Predicting Recidivism

Risk assessment devices are being increasingly used throughout the country. Most research has evaluated the use of such instruments with adults. The purpose of this study was to evaluate the validity of the Wisconsin Juvenile Probation and Aftercare Risk instrument that is being promoted as a “model system” for other states to use. This study used discriminant analysis to examine the extent to which the Wisconsin instrument could discriminate between recidivists and nonrecidivists. The results found that only one of the eight variables was significantly related to recidivism. The instruments' total risk score could not discriminate between the recidivist and nonrecidivist groups, bringing into question its validity. The implications of these results are discussed in terms of aftercare management of juvenile offenders.