U.S. flag

An official website of the United States government, Department of Justice.

Sacrificing Accuracy for Transparency in Recidivism Risk Assessment: The Impact of Classification Method on Predictive Performance

NCJ Number
250851
Journal
Corrections Volume: 1 Issue: 3 Dated: May 2016 Pages: 155-176
Author(s)
Date Published
July 2016
Length
22 pages
Annotation

Using multiple performance metrics and measures of recidivism on 40,740 Minnesota offenders released from prison between 2006 and 2011, this study evaluated the performance of prediction models developed with both Burgess methodology and supervised learning algorithms (i.e., statistical and machine learning algorithms).

Abstract

Recent studies have compared the performance of machine learning algorithms versus logistic regression models in predicting recidivism. Existing research, however, has not compared their performance to Burgess methodologya transparent, simplistic and summative classification technique used to develop some of the most widely used risk and needs assessment instruments currently used in corrections. The results of the current study show that, compared to the best supervised learning classifiers, use of Burgess methodology yielded inferior performance in terms of predictive discrimination, accuracy, and calibration. (Publisher abstract modified)

Date Published: July 1, 2016