Assessing the Implications of a Structured Decision‐Making Tool for Recidivism in a Statewide Analysis

Abstract

Research Summary The Florida Department of Juvenile Justice has implemented a disposition matrix to guide recommendations made by juvenile probation officers to the court. This study examines whether recidivism rates for dispositions/placements made within the suggested range of this matrix differ from those outside of the suggested range. Using a sample of 38,117 juvenile offenders, we found that the dispositions/placements within the suggested range had an average recidivism rate of 19.4%, whereas those whose dispositions were outside the range had an average recidivism rate twice as high (38.7%). Furthermore, dispositions/placements that were the least restrictive option within the suggested range performed best. Dispositions above the suggested range (more restrictive) performed poorly, although those below the suggested range (less restrictive than suggested) performed the worst. These results held for males and females, across race/ethnicity, and across risk to reoffend levels. Policy Implications Implementation of structured decision‐making tools leads to questions from stakeholders and front‐line staff charged with using those tools regarding their effectiveness. Research and theory‐based justifications do not hold the weight actual data from the implementation population provide. These tools help control costs, facilitate planning, and can improve outcomes. Monthly monitoring of adherence rates, development of override and management oversight protocols, and regular feedback to front‐line staff are critical components of success.

When Does Race Matter in Juvenile Court Outcomes? A Test of the “Type of Offense” Hypothesis

“…Baglivio et al. () show that the use of a disposition matrix for sentencing (that included the use of a risk assessment) produced more favorable outcomes for youth at all risk levels than sanctions outside of the matrix. However, the problem with this approach is that sentencing guidelines may restrict judges from considering factors that they believe are very relevant to a specific case.…”

Does Treatment Quality Matter?

“…The Comprehensive Strategy for Serious, Violent, and Chronic Juvenile Offenders and the notion of progressively graduated sanctions (Howell, ; Wilson and Howell, ), encompassed by the “risk principle” of matching the intensity of services to the risk to reoffend level of the youth, provides a core model of how cases should be handled within a juvenile justice system. Reserving the deepest end placements for the highest risk youth who have been unsuccessful in less intensive placements is not only fiscally responsible, by targeting limited resources to those most in need of services and who pose the highest risk to society, but has been demonstrated to optimize overall system effectiveness (Andrews and Bonta, ; Andrews and Dowden, ; Andrews, Zinger, Hoge, Bonta, Gendreau, and Cullen, ; Baglivio, Greenwald, and Russell, ; Lipsey, ; Lowenkamp and Latessa, ). As poorly implemented interventions lead to undermining the entire evidence‐based approach, a central concern in this framework is the notion of whether interventions are implemented effectively.…”

“…In 2011, the FDJJ became involved in the Juvenile Justice System Improvement Project (JJSIP), a grant‐funded opportunity for juvenile justice system reform through the Georgetown University Center for Juvenile Justice Reform and Vanderbilt University's Peabody Research Institute . The project focused on two elements of reform: (1) ensuring a system of graduated sanctions from prevention through residential placement and step‐down aftercare services, “right‐sizing” resources across the continuum, as well as the design, implementation, and validation of a disposition recommendation matrix to guide probation officer recommendations to the juvenile court (Baglivio et al., ), and (2) the implementation of Vanderbilt University's Standardized Program Evaluation Protocol (SPEP™), a standardized measure of intervention effectiveness developed by Dr. Lipsey based on meta‐analysis of more than 500 high‐quality studies of juvenile offender interventions (Howell et al., ; Lipsey and Howell, ). Lipsey (), in his meta‐analytic work, indicated that the variation in the effectiveness of interventions, with respect to reducing recidivism, is mostly attributable to focusing on higher risk youth (the risk principle), the quality of the implementation, the dosage provided, and the type of service being delivered (e.g., cognitive–behavioral interventions are more effective than individual counseling in reducing subsequent offending, on average).…”

“…For example, one variable in the CPAI involves whether there is a validated risk assessment to ensure that only higher risk youth are placed in residential programs, after community‐based alternatives have been exhausted. In Florida, this has been a mandatory and standardized feature across the state since 2012, and this is done in accordance with a validated disposition matrix (Baglivio, Greenwald, et al., ). Also, the CPAI considers whether there is a regular reassessment of criminogenic risk and needs throughout placement—this too is mandatory across Florida residential programs, and its use has been empirically examined (Baglivio, Wolff, Jackowski, and Greenwald, ; Hay, Widdowson, Bates, Baglivio, Jackowski, and Greenwald, ).…”

Considering Race and Gender in the Validity of Juvenile Justice Risk

“…), while services within those restrictiveness levels (such as anger management, substance abuse treatment, etc.) should be based on the individual juvenile's top‐ranked criminogenic needs (Baglivio, Greenwald, and Russell, ). As such, it becomes critical to understand whether the predictive ability of a given risk assessment is equal across all subpopulations on which the tool is employed.…”