Boost and Flag: Disentangling Repeat/Near Repeat Crime Patterns from Risk Heterogeneity

• In Event: New Directions in the Analysis of Repeat/Near Repeat Crime

Abstract

Researchers have documented that spatiotemporal patterning of crime is due, in part, to a contagion effect whereby some crimes serve as instigators that increase the likelihood of subsequent crimes (boost hypothesis). A limitation of prior research centers on distinguishing whether this contagion effect is causal or results from unobserved heterogeneity in relative risk (flag hypothesis). A recent study partially addressed this concern and provided evidence that levels and dimensions of structural disadvantage and disorganization in the community at-large condition repeat/near repeat crime patterns. An emerging literature has demonstrated the utility of Risk Terrain Modeling (RTM), a spatial econometric methodology, to developing reliable measures of relative risk based on the built environment that can accurately forecast crime for micro areas. Recently, RTM was utilized to develop an aggregate neighborhood risk of crime (ANROC) measure that reflects heterogeneity in risk across neighborhoods. Drawing on multiple spatiotemporal analytical methods and exploring the influence of community disadvantage and disorganization, the current study jointly assesses the boost and flag hypotheses. Specifically, we extend prior research distinguishing contagion from relative risk by exploring the conditional nature of near repeat crime patterns utilizing both social and demographic characteristics of the community and an ANROC measure.

Authors

• Shaun A Thomas, University of Arkansas

• Grant Drawve, University of Arkansas

• Matthew Nobles, University of Central Florida