaviors, such as drinking, early sexual behavior, smoking, and “hanging out, ” that reflect natural desires yet can yield negative outcomes both personally and for others (e.g., teen childbearing, low birth weight, low achievement, poor health). For example, not only do mixed-use neighborhoods offer greater opportunities for expropriative crime, they offer increased opportunity for children to congregate outside the home in places conducive to peer-group influence (Sampson 1999a,b).
In short, because illegal and “deviant” activities feed on the spatial and temporal structure of routine legal activities (e.g., transportation, work, entertainment, and shopping), the differential land use of neighborhoods is a key to comprehending the ecological distribution of situations and opportunities conducive to a wide range of potentially adverse behaviors. In particular, the ecological placements of bars, liquor stores, strip-mall shopping outlets, subway stops, and unsupervised play spaces play a direct role in the distribution of high-risk situations. The ecology of routine activities is not usually thought of in the “neighborhood effects” literature, much less as a mechanism. We suggest, however, that it holds considerable promise as an explanatory factor, especially in combination with a theory of social capital. For instance, decisions to locate high-risk businesses are often targeted precisely to lower-income communities known to lack the organizational capacity to resist. Thus one arena where collective efficacy is likely to matter is in the differential ability of neighborhoods
to organize against local threats such as disorderly bars, licensing of new liquor stores, and the mixing of strip malls with residences and schools.
METHODOLOGICAL CHALLENGES AND RESEARCH FRONTIERS
Despite promising leads from existing research, several limitations must be addressed if scientific knowledge on social capital and health is to progress. First, aside from the experimental designs discussed above, neighborhood research has been largely silent on the issue of differential residential selection of individuals. This selection issue raises not only a methodological challenge to causal inference but also important substantive questions about how individuals sort themselves into neighborhoods. This sorting process surely has a bearing on how social capital gets produced. Closely related is the issue of “endogeneity” bias -does social capital affect health, or does health status determine the need and capacity for a neighborhood to produce social capital?
Second, health environments are not limited to geographical communities. Families, the workplace, religious institutions, and peer groups, to name just a few, generate their own collective properties that bear on health. As noted earlier, for example, friendship ties and family social support networks have been found to promote individual health (Berkman and Syme 1979). Nor are the relevant health environments limited to urban settings and areas of disadvantage. Most of the United States' population lives in suburban areas, and the relationship of socioeconomic status and health holds at the upper end of the socioeconomic distribution as well as the lower end (Robert 1999). Yet much of the extant research literature is limited to the study of poverty in inner-city communities, underscoring the need to assess suburban and rural contexts.
Third, there is a need to further develop multi-level methodologies for contextually based research. Health data collected at nested levels of aggregation (e.g., neighborhood, city, state) pose important challenges to the standard analytic procedures that are prevalent among health researchers. In particular, the use of multi-level models has yet to fully incorporate the analysis of spatial interdependence. If social capital is a public good then there is reason to believe that its benefits may diffuse over arbitrary neighborhood boundaries. Spatial processes may also influence some health-related behaviors, such as drug use and utilization of health services. A methodological challenge is thus to integrate multi-level methods with spatial dynamics.
A fourth research frontier is to build strategies for the direct measurement of social mechanisms and collective properties hypothesized to predict health (Mayer and Jencks 1989). Raudenbush and Sampson (1999) argue that while interest in the social sciences has turned to an integrated scientific approach that emphasizes individual factors in social context, a mismatch has arisen in the quality of measures. Decades of psychometric and biological research have produced individual-level measures that often have excellent statistical properties. By contrast, much less is known about measures of ecological settings. Neigh-
borhood-level research is dominated by poverty and other demographic characteristics drawn from census data or other government statistics that do not provide information on the collective properties of administrative units.
A major research frontier is thus methodological—the science of ecological assessment (“ecometrics”) of social environments relevant to health (Raudenbush and Sampson 1999). A major component of ecometrics is the development of systematic procedures for directly measuring community social processes, such as in population-based health surveys and systematic social observation of community environments. For example, the latter approach has used videotaping techniques to capture aspects of microcommunity environments (e.g., street-blocks) that bear on health risks (e.g., garbage in the streets, public intoxication, unsafe housing). Lochner et al. (1999) suggest other inventive measures of mutual trust and social cohesion that can be obtained through direct observation, such as the proportion of gas stations that require customers to pay before they pump their gas. Field-based experimental techniques, such as dropping stamped envelopes on the street and counting how many get returned, could also be used in communities to observe collective efficacy “in action.”
IMPLICATIONS AND DIRECTIONS
Consideration of the collective properties of social environments promises a deeper understanding of the etiology of health outcomes and the development of community-based prevention strategies. Indeed, the health sciences can “capitalize on social capital” by thinking creatively about the implications of extant research for community-based prevention strategies. Because community contexts are important units of analysis in their own right, we suggest the need for concrete “community-level” strategies that have been neglected in the health field. We thus turn to some possible directions for future research and community intervention efforts.
Data Collection and Methodology
An important “first step” in fostering intervention from a social capital perspective would be to support the systematic collection of benchmark data on social environments that can be compared across communities (see Sampson 1999c). The goal would be to develop a standardized approach to the collection and dissemination of data that individual communities could use to evaluate where they stand in regard to national and/or regional norms. Similar to school “report cards” that are used to track the progress of educational reform, a standardized approach to assessing collective properties would eventually allow
cities and local communities to gauge how well or poorly they are doing on a variety of health-related dimensions.7
An exemplar that might be used as a framework on which to build is the “Sustainable Seattle” project, where some 40 indicators have been collected for use as a benchmark to gauge the progress of Seattle in meeting various goals of public and civic health (Sampson 1999c). An innovative combination of archival records, census data, and surveys have been used to compile sustainability trends across five basic areas—environment (e.g., air quality), population and resources (e.g., fuel consumption), economy (e.g., housing affordability, poverty), youth and education (e.g., high school graduation, literacy), and community health (e.g., low birth weight, neighborliness). The Leaders Roundtable in Portland, Oregon has undertaken a similarly ambitious initiative (The Caring Community) that has collected data on community health using a combination of focus groups, surveys, key stakeholder interviews, and document reviews. We also noted above some strategies for new data collection involving social observation and field experiments. If measurement standards and systematic procedures at the national level could be developed, communities could then use benchmark data to develop early warning signs with respect to changes in the quality of health environments.
Alongside such data collection, there is a need for strategic investment in methodologies that are central to building an infrastructure capable of supporting the assessment and analysis of community social capital on a systematic and flexible scale. In addition to multi-level spatial methods, a practical move is to invest in Geographical Information Systems (GIS) and support the geographical linkage of ongoing data collection efforts in the health sciences. For instance, data on health outcomes can now be linked virtually in real time to address-level data bases on employment, density of liquor stores, mixed land use, and building code violations. Such “gee-coding” would support the ability to use existing health records to construct community health profiles, thereby aiding in the development of benchmark standards. A principal advantage of GIS is community profiling and the ability to overlay multiple health-related phenomena (e.g., deaths, cancer clusters, and accident hot spots) in time and space (see also below).
Many urban communities across the U.S. are witnessing unprecedented changes in their social environments resulting from the devolution of public housing. Especially in large cities, families are being relocated and entire housing projects are being dispersed. The quasi-experimental nature of these changes provides opportunities to learn about the connection between health outcomes
Robert Putnam has recently proposed launching just such a benchmark survey, both nationwide and in about 40 American communities, with the goal of assessing baseline levels of social capital. The possibilities for coordinating between his and other efforts seem rich with promise.
and environmental change. We noted some early evidence on the significance of such changes, and recommend that, wherever possible, health considerations be taken into account in ongoing and new evaluations. The experimental nature of many housing-related interventions provides strategic leverage in addressing individual selection bias, hence improving our understanding of the connection between communities and health in a way not possible with passive, observational designs. By integrating “ecometric” strategies for collecting theoretically relevant data on the collective properties of social environments with the random assignment of individuals to new social contexts, researchers are in a better position to sort out selection mechanisms and social causation mechanisms in health outcomes.
Finally, our paper implies that it is possible to focus on designing prevention strategies from the lens of research on social capital and communities. Traditional thinking about disease has emphasized behavioral change among individuals as a means to reduce disease risk. For example, smoking interventions have targeted smokers and include hypnosis, smoking cessation programs, and nicotine patches. Environmental approaches look instead to macrolevel factors such as taxation policies, regulation of smoking in public places, and restriction of advertising in places frequented by adolescents (Yen and Syme 1999). Such approaches appear to have had notable successes in reducing the aggregate level of cigarette consumption. It follows that community-level prevention that attempts to change places and social environments rather than people may yield similar payoffs that complement traditional individual and disease-specific approaches. Although scant, there is some evidence that community-based interventions may be successful in promoting prenatal health care and children 's health. An evaluation by the National Academy of Sciences (1981:58) reported that community-based interventions were significantly associated with lower rates of infant mortality and premature birth. Similarly, Wallace and Wallace (1990:417–418) argue that governmental community intervention helped to reduce the infant mortality rate in New York from 1966 to 1973. Hence there is reason to believe that connecting health initiatives for children with strategies for community social organization is promising (Sampson 1999a).
Drawing on social capital and routine activity theory as explicated in this paper, we would argue that advances in computer mapping technology be used to promote safety by exploiting information on ecological “hot spots.” In Chicago, Block (1991) pioneered the use of what is termed an “early warning system” for gang homicides. By plotting each homicide incident and using mapping and statistical clustering procedures, the early warning system allows police to identify potential neighborhood crisis areas at high risk for suffering a “spurt” of gang violence. With rapid dissemination of information, police can intervene in hot spots to quell emerging trouble. Places may also be modified or put under periodic surveillance to reduce the opportunities for trouble to occur. Sherman
and colleagues (Sherman 1994; Sherman et al., 1989:48) have reviewed “hot-spot” neighborhood interventions such as differential patrol allocations by place, selective revocation of bar licenses, and swift removal of vacant “crack” houses. The idea of hot spots suggests a neighborhood-level response that may be more effective than policies that simply target individuals or families. By responding proactively to neighborhoods and places that disproportionately generate crimes and adverse health events (e.g., high incidence of heat related deaths, alcohol fatalities), intervention strategies can more efficiently stave off “epidemics ” and their spatial diffusion. Also, neighborhood strategies to monitor the ecological placements of bars, liquor stores, strip-mall shopping outlets, subway stops, and unsupervised play spaces may play an important role in controlling the distribution of high-risk situations for crimes and accidents.
Another intervention strategy implied by this paper is to foster activities and policies that directly promote social capital by encouraging community social interaction, in the hopes that it will in turn lead to greater cohesion, shared expectations, and social control. Such initiatives include both “bottom-up” and “top-down” approaches (Kawachi and Berkman 2000). Community organizing through voluntary associations is a prime example of a bottom-up approach. In the field of crime prevention, these efforts often take the form of collective surveillance programs such as neighborhood watch and citizen patrols. Unfortunately, there is no convincing evidence that attempts at informal community surveillance can successfully “implant” social organization in neighborhoods where such processes are naturally lacking, but there have also been very few rigorous evaluations of such efforts (Rosenbaum et al., 1998:51). Social capital theory suggests that the bottom up approach would be more viable if community organizations form partnerships with other organizations both inside and outside their community (Meares and Kahan 1998).
There is some encouraging news in this regard from the evaluation of community policing interventions designed to increase the involvement of local residents, especially in those communities that have called upon faith-based organizations to be partners in the co-creation of social order (Meares and Kahan 1998). One of the major goals of community policing is for the police to act as a catalyst in sparking among residents a sense of local ownership over public space and greater activation of informal social control. An organizational strategy designed to accomplish this outcome is the “beat meeting ” —regularly scheduled meetings of the police with residents of their beats. Early evidence from the Chicago Alternative Policing Strategy (CAPS) suggests that beat meetings were one of the most visible and unique features of community policing. About 25 residents and five officers attended per meeting, with attendance highest in African-American and minority neighborhoods. Skogan and Hartnett's evaluation (1997:160) estimated that residents turned up on almost 15,000 occasions to discuss local problems with the police. To be sure, the news is not all positive. Skogan and Hartnett (1997:125, 130) found that the police took the lead in almost all beat meetings; despite much prodding, it was difficult to sustain resident input and to induce collective problem solving among residents.
Another model for encouraging grassroots participation in community initiatives is for public health researchers to collaborate with community members and organizational representatives in all aspects of the research process, in what has become known as “community-based research” (Minkler and Wallerstein 1998; Israel et al., 1998; Wallerstein 1999). One of the core ideas of this approach is that the research process can become a vehicle through which communities empower themselves (and thus become more collectively efficacious) by promoting “the reciprocal transfer of knowledge, skills, capacity, and power” (Israel et al., 1989:179) and by facilitating the formation of linkages between local agencies and organizations. As is the case in the crime prevention literature, there has been a shortage of high-quality evaluations of such efforts, making their success difficult to gauge even though the idea of building upon existing community resources and encouraging linkages between institutions is theoretically compelling.
In sum, community-level interventions to increase neighborhood “social capital” are hard to implement and have achieved only limited success in the areas that need them the most—poor, unstable neighborhoods with high crime rates (Hope 1996; Skogan and Hartnett 1997). Short-run interventions that try to change isolated or specific behaviors without confronting their common antecedents are also susceptible to failure. The paradox is that self-help strategies to “build community” give priority to the very activities undermined by the social isolation and spatial vulnerability of unstable and economically deprived neighborhoods (Hope 1996:24, 51). Neglecting the spatial dynamics of neighborhood social organization and the vertical connections (or lack thereof) that residents have to extracommunity resources obscures the structural backdrop to social capital (Sampson et al., 1999; Sampson 1999a). Overall, then, social capital is at once promising and yet no panacea, and accordingly should not be uncritically adopted as an isolated intervention strategy. “Top-down” or systemic policies, such as the promotion of housing-based neighborhood stabilization, the deconcentration of poverty, code enforcement, improvement of municipal health services, and fostering the community 's organizational base, need to be creatively linked to “bottom-up” strategies that involve local residents and institutions in partnership.
Almgrem, Gunnar, Avery Guest, George Imerwahr, and Michael Spittel. 1998. “Joblessness, Family Disruption, and Violent Death in Chicago, 1970 –1990. Social Forces 76:1465–1493.
Anderson, Roger T., Paul Sorlie, Eric Backlund, Norman Johnson, and George A. Kaplan. 1997. “Mortality Effects of Community Socioeconomic Status.” Epidemiology 8:42–47.
Astone, Nan, C.Nathanson, Robert Schoen, and Young Kim. 1999. “Family Demography, Social Theory, and Investment in Social Capital. ” Population Development and Review 25:1–31.
Bandura, Albert. 1997. Self Efficacy: The Exercise of Control. New York: W.H.Freeman.
Berkman, Lisa and S.L.Syme. 1979. “Social Networks, Host Resistance, and Mortality: A Nine-Year Follow-Up Study of Alameda County Residents.” American Journal of Epidemiology 109:186–204.
Block, Carolyn. 1991. “Early Warning System for Street Gang Violence Crisis Areas: Automated Hot Spot Identification in Law Enforcement.” Chicago: Illinois Criminal Justice Information Authority.
Bourdieu, Pierre. 1986. “The Forms of Capital.” Pp. 241–258 in Handbook of Theory and Research for the Sociology of Education, edited by J.Richardson. New York: Greenwood Press.
Brooks-Gunn, Jeanne, Greg J.Duncan, and Lawrence Aber (eds.). 1997a. Consequences of Growing up Poor. New York: Russell Sage Foundation.
—. 1997b. Neighborhood Poverty: Policy Implications in Studying Neighborhoods . New York: Russell Sage Foundation.
Bursik, Robert J. and Harold Grasmick. 1993. Neighborhoods and Crime: The Dimensions of Effective Community Control. New York: Lexington.
Cohen, Lawrence and Marcus Felson 1979. “Social Change and Crime Rate Trends: A Routine Activity Approach. ” American Sociological Review 44:588–608.
Coleman, James S. 1988. “Social Capital in the Creation of Human Capital.” American Journal of Sociology 94:S95–S120.
—. 1990. Foundations of Social Theory. Cambridge, MA: Harvard University Press.
—. 1994. “My Vision for Sociology.” Society.
Diehr, Paula, Thomas Koepsell, Allen Cheadle, Bruce M.Pstay, Edward Wagner, and Susan Curry. 1993. “Do Communities Differ in Health Behaviors?” Journal of Clinical Epidemiology 46:1141–1149.
Diez-Roux, Ana V. 1998. “Bringing Context Back into Epidemiology: Variables and Fallacies in Multilevel Analysis.” American Journal of Public Health 88:216–222.
Diez-Roux, Ana V., Javier Nieto, Carles Muntaner, Herman A.Tyroler, George W.Comstock, Eyal Shahar, Lawton S.Cooper, Robert L.Watson, and Moyses Szklo. 1997. “Neighborhood Environments and Coronary Heart Disease: A Multilevel Analysis.” American Journal of Epidemiology 146:48–63.
Duncan, Craig, Kelvyn Jones, and Graham Moon. 1999. “Smoking and Deprivation: Are There Neighborhood Effects?” Social Science and Medicine 48:497–505.
Elliott, Delbert, William Julius Wilson, David Huizinga, Robert J.Sampson, Amanda Elliott and Bruce Rankin. 1996. “Effects of Neighborhood Disadvantage on Adolescent Development.” Journal of Research in Crime and Delinquency 33:389–426.
Faris, Robert and Warren H.Dunham. 1939 (1965, 2nd edition). Mental Disorders in Urban Areas: An Ecological Study of Schizophrenia and Other Psychoses . Chicago: University of Chicago Press.
Fischer, Claude. 1982. To Dwell Among Friends: Personal Networks in Town and City. Chicago: University of Chicago Press.
Goldberger J, Wheeler GA, Sydenstrycker E. 1920. “A Study of the Relation of Family Income and Other Economic Factors to Pellagra Incidence in Seven Cotton Mill Villages of South Carolina in 1916.” Public Health Reports 35:2673–2714.
German, Bridget K. 1999. “Racial and Ethnic Variation in Low Birthweight in the United States: Individual and Contextual Determinants.” Health and Place 5:195–207.
Granovetter, Mark. 1973. “The Strength of Weak Ties.” American Journal of Sociology 78:360–380.
Haan, Mary, George A.Kaplan, and Terry Camacho. 1987. “Poverty and Health: Prospective Evidence from Alameda County Study. ” American Journal of Epidemiology 125(6):989–998.
Hope, Tim. 1995. “Community Crime Prevention.” Pp. 21–89 in Building a Safer Society edited by Michael Tonry and David Farrington. Chicago: University of Chicago Press.
House, James S., Debra Umberson, and Karl R.Landis. 1988. “Structures and Processesof Social Support.” Annual Review of Sociology 14:293–318.
Hunter, Albert. 1985. “Private, Parochial and Public Social Orders: The Problem of Crime and Incivility in Urban Communities.” In The Challenge of Social Control, edited by Gerald Suttles and Mayer Zald. Norwood, NJ: Ablex.
Israel, Barbara A., Amy J.Schulz, Edith A.Parker, and Adam B.Becker. 1998. “Review of Community-Based Research: Assessing Partnership Approaches to Improve Public Health.” Annual Review of Public Health 19:173–202.
Janowitz, Morris. 1975. “Sociological Theory and Social Control.” American Journal of Sociology 81:82–108.
Jones, Kelvyn and Craig Duncan. 1995. “Individuals and Their Ecologies: Analysing the Geography of Chronic Illness within a Multilevel Modeling Framework.” Health and Place 1(1):27–40
Katz, Lawrence, Jeffrey Kling, and Jeffrey Liebman. 1999. “Moving to Opportunity in Boston: Early Impacts of a Housing Mobility Program.” Paper presented at the conference, “Neighborhood Effects on Low Income Families,” Joint Center for Poverty Research, Northwestern University/University of Chicago, Chicago, IL.
Kawachi, Ichiro and Lisa F.Berkman. 2000. “Social Cohesion, Social Capital, andHealth.” In Social Epidemiology, edited by Lisa F.Berkman and Ichiro Kawachi. New York: Oxford University Press.
Kawachi, Ichiro, Bruce P.Kennedy, and Roberta Glass. 1999. “Social Capital and Self-Rated Health: A Contextual Analysis” American Journal of Public Health 89:1187– 1193.
Kawachi, Ichiro, Bruce Kennedy, Kimberly Lochner, and Deborah Prothrow-Stith. 1997. “Social Capital, Income Inequality, and Mortality.” American Journal of Public Health 87:1491–1498.
Kawachi, Ichiro. Kennedy, Bruce P.Wilkinson, Richard G. 2000. The Society and Population Health Reader: Income Inequality and Health. New York: The New Press.
Krieger Nancy. 1994. “Epidemiology and the Web of Causation: Has Anyone Seen the Spider? ” Social Science and Medicine 39:887–903.
Kornhauser, Ruth. 1978. Social Sources of Delinquency. Chicago: University of Chicago Press.
LeClere, Felicia B., Richard G.Rogers, and Kimberley Peters. 1998. “Neighborhood Social Context and Racial Differences in Women's Heart Disease Mortality.” Journal of Health and Social Behavior 39:91–107.
Lochner, Kimberly, Ichiro Kawachi, and Bruce P.Kennedy. 1999. “Social Capital: A Guide To Its Measurement.” Health and Place 5:259–270.
Logan, John and Harvey Molotch. 1987. Urban Fortunes: The Political Economy of Place. Berkeley, CA: University of California Press.
Malmstrom, Marianne, Jan Sundquist, and Sven-Erik Johansson. 1999. “Neighborhood Environment and Self-Reported Health Status.” American Journal of Public Health 89:1181–1186.
Manski, Charles. 1995. Identification Problems in the Social Sciences. Cambridge, MA: Harvard University Press.
Massey, Douglas S. 1996. “The Age of Extremes: Concentrated Affluence and Poverty in the Twenty-First Century.” Demography 33:395–412.
Massey, Douglas S. and Nancy Denton. 1993. American Apartheid: Segregation and the Making of the Underclass. Cambridge, MA: Harvard University Press.
Mayer, Susan E. and Christopher Jencks. 1989. “Growing up in Poor Neighborhoods: How Much Does it Matter? Science 243:1441–1445.
Meares, Tracey and Dan Kahan. 1998. “Law and (Norms of) Order in the Inner City.” Law and Society Review 32:805–838.
Minkler, Meredith and Nina Wallerstein. 1988. “Improving Health through Community Organization and Community Building: A Health Education Perspective.” Pp. 30–54in Community Organization and Community Building for Health. New Brunswick, NJ: Rutgers University Press.
Morenoff, Jeffrey and Robert J.Sampson. 1997. “Violent Crime and the Spatial Dynamics of Neighborhood Transition: Chicago, 1970–1990.” Social Forces, 76:31–64.
National Academy of Sciences. 1981. Toward a National Policy for Children and Families. Washington, D.C.: United States Government Printing Office.
O'Campo, Patricia, Xiaonan Xue, Mei-Cheng Wang, and Margaret O'Brien Caughy. 1997. “Neighborhood Risk Factors for Low Birthweight in Baltimore: A Multilevel Analysis.” American Journal of Public Health 87:1113–1118.
Park, Robert. 1916. “The City: Suggestions for the Investigations of Human Behavior in the Urban Environment.” American Journal of Sociology 20:577–612.
Portes, Alejandro. 1998. “Social Capital: Its Origins and Applications in Modern Sociology. ” Annual Review of Sociology 24:1–24.
Portes, Alejandro and Julia Sensenbrenner. 1993. “Embeddedness and Immigration: Notes on the Social Determinants of Economic Action.” American Journal of Sociology 98:1320–1350.
Putnam, Robert. 1993. “The Prosperous Community: Social Capital and Community Life.” The American Prospect (35–42).
Raudenbush, Stephen and Robert J.Sampson. 1999b.“‘Ecometrics': Toward A Science of Assessing Ecological Settings, with Application to the Systematic Social Observation of Neighborhoods.” Sociological Methodology 29:1–41.
Robert, Stephanie. 1998. “Community-Level Socioeconomic Status Effects on Adult Health.” Journal of Health and Social Behavior 39:18–37.
—. 1999. “Socioeconomic Position and Health: The Independent Contribution of Community Socioeconomic Context.” Annual Review of Sociology 25:489–516.
Rosenbaum, Dennis P., Arthur J.Lurigio, and Robert C.Davis. 1998. The Prevention of Crime: Social and Situational Strategies. Belmont, CA: Wadsworth.
Sampson, Robert J. 1999a. “How Do Communities Undergird or Undermine Human Development? Relevant Contexts and Social Mechanisms.” In Does It Take a Village? Community Effects on Children, Adolescents, and Families, edited by Alan Booth and Nan Crouter. Mahwah, New Jersey: Lawrence Erlbaum. In press.
—. 1999b. “Crime and Public Safety: Insights from Community-Level Perspectives on Social Capital.” Paper prepared for the Ford Foundation conference on “Social Capital and Poor Communities: Building and Using Social Capital to Combat Poverty,” New York City, March 11–13th, 1999.
—. 1999c. “Collective Properties and Healthy Communities.” Draft chapter for the National Academy of Sciences, Panel on “Future Directions of Social and Behavioral Sciences Research at the National Institutes of Health.” Burton Singer, Chair. Washington, D.C.: National Academy of Sciences.
Sampson, Robert J., Stephen Raudenbush, and Felton Earls. 1997. “Neighborhoods and Violent Crime: A Multilevel Study of Collective Efficacy.” Science 277:918–924.
Sampson Robert J.Jeffrey Morenoff, and Felton Earls. 1999. “Beyond Social Capital: Spatial Dynamics of Collective Efficacy for Children.” American Sociological Review 64:633–660.
Sandefur, Rebecca and Edward O.Laumann. 1998. “A Paradigm for Social Capital.” Rationality and Society 10:481–501.
Schwartz S. 1994. “The Fallacy of the Ecological Fallacy: The Potential Misuse of a Concept and the Consequences.” American Journal of Public Health 84(5):819–824.
Schwartz, S., E.Susser, and M.Susser. 1999. “A Future for Epidemiology?” Annual Review of Public Health 20:15–33.
Shaw, Clifford, and Henry McKay. 1942. (2nd edition, 1969). Juvenile Delinquency and Urban Areas. Chicago: University of Chicago Press.
Sherman, Lawrence. 1995. “The Police.” In Crime, edited by J.Q.Wilson and J.Petersilia. San Francisco, CA: ICS Press.
Sherman, Lawrence, Patrick Gartin, and M.Buerger. 1989. “Hot Spots of Predatory Crime: Routine Activities and the Criminology of Place.” Criminology 27:27–56.
Skogan, Wesley and Susan Hartnett. 1997. Community Policing, Chicago Style. New York: Oxford University Press.
Shy, C.M. 1997. “The Failure of Academic Epidemiology: Witness for the Prosecution. ” American Journal of Epidemiology. 145:479–487.
Sooman, Anne and Sally Macintyre. 1995. “Health Perceptions of the Local Environment in Socially Contrasting Neighbourhoods in Glasgow.'” Health and Place 1(1):15–26.
Susser, Mervyn. 1994a. “The Logic in Ecological: I. The Logic of Analysis.” American Journal of Public Health 84:825–829.
Susser, Mervyn. 1994b. “The Logic in Ecological: II. The Logic of Design.” American Journal of Public Health 84:830–835.
Wallace, Rodrick and Deborah Wallace. 1990. “Origins of Public Health Collapse in New York City: The Dynamics of Planned Shrinkage, Contagious Urban Decay and Social Disintegration. ” Bulletin of the New York Academy of Medicine 66(5):391–434
Wallerstein, Nina. 1999. “Power between Evaluator and Community: Research Relationships within New Mexico's Healthier Communities.” Social Science and Medicine 49(1):39–53.
Wilson, William Julius. 1987. The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. Chicago: University of Chicago Press.
—. 1996. When Work Disappears. New York: Knopf.
Woolcock, Michael. 1998. “Social Capital and Economic Development: Toward a Theoretical. Synthesis and Policy Framework.” Theory and Society 27:151–208.
Yen, Irene H. and S.L.Syme. 1999. “The Social Environment and Health: A Discussion of the Epidemiologic Literature.” Annual Review of Public Health 20:287–308.
Yen, Irene H. and George Kaplan. 1999a. “Poverty Area Residence and Changes in Depression and Perceived Health Status: Evidence from the Alameda County Study.” International Journal of Epidemiology 28:90–94.
—. 1999b. “Neighborhood Social Environment and Risk of Death: Multilevel Evidence from the Alameda County Study.” International Journal of Epidemiology 28:90–94.