ABSTRACT
In forensic practice, clinicians often need to apply risk assessment results to complex correctional decisions, but a strong empirical base to support some practices (such as assessing strengths or measuring change) is limited and still emerging. Using data from 195 adult men (M age = 32.02, SD = 8.53) who completed a custody-based violence prevention rehabilitation program, we supplemented existing static and dynamic risk factor ratings recorded prospectively by correctional staff by conducting retrospective file reviews and coding additional dynamic risk factors and strengths (including motivation and program engagement). We observed small average reductions in dynamic risk ratings and increases in strength ratings. After controlling for static risk scores, increases in client motivation and program engagement predicted reduced violent recidivism. Results from multivariate models suggested clients’ interest in engaging with post-custodial support for staying crime-free was critical for post-release success. Although our methodology was innovative in this context, inconsistent interrater reliability (ICC1 = .05–.74) and results suggest clinicians should cautiously interpret change scores and strength factor ratings. Rehabilitation research would benefit from partnerships with corrections agencies to design rigorous prospective research on client motivation and program engagement.
Acknowledgements
The views expressed are those of the authors and not necessarily those of the Department of Justice and Community Safety (Corrections Victoria). We thank Corrections Victoria and Victoria Police staff for their partnership in providing the data used in this article. We acknowledge Daniel Shea, Helen Casey, and David Ballek for their assistance with this research.
Data availability statement
Corrections Victoria owns the data described in this article; we used these data with their permission and can share the data only with written permission from Corrections Victoria.
Disclosure statement
No potential conflict of interest was reported by the author(s).