Predictors of recidivism following release from custody: a meta-analysis

Abstract

ABSTRACT The reliable identification of those offenders at greatest risk of post-release recidivism is critically important given the emotional and financial costs associated with offending behaviour. The aim of the current study was to synthesise the available literature on risk predictors to identify which factors are predictive of recidivism in adult offenders, in the four years following release from custody. After systematically reviewing the literature and selecting those at least risk of bias, 43 high quality studies were subjected to meta-analysis. Sufficient data pertaining to 21 factors were available. Consistent with Bonta and Andrews [(2017). The psychology of criminal conduct (6th ed.). Routledge], prominent factors associated with the ‘central eight’ risk domains for general recidivism, particularly those indicative of antisocial potential, produced the largest effect sizes. These included factors such as an extensive criminal history (e.g. number of previous incarcerations), rule violations whilst under supervision, and holding procriminal attitudes. Overall, static risk factors were superior to dynamic in predicting recidivism. We explore these findings in the context of the limitations of the risk predictor literature and argue that ongoing behavioural monitoring is a promising means of identifying real-time changes in the antisocial potential of prisoners released to the community .