Title

A LONGITUDINAL ANALYSIS OF THE PERSONALITY AND BEHAVIORAL EFFECTS OF PARTICIPATION IN THE CONNECTICUT WILDERNESS SCHOOL: A PROGRAM FOR DELINQUENT AND PRE-DELINQUENT YOUTH

Date of Completion

January 1981

Keywords

Psychology, Clinical

Degree

Ph.D.

Abstract

Six months after completing the Connecticut Wilderness School program for delinquent and pre-delinquent youth, 104 graduates were contacted to assess personality and behavioral changes. A multi-source design was used, collecting data from agency, parent, and Wilderness School staff, as well as from subject self-report questionnaires. These data were combined with data collected before and after the nineteen day course (Gaston 1978). Data were collected on 92% of the subject sample. A post-hoc only control sample of 50 subjects was obtained at the time of follow-up.^ Self-Report measures included the Nowicki-Strickland Internal/External Locus of Control(I/E), a Self-Rating Scale assessing self-confidence and self-image, and a Deviant Behavior Scale assessing frequency of deviant behavior in school, home, and community settings. A structured interview and a measure assessing the amount of change in different areas were also administered.^ Agency data included the Peterson Quay Behavior Problem Checklist and a questionnaire including demographic data and behavioral ratings. Parents completed a questionnaire regarding changes in personality, attitudes, and behavior, and Wilderness School staff maintained records regarding the amount and type of follow-up contact.^ Multiple regression analyses indicated that treated youngsters were more internal (locus of control), and more positive in their self-rating than control subjects. Treated youngsters also exhibited significantly less deviant behavior and fewer arrests than controls. Age was related to internality increases, and a significant sex effect was found on the deviant behavior scale, indicating females to be less deviant across groups than males.^ Agency Ratings on the Behavior Problem Checklist indicated less problem behavior for the treated group overall, with significant between-group differences on total score, and Personality Problem, Inadequacy/Immaturity, and Conduct Problem subscales. No differences were found on the Socialized Delinquency subscale.^ Longitudinal Analyses revealed no regression in internality scores from post-course to follow-up, and low correlations between I/E and behavioral measures. Some correlation of behavior and self-rating was found. Repeated measures analyses of variance resulted in significant trials effects for I/E, and for the Behavior Problem Checklist, including all four subscales. A sex main effect was found for BPC total score, and for the Conduct Problem subscale.^ Correlational analyses of the agency, parent and self-report data reveal strong agreement on the overall effects of the program, with less agreement on individual areas of change.^ The amount of parent involvement in follow-up services was significantly related to increases in subjects' internality.^ The results indicate that this program has a measurable effect on reducing delinquent behavior in the treated group. It appears to be most successful for females, and for youngsters in the early stages of delinquency. ^

COinS