The current study examined the predictive validity of the SAVRY in African American and White recently adjudicated juvenile offenders in Louisiana. The sample consists of 267 community-based, male juvenile offenders, whom were tracked for an average follow-up period of 18 months. Receiver operating characteristic (ROC) analyses on the overall sample and African Americans, specifically, showed the numeric score predicted recidivism. Chi-square analyses found the SRR did not have a significant relationship with reoffending for general recidivism petitions. However, it was significant for all other forms of recidivism in the overall sample and African Americans. Hierarchical Cox regressions identified significant differences in time to reoffenses (all forms and contexts) and SAVRY scores. The study concludes that the SAVRY shows promise as a cultural invariant risk assessment tool that can predict general and violent recidivism for African American juveniles, although additional research is required to more confidently support such a claim.
Woods, Scarlet Paria, "Predictive Validity of the SAVRY within a Diverse Population of Juvenile Offenders" (2013). Psychology Theses. 22.