Predicting Criminal Recidivism in Adult Male Offenders: Researcher Versus Parole Officer Assessment of Dynamic Risk

In an attempt to bridge the gap between research and practice in the domain of criminal risk assessment, this study compared the predictive accuracy of dynamic risk assessments attained via an exhaustive research protocol to that achieved by the proxy ratings generated by parole officers. After an i...

Full description

Saved in:
Bibliographic Details
Published in:Criminal justice and behavior 2010-08, Vol.37 (8), p.860-882
Main Authors: Jones, Natalie J., Brown, Shelley L., Zamble, Edward
Format: Article
Language:English
Subjects:
Accuracy
Adults
Comparative analysis
Criminal justice
Men
Recidivism
Regression analysis
Risk assessment
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In an attempt to bridge the gap between research and practice in the domain of criminal risk assessment, this study compared the predictive accuracy of dynamic risk assessments attained via an exhaustive research protocol to that achieved by the proxy ratings generated by parole officers. After an initial prerelease assessment, 127 male offenders under community supervision in Ontario, Canada, were assessed by parole officers and researchers at three different intervals (i.e., 1, 3, and 6 months postrelease). Cox regression survival analyses with time-dependent covariates and receiver operating characteristic analyses revealed moderate to high levels of predictive accuracy in both research-based and parole officer ratings (area under the curve [AUC] = .79 and .76, respectively). The strongest prediction model combined the research-based time-dependent dynamic factors with static items (AUC = .86), thus offering provisional support for the inclusion of prospectively rated dynamic factors in risk assessment protocols.
ISSN:0093-8548
1552-3594
DOI:10.1177/0093854810368924