SAVRY Predictive Validity of Mississippi Justice-Involved Youth Recidivism: A Latent Variable Approach

Abstract

Recidivism, and its contributing factors, remains a primary concern among juvenile justice practitioners. The literature has identified numerous policies and practices that reduce reoffending. Among these is use of validated assessment tools to identify risk level and treatment needs of youth. Using six years of juvenile court processing data, the authors examine the predictive validity of the Structured Assessment of Violence Risk in Youth (SAVRY), a structured professional judgement tool used to predict risk of violence and recidivism among youth aged 12 to 18. Findings support the use of the SAVRY to predict general recidivism, especially among males and highlights the importance of conducting gender-specific analyses. Analyses did not identify patterned race/ethnicity differences in risk. Our findings recommend the continued use of the SAVRY as an evidence-based instrument.