Risk assessment of sex offenders

Abstract

This essay discusses the assessment of recidivism risk in sex offenders. It begins with definitions of critical terms and concepts. A number of approaches to risk assessment are described. Validated risk instruments are reviewed, with a focus on their reliability and accuracy in predicting recidivism. Actuarial assessment of risk is described as a two-stage process. In the first stage, offenders are assessed and assigned to a risk level or stratum. In the second stage, the probability of risk over a follow-up period is estimated based on the offender’s risk ranking. The essay discusses calibration in the context of Bayes’ theorem, which reveals critically important realities involving base rates and the use of currently available standardization samples in determining a final estimate of recidivism likelihood. The essay concludes with a glimpse into the future of risk assessment and predictions about the next stage in evidence-based risk assessment of sex offenders.