Social Cohesion and Informal Social Control: Exploring the Factor Structure of Collective Efficacy Among Two Diverse Samples

Background: Collective efficacy is a widely studied phenomenon associated with multiple youth outcomes including exposure to violence, child maltreatment, substance use, mental health, and delinquency. Despite its widespread use in research, social work interventions struggle to raise collective efficacy. Commonly defined as a shared belief in neighbors’ conjoint capability for action to achieve its intended effect, collective efficacy is comprised by two concepts: social cohesion and informal social control.

Social cohesion is defined as trust among neighbors, shared values, and reciprocal helping behavior, whereas informal social control reflects the residents’ actions-to prevent delinquency in the local area. Many studies combine social cohesion and informal social control into a single construct of collective efficacy. However, recent research suggests that social cohesion and informal social control should be treated as separate concepts. This study contributes to the study of neighborhood processes by examining the factor structure of collective efficacy across two diverse samples.

Method: Data were drawn from two sources: (1) the Seattle Neighborhoods and Crime Survey (SNCS), and (2) the Fragile Families and Child Well‐Being study (FFCW). The SNCS was a random sample of Seattle residents designed to test multilevel theories of neighborhood social organization. The FFCW study is a longitudinal survey that follows an ethnically and racially diverse birth cohort of children and their families -through age 9; three-quarters of the families are not married.  In both data sets, collective efficacy was measured using items that are based on the original measure developed by Sampson and colleagues. Confirmatory factor analyses (CFA) were performed using MPlus 7.0 to examine the underlying factor structure of collective efficacy.

Results: The SNCS sample included 2,512 individuals averaging 46.5 (SD = 15.2) years old, 49.7% (n = 1,248) were female, most were (79.9%), White, (71.3%) currently employed, and nearly half (45.3%) were married. The FFCW includes 3127 mothers who participated in the 3^rd wave.  Participants had an average age of 28.0 years (SD = 6.07), 30.9% of mothers identified as White, 49.8% Black, and 16.8% Other. The majority of the mothers were not married (67%) and lived at 50% or greater of the Federal Poverty Level (76.6%).  CFA results indicated that collective efficacy is better represented as a two‐factor solution in both studies.  Model fit statistics for both samples demonstrated poor fit; SNCS sample were: χ²₁₉ = 180.795, RMSEA= 0.058, CFI = 0.991, TLI = 0.986, whereas fit statistics for the FFCW sample were χ²₃₅= 5555.92 (p<.000), RMSEA= .225, CFI= .8553, TLI= .810, WRMR=7.908.  Model fit for the FFCW improved significantly upon specifying social cohesion and informal social control as separate factors: Χ²₃₄= 1322.482, RMSEA= .110, CFI= .966, TLI= .954, WRMR=3.222.

Conclusions: Although traditionally conceptualized as concept single construct, findings from this study suggest that collective efficacy is better modeled as separate factors - rather than as a summary measure. Modeling collective efficacy as a two‐factor solution will allow researchers to examine their unique influence on outcome variables.   Doing so will allow researchers to uncover factors that can help turn social cohesion into collective action.