Identifying Common Patterns in Risk Assessments to Inform Condition Setting

• In Event: Condition-Setting Practices in Probation

Abstract

According to the RNR model, results from risk assessments should be used in condition setting. However, in practice, this often is not the case. In an effort to provide generic insight that can be used by courts, even prior to risk assessment, we provide descriptive analyses of the risk assessments. Based on ORAS reports from a statewide probation agency, encompassing over 25,000 cases, we will characterize common patterns by race, gender, and case characteristics, which provides a preliminary, baseline understanding of the risk and needs for individuals in the criminal legal system. For example, we find that women have higher risk for education, employment and financial situation; family & social support; substance use; and neighborhood problems, while men have higher risk for criminal history, criminal attitudes and behavioral patterns, and peer associations. Additionally, we find that younger individuals tend to have higher total and scale scores, suggesting that risk levels decline with age. We also find racial/ethnic disparities in risk scores, with Black and Latinx individuals generally scoring higher across multiple scales compared to White individuals. Superior Court cases tend to have higher scores in peer associations and attitudes/behavior, while DUI cases have lower ORAS scores overall. We discuss practice implications for these common patterns.

Authors

• JoAnn Lee, The University at Buffalo

• Feyza Koseoglu Darilmaz, Center for Advancing Correctional Excellence, George Mason University