Risk Metrics and the Five-Level System for the Youth Level of Service/Case Management Inventory (YLS/CMI): A Comparison of Two Samples of Justice-Involved Youth in Canada

Risk assessments are integral in the criminal justice system for predicting reoffending and guiding treatment. Challenges arise in using diverse risk tools due to difficulties comparing results across instruments with varying risk measurement methodologies. To address this, the Council of State Governments (CSG) Justice Center developed guidelines, applying metrics like percentiles, risk ratios, and recidivism rates to generate five risk levels (Hanson et al., 2017). While these levels have been applied to adult measures, their application to youth assessments remains unexplored. The Youth Level of Service/Case Management Inventory (YLS/CMI) is a widely used youth risk assessment tool, categorizing total scores into Low, Moderate, High, and Very High risk based on percentiles. However, percentiles convey a youth's position without directly linking to reoffense rates or communicating reoffending probability, requiring additional risk communication metrics. This dissertation aimed to provide risk metrics, separately by gender, for the YLS/CMI, with secondary aims to implement the CSG Justice Center’s Five-Level system and derive standardized risk levels. The final objective was to measure construct validity via psychopathy, aggression, and pride in delinquency measures. Using data from two justice-involved youth samples in Ontario, Canada – a probation sample (N = 540) and a forensic assessment sample (N = 880) – the study revealed that approximately half of the sample reoffended within two years, with similar rates in both samples. Percentiles highlighted YLS/CMI score distribution variations, emphasizing context dependence. Risk ratios were notable for Low Risk males in the community sample, indicating a 68% lower likelihood of reoffending than Moderate risk individuals. Recidivism rates varied across YLS/CMI risk categories and sample types, with significant differences in mean total scores observed across the Five-Level system in both samples. Construct validity in the forensic sample demonstrated significant differences in psychopathy, aggression, and pride in delinquency across levels. The discussion encompasses developmental factors, emphasizing the age-crime curve and addressing the distinctive needs of justice-involved youth. Recommendations include extending the validation of the five levels to youth risk tools, establishing local norms, and investigating the Five-Level system’s applicability across diverse settings.

Department dc:contributor.department

Applied Psychology and Human Development

Year dc:date.issued

2024

Author dc:creator

• Khan, Maria

Advisors dc:contributor.advisor

• Peterson Badali, Michele
• Skilling, Tracey

Handle dc:identifier.uri

http://hdl.handle.net/1807/140139

OAI identifier oai:identifier

oai:utoronto.scholaris.ca:1807/140139