© 2025 CIRANO. All rights reserved.
William Arbour, assistant professor at Université de Montréal, and Guy Lacroix, full professor at Université Laval, both CIRANO researchers, published the article Beyond Traditional Risk Scores: Tackling LS/CMI Offender. Misclassifications with Machine Learning, co-authored with Sébastien Brouillette-Alarie, Guy Giguère and Steeve Marchand, in the Journal of Quantitative Criminology.
This paper investigates the accuracy of offender risk assessment scoring methods. The authors study the degree of misclassification resulting from the conventional practice of aggregating individual items to derive risk scores and categories. They document which types of offenders are prone to misclassification, particularly in relation to age and gender.