The ability to predict recidivism was investigated with 130 male juvenile delinquents committed to a state training school. These youth had been arrested for felony offenses, and typically had long criminal careers. The current study was designed to improve the predictability of recidivism by determining those classes of variables that best delineate recidivists versus nonrecidivists. It was hypothesized that recidivists would exhibit more hostile and aggressive behavior, more antisocial characteristics and behavior, and have more involvement with illegal drugs. It was also hypothesized that nonrecidivists would show higher academic and intellectual functioning, more psychological distress, and better adjustment to the current placement. / Two groups of analyses were conducted. First, two-way MANOVAs were conducted on each a priori categorization of variables to detect differences between recidivists and nonrecidivists. Second, a factor analysis was conducted to empirically classify the variables. A stepwise discriminant function analysis was then conducted using the factor scores for each student to determine the equation that best determined group membership. / The results of the study indicate that it is possible to accurately predict recidivism in a population of chronic, severe, juvenile offenders using classes of behavior rather than individual variables. A past history of antisocial behavior was found to play the strongest role in the prediction of recidivism. The students' adjustment to the program, intellectual achievement, and psychological distress were found to play only minor roles in the prediction of recidivism, though their inclusion in the equation significantly improved the prediction rate. / Source: Dissertation Abstracts International, Volume: 53-10, Section: B, page: 5443. / Major Professor: Wallace A. Kennedy. / Thesis (Ph.D.)--The Florida State University, 1992.
| Identifer | oai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_76775 |
| Contributors | Duncan, Renae D., Florida State University |
| Source Sets | Florida State University |
| Language | English |
| Detected Language | English |
| Type | Text |
| Format | 79 p. |
| Rights | On campus use only. |
| Relation | Dissertation Abstracts International |
Page generated in 0.002 seconds