A Validation Study of the Positive Achievement Change Tool (PACT)
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The Juvenile Justice and Delinquency Prevention Act (JJDPA, 2002) mandates the use of risk assessments in juvenile justice proceedings. The Texas Juvenile Justice Department (TJJD, 2015) has approved the use of the Positive Achievement Change Tool (PACT), a relatively recently developed juvenile risk assessment instrument. Although promising, much of the published validation studies of the PACT have been limited to the population in which it was developed (Baglivio, 2009; Baglivio & Jackowski, 2013; Winokur-Early, Hand, & Blankenship, 2012). The current study examined the PACT’s predictive validity for recidivism in a sample of Texas juvenile offenders (n = 549). Additional analyses assessed for any gender/ racial differences in the PACT risk scores’ predictive abilities as well as explored the predictive validity of these scores for continued probation involvement and specific treatment outcomes. Initial PACT assessment data were collected on juveniles aged 10 to 18. Outcomes on re-offending, probation involvement, and treatment referrals within 12-months of the initial PACT administration were also obtained. The small sample of recidivists precluded some gender analyses, such that, of the 77 juveniles who recidivated, only 10 were female. Findings suggest that the Overall Risk to Reoffend level (AUC = .621), Criminal History (AUC = .612), and Social History (AUC = .603) scores demonstrated similar predictive abilities for the total sample. Observed AUC effect sizes for the male-only, minority-only, and non-minority samples were also in the small to moderate range. Multiple regression analyses also consistently revealed High Risk juveniles were between three to five times more likely to recidivate than Low Risk juveniles. The Overall Risk to Reoffend level also predicted and differentiated between continued probation involvement, the number of different intervention referrals received, and total intervention referrals. The risk levels did not predict total intervention success. The Criminal History and Social History domain scores were also variable predictors of probation and treatment outcomes. Overall, the results of this study provide continued support for the use of the PACT as a valid measure of predicting risk for recidivism with Texas juvenile offenders. Implications for use and implementation of risk/needs assessment and future research are discussed.