Targeted Interventions Gaining Enhanced Reentry (TIGER): Development of a Gender-Specific Multiband Actuarial Risk Assessment

• In Event: Gendered Aspects of Reentry Experiences

Abstract

In partnership with the Louisiana Department of Public Safety & Corrections (LA DPS&C), the Mississippi State University National Strategic Planning & Analysis Research Center (NSPARC) sought to refine and expand the Targeted Interventions Gaining Enhanced Reentry (TIGER) Risk-Needs-Responsivity System. Developed through the 2014 Justice Reinvestment Initiative administered by the U.S. Bureau of Justice Assistance (BJA), the TIGER RNR System utilizes two core components to inform offender treatment pathways: 1) a series of narrowband, population-specific actuarial risk assessments examining static risk factors; and 2) a psychometrically validated criminogenic needs assessment examining dynamic risk factors. Though highly predictive, the TIGER Risk Assessment suite solely assesses probability of general three-year recidivism due to the nature of consisting of narrowband assessments, rather than providing more nuanced predictions pertaining to offense-specific recidivism and technical violations.

The current study aims to address this limitation through the development of a series of gender-specific multiband actuarial risk assessments using period-specific populations of individuals under the purview of the LA DPS&C. To this end, supervised machine learning protocols were used to assess the propensity of: a) three-year general recidivism; b) three-year violent recidivism; c) three-year property recidivism; d) three-year drug recidivism; and e) three-year technical revocation.

Authors

• Timothy Thomas Reling, Mississippi State University - NSPARC

• Yanbing Tang, Mississippi State University - NSPARC

• Long Tian, Mississippi State University - NSPARC

• Mounica Irrinki, Mississippi State University - NSPARC

• Robert Vehock, Louisiana Department of Public Safety & Corrections