Juvenile Risk and Needs Assessment and Ethnicity: Beyond Predictive Validity

• In Event: Contemporary Questions in Juvenile Risk and Needs Assessment: Results from a Multi-Site Study

Abstract

The increased use of actuarial risk and needs assessments to predict recidivism and guide case management in the juvenile justice system has generated a need for ongoing research investigating the validity of these tools. Prior research has primarily focused on examining the validity of a given assessment in terms of predicting recidivism. Recent research has suggested that important differences may exist in the degree of predictive validity across racial and ethnic subgroups; however, findings in this body of research are mixed and the underlying mechanisms producing differences are not well understood. A possible explanation for the observed differences by race is that the items tapping into different risk and need domains may operate differently across racial subgroups. Validation studies often stop short of evaluating the construct validity of the measures captured in the tools (i.e., domain), and instead focus exclusively on predicting recidivism. This study compares the predictive and measurement validity of a juvenile risk assessment for youth in residential placement, focusing on male Caucasian and Hispanic subsamples (n~500). The implications of the findings for juvenile risk assessment and case management are discussed.

Authors

• Jamie Newsome, University of Cincinnati Corrections Institute

• James McCafferty, Kennesaw State University

• Amber Mandalari, University of Cincinnati

• Christopher Sullivan, University of Cincinnati