Recidivism and Social Interactions

Abstract

Using a national sample, this article identifies the risk factors for recidivism among female, male, black, white, and Hispanic felony probationers. The individual hazard function is assumed to depend on individual and neighborhood characteristics, as well as on social interactions among probationers. In selecting the covariates from a set of potential candidates, Bayesian model averaging is used to account for both model uncertainty and the subsequent inference. The results point to social interactions as one of the most significant factors affecting recidivism among all gender, ethnicity, and race groups. When a frailty parameter is added to account for the possibility of unobserved risk factors shared by probationers within neighborhoods, the empirical results remain robust indicating negligible unobserved neighborhood-level heterogeneity.