Mobility Network-based Measurement of Local Collective Efficacy and its Consequences for the Spatial Patterning of Violent Crime

Abstract Objectives Estimate the extent to which local variation in collective efficacy, a measure of social cohesion and norms toward intervention among individuals, is associated with the sub-neighborhood spatial patterning of violent crime in Columbus, OH. Methods Using estimates of local collective efficacy derived from survey data on individuals’ perceptions of collective efficacy in the neighborhoods and at their routine activity locations collected as part of the Adolescent Health and Development in Context Study and incident-level, point-referenced crime data from the Ohio Incident-Based Reporting System, we fit inhomogeneous Poisson process models. Results We find that net of neighborhood-level collective efficacy, a one standard deviation increase in deviation in the local collective efficacy score from the neighborhood average local collective efficacy score is associated with a decline in the violent crime intensity by a factor of 0.858. Conclusions Affirming Jane Jacobs’ arguments about informal social control dynamics at fine-grained levels, our study illustrates that collective efficacy operates and can be measured at sub-neighborhood levels across the entirety of a city. Our findings highlight the need to measure social processes at finer spatial scales in order to understand their effects on crime. Future research should prioritize collecting additional fine-grained data on social processes using individuals’ perceptions of the locations they frequent as part of their everyday mobility patterns.