NCJ Number
211343
Journal
Criminal Justice and Behavior Volume: 32 Issue: 5 Dated: October 2005 Pages: 479-510
Date Published
October 2005
Length
32 pages
Annotation
This study assessed and compared the validity of several contemporary violence risk assessment measures in a sample of criminal offenders recently released from incarceration.
Abstract
There are several violence risk assessment instruments that have been developed and tested to various degrees in recent years, with some success. Few studies have compared these various violence risk assessment instruments and procedures together in one sample. Specifically, no studies have compared the Violence Risk Appraisal Guide (VRAG), Psychopathy Checklist-Revised (PCL-R) and the current version of the Historical-Clinical-Risk Management-20 (HCR-20). In addition, the Violent Offender Risk Assessment Scale (VORAS) has not been compared to any of these other instruments. This research study sought to extend the developing body of research on risk assessment by addressing some of its limitations. The study evaluated the predictive validity of multiple measures of violence risk in a sample of 118 male offenders from Federal correctional institutions released to supervision in Western Canada between 1989 and 1994. These measures included the VRAG, HCR-20, VORAS, and two measures of psychopathy that often form parts of risk assessment batteries, the structured final risk judgments, the latter of which were recommended to be used in actual practice. Participants were selected for inclusion on the basis of known outcome status, violent recidivism versus no violent recidivism after release. The multivariate analyses showed that HCR-20 indices were consistently related to violence and that the VRAG entered some analyses. The findings are inconsistent with a position of strict actuarial superiority, as HCR-20 structured risk judgments, an index of structured professional or clinical judgment, were as strongly related to violence. Study limitations and implications are presented and discussed. Tables and references