The Ohio Risk Assessment System 3.0: The Validation of Violence Predictor Tools

• In Event: Impacting Criminal Justice Outcomes: A Review of New Research on Supportive Practices

Abstract

Practitioners and researchers have advocated for an increased focus on the identification of risk of violent recidivism, citing the pain of both personal and societal costs associated with violent incidents. Prior research has uncovered that risk assessment tools can be developed to predict violent recidivism at an acceptable level. However, tools designed specifically for violence recidivism prediction are not yet widely used in corrections, highlighting the need for further efforts in this area. Recently, three violence predictor tools were developed as part of a widely used risk assessment system – the Ohio Risk Assessment System. These violence tools included a community supervision, a prison intake, and a reentry violence predictor. The goal of this study is to highlight the development and validation of the Ohio Risk Assessment System – Violence Predictor Tools using a sample of over 275,000 individuals under correctional supervision. Using a cross-validation technique, the results of the current study suggest the violence predictor tools reached acceptable levels of predictive validity. The research and practical implications of these findings for the prediction of violence recidivism will be discussed.

Authors

• Christopher D'Amato, University of Cincinnati Corrections Institute

• Myrinda Smith, University of Cincinnati Corrections Institute