Collective Efficacy and Crime

In this article, we explore the literature on collective efficacy and crime, but pay attention to three major challenges that confront this literature, and researchers moving forward with the concept in studies of neighborhoods. These challenges are (1) precisely defining and measuring the concept of collective efficacy; (2) determining whether the notions of general cohesion and trust in neighborhoods are really components of collective efficacy, or determinants of it; and (3) whether neighborhood crime can have feedback effects on the level of collective efficacy. Collective efficacy is a construct that originally comes out of the psychology literature and the writings of Bandura (1982, 1986). Although the psychology literature has explored the empirical effects of collective efficacy for various relatively small groups, arguably its largest impact on social science literature comes from the introduction of the concept to the neighborhoods and communities literature by Sampson et al. (1997). A voluminous literature that is largely sociological and criminological has explored the effect of collective efficacy on a number of community processes, including neighborhood crime rates. In this article, we explore the literature on collective efficacy and crime, but pay attention to three major challenges that confront this literature, and researchers moving forward with the concept in studies of neighborhoods. These challenges are (1) precisely defining and measuring the concept of collective efficacy; (2) determining whether the notions of general cohesion and trust in neighborhoods are really components of collective efficacy, or determinants of it; and (3) whether neighborhood crime can have feedback effects on the level of collective efficacy. A Brief History of the Concept of Collective Efficacy Collective efficacy was initially introduced as a concept by Bandura (1982, 1986). Building on his notion of self-efficacy, Bandura noted that “Perceived collective efficacy will influence what people choose to do as a group, how much effort they put into it, and their staying power when group efforts fail to produce results” (Bandura, 1982: p. 143). His early consideration of the idea mostly conceptualized collective efficacy as a property of very large groups, such as nations and social movements (Bandura, 1986: pp. 449–453). This spawned a number of studies in the psychology literature in the 1990s and into the 2000s focusing on collective efficacy as a property of small groups, such as sports teams and work organizations (Zaccaro et al., 1995). Collective efficacy is conceptualized in this literature as a fundamentally group phenomenon that “. fosters groups’ motivational commitment to their missions, resilience to adversity, and performance accomplishments .,” and an emergent phenomenon in which “the locus of perceived collective efficacy resides in the minds of group members” (Bandura, 2000: p. 76) The key twist in the trajectory of collective efficacy as a concept in the social sciences arguably came when Sampson and colleagues introduced the concept to the neighborhoods and communities literature. In what has since become a citation classic, their study in Science introduced the concept as a property of neighborhoods that might be able to reduce the level of violence (Sampson et al., 1997). This seminal study focused on neighborhoods in the city of Chicago, and proposed defining collective efficacy as the combination of a general sense of residents’ willingness to provide informal social control along with high levels of trust and solidarity among residents. They argued for the plausibility of this broader conceptualization of collective efficacy compared to that developed in the psychology literature because “the willingness of local residents to intervene for the common good depends in large part on conditions of mutual trust and solidarity among neighbors” (Sampson et al., 1997: p. 919). The cross-sectional study indeed found that neighborhoods with higher levels of collective efficacy exhibited lower levels of violence (whether measured as perceived violence, violent victimization, or homicide events). This single study has fostered an army of subsequent research. Sampson himself has further refined the concept of collective efficacy in subsequent work. In particular, he has argued that it is a concept that helps focus on social mechanisms, and moves away from a risk-factor approach (Sampson, 2006a,b). In this sense, he considers it a mediating variable between neighborhood structural characteristics and crime. He emphasizes that this concept is fundamentally about repeated interactions, which then impact expectations about the future (residents’ beliefs about the actions of their fellow residents) (Sampson, 2006a,b). Sampson’s work on collective efficacy and neighborhood crime can be seen in his recent book that chronicles the Project on Human Development in Chicago Neighborhoods (PHDCN) as a large study of neighborhoods in Chicago (Sampson, 2012). Collective Efficacy: Challenges of Measurement Measuring informal social control in neighborhoods is actually quite a challenging problem. The challenge is that informal social control activity is only on display in moments in which there are challenges to the social order. This potential response on the part of residents is typically hypothetical, or latent. International Encyclopedia of the Social & Behavioral Sciences, 2nd edition, Volume 4 http://dx.doi.org/10.1016/B978-0-08-097086-8.45045-2 169 Thus, attempting to measure the behavior of informal social control is difficult because of its rarity. This is both a methodological challenge (for measurement) as well as a theoretical challenge. More commonly, studies have relied on questions that are hypothetical about how the respondent would behave in a particular instance (Warner, 2007). In the case of studies measuring collective efficacy, the hypothetical questions ask respondents how they believe their neighbors would behave in various situations (Sampson et al., 1997). Answering such questions will sometimes be difficult if residents have never seen their neighbors be confronted by situations requiring informal social control behavior and require them to speculate on the answer. This speculation is particularly likely in neighborhoods with low levels of crime or disorder (St. Jean, 2007: p. 211). Besides these issues around uncertainty is the challenge of measuring a group concept by surveying individuals. This issue remains unresolved in the literature, as Zaccaro et al. (1995) note, “. a more useful approach to defining collective efficacy is to consider both judgments of members’ abilities and perceptions of how well group members work together in achieving collective outcomes” (p. 309). Other approaches have instead attempted to measure collective efficacy using behavioral measures. For example, one study suggested that parental monitoring may represent collective efficacy in action (Rankin and Quane, 2002). Another study used the presence of registered voters as a proxy for the collective efficacy in a streetblock (Weisburd et al., 2012). One unique and clever approach to measuring collective efficacy in neighborhoods constructed a website that allowed residents of neighborhoods to start webpages, and then assessed the usage of the pages as one measure of collective efficacy (Hampton, 2010). An interesting finding was that there were two extremes of neighborhoods that generated such webpages – new middle-class suburban neighborhoods and truly disadvantaged neighborhoods – suggesting that such a measure may be tapping some different processes than the more conventional measures of collective efficacy. Distinguishing between Cohesion and Informal Social Control: One Construct, or Causally Related? Although Sampson and colleagues defined collective efficacy in neighborhoods as being a combination of both cohesion/trust and expectations of informal social control, there is debate about this strategy. It is worth noting that the psychology literature has typically conceived of cohesion and collective efficacy as two separate constructs, which may separately impact one another in a causal fashion (Zaccaro et al., 1995). Empirical evidence in the communities and crime literature suggests that these may indeed be distinct constructs in many locations. Horne (2004) notes that the two measures are not always highly correlated and that high levels of cohesion do not necessarily increase norm enforcement. A study in Chicago suggested that informal social control attitudes mediated the relationship between neighborhood cohesion and crime (RhinebergerDunn and Carlson, 2011). A study of Tianjin, China found a correlation of just 0.43 between neighborhood cohesion and informal control, suggesting that they are in fact distinct constructs (Zhang et al., 2002). A study that split cohesion and expectations of informal social control found that while cohesion had a negative effect on perceived incivilities in the neighborhood, expectations of informal social control had no impact (Reisig and Cancino, 2004). A study of Mesa, AZ found minimal factor analysis evidence for combining these two constructs, and also found that only cohesion had a negative effect on violent crime when both were included in the model simultaneously (Armstrong et al., 2010). A study of Brisbane neighborhoods using a confirmatory factor analysis strategy also found no evidence for combining these two constructs; furthermore, this study used measures capturing collective efficacy for three distinct tasks, and all of them were distinct from a measure of cohesion (Wickes et al., 2013). Another study measured neighborhood attachment rather than cohesion, and found that higher levels of attachment led to higher levels of expectations of informal social control (Burchfield, 2009; Silver and Miller, 2004). Akeyassumptionof theneighborhoodcollective efficacy literature is that these norms and attitudes within a neighborhood that lead residents to perceive higher levels of collective efficacy then translate into actual higher levels of informal social control behavior. Thus, “Social movements and collective efficacy theory share a common orienting framework – a focus on themobilization of action for an intended purpose” (Sampson et al., 2006: p. 4). As Steenbeek andHipp (2011: p. 839) noted: “expectations that others will intervene (potential social control) need not necessarily result in people actually intervening more (actual social control behavior), even though this is implicitly assumed by social disorganization theory.” However, only a few studies have addressed this question empirically, and the evidence so far appears somewhat weak. For example, one study in the Netherlands did not have a robust measure of collective efficacy, but diduse longitudinal data to test andfind that thepotential for social control (measured by a single question asking about residents’ willingness to improve the neighborhood) in fact did not increase social control behavior two years later (Steenbeek and Hipp, 2011). A study of neighborhoods in Brisbane focused on residents who reported a high level of disorder in the neighborhood, and found that higher levels of reported collective efficacy did not in fact lead to more actual informal social control behavior (Wickes et al., 2012). Instead, this study found that higher levels of neighboring increased the likelihood of engaging in informal social control behavior. As another example, a study using an experimental design concluded that norms are more likely to be enforced when there are benefits that increase social interaction, which suggests that cohesion might be more important for fostering action than residents’ perceptions of informal social control potential (Horne, 2004). Studies Assessing the Consequences of Collective Efficacy in Neighborhoods A number of studies have assessed the cross-sectional relationship between neighborhood collective efficacy and crime, with studies almost always finding a negative relationship. The seminal study of Chicago found such an effect (Sampson et al., 1997), and subsequent studies in Chicago have also detected this negative relationship (Sampson and Raudenbush, 1999; Morenoff et al., 2001; 170 Collective Efficacy and Crime