Classifying Violent versus Non-Violent Offending in a Diverse Sample of Adolescent Juvenile Delinquents

Abstract

The purpose of the study was to examine whether a set of theoretically and empirically established risk factors could contribute to the classification of violent versus non-violent offending in an ethnically diverse sample of male and female adolescent juvenile delinquents. Variables examined included economic disadvantage, grade point average (GPA), sex, dependency involvement, special education diagnosis, specific learning disability diagnosis, emotional disability diagnosis, history of illegal substance violation, and recidivism. Observed versus expected frequencies of violent versus non-violent offending across different ethnicity categories were also examined. It was hypothesized that economic disadvantage and grade point average would not significantly contribute to the classification of violent versus non-violent offense group membership in juvenile delinquents. It was hypothesized that sex, dependency involvement, special education diagnosis, specific learning disability diagnosis, emotional disability diagnosis, history of illegal substance violation, and recidivism would significantly contribute to the classification of violent versus non-violent offense group membership in juvenile delinquents. It was also hypothesized that there would be no significant differences between observed versus expected frequency of violent versus non-violent offending across ethnicity categories.A discriminant analysis retained five of nine variables in the final stepwise model, including recidivism, illegal substance violation history, special education diagnosis, emotional disability diagnosis, and specific learning disability diagnosis. While variables in the analysis significantly contributed to the classification of non-violent group membership, the model yielded low hit ratio for classification of cases into the violent group. A chi-square analysis was also conducted to examine whether there were significant differences between observed and expected frequencies across different ethnicity categories with regard to violent versus non-violent offense group membership. No significant association existed between ethnicity and the juvenile delinquents' expected versus observed frequency of violent versus non-violent offender group membership. Implications of these findings, limitations of the study, and areas for future research are discussed.