Predictors of success at a rural juvenile offender facility

Abstract

Although risk factors contributing to failure in treatment of young offenders have been studied extensively, little is written about what effects success. This study on the latter takes advantage of data obtained at a local treatment facility. This study uses statistical strategies to compare 7 different variables from a set of archival data with the outcome variable, which is 'success in treatment'. The seven independent variables are ethnicity, age at entry to treatment, pre-release pass (PRP), days in treatment, FAS/FAE, sexual offender, and psychiatric diagnosis. This data has been accumulated by a clinician at the facility and offered to the investigator for the purpose of this project. The first stage of the analysis was to correlate all of the 7 variables with the outcome variable (success/no success). The variables with the strongest association were selected, and then correlated with each other. Variables shown to be correlated with success were further studied using a Logistic Regression analysis. The results of the statistical analysis showed that non-minority status was the only variable to be clearly associated with success.