A Comparison of the Violence Risk Appraisal Guide, Psychopathy Checklist, and child and Adolescent Taxon Scale: Predictive Utility And Cross Cultural Generalizable

Abstract

The Violence Risk Appraisal Guide (VRAG) is a widely utilized measure for estimating the risk of violent reoffending among forensic populations. However, completing the VRAG can be a lengthy process as it requires entering scores from a second test, the Psychopathy Checklist Revised (PCL-R), which must be administered separately and requires hours to complete. In order to reduce scoring time, the authors of the VRAG have developed a brief checklist, the Child and Adolescent Taxon Scale (CATS), which can be used in place of the PCL-R as a more efficient method of assessing psychopathy (Quinsey et al., 2004). Previous research has shown the CATS can identify antisocial individuals and yields similar VRAG risk estimates when substituted for the PCL-R (Glover et al., 2002; Quinsey et al., 1998). However, these investigations employed predominantly Caucasian samples, and evidence supporting the validity of the CATS with ethnically diverse populations is presently lacking. This dissertation research addressed these concerns by examining the predictive utility and cross-cultural generalizability of VRAG scores calculated using the CATS with a more racially diverse sample of forensic psychiatric patients. In addition, the utility of the CATS as a stand-alone measure of psychopathy was examined. The relationship between CATS, VRAG, and PCL-R scores was assessed, and the instruments were compared in terms of their ability to predict the length of time African American and Caucasian patients were treated in a maximum security hospital before being approved for a transfer to a less restrictive setting. As expected VRAG probability estimates for recidivism did not differ depending on whether the CATS or the PCL-R was used as the index of psychopathy. In addition, the CATS showed good concurrent validity with the PCL-R, and no significant race related scoring differences were observed. Finally, the CATS was the only risk assessment measure able to predict the length of time before participants were approved for transfer to a less restrictive setting. Findings are discussed in terms of the implications for clinical-forsensic practice.