| Summary: | Thesis (M.Sc.)--Memorial University of Newfoundland, 1992. Psychology Bibliography: leaves 102-110. -- Error in pagination: page ix mislabeled as iv. The construct validity and the ability of the Offer Self-Image Questionnaire for Adolescents (OSIQ, Offer, Ostrov & Howard, 1982) to differentiate between clinical and normal groups were examined. Subjects were 197 students recruited from schools and 30 clinical subjects recruited from an outpatient adolescent counselling service in St. John’s, Newfoundland. A factor analysis was conducted to assess the validity of the scale structure of the OSIQ. This analysis revealed that the 12 scales which are purported to make up the OSIQ could not be obtained from a factor analysis of the correlation matrix of items. Further analyses revealed that the secondary factors, originally reported by Offer (1969) were only partially replicated with the present sample. Specifically, these factors were closely matched in the analysis of the female subjects, but were not found in analyses of male subjects. Nor were these factors found in analyses of the clinical or school subjects. The ability of the OSIQ to predict clinical status of the subjects was also investigated. The Emotional Tone, F (1, 169) = 14.25, p < .001; Family Relations, F (1, 169) = 12.22, p = .001; and Psychopathology, F (1, 169) = 11.48, p = .001, scales were able to distinguish school subjects who sought counselling in the past year ("school/clinical”) from those who did not ("normal"). As well, the Family Relations, F (1, 179) = 4.63, p < .05, and Idealism, F 1, 179) = 6.30, p < .05, scales were found to differentiate the "clinical" group and the "normal" group. However, the Idealism scale showed that those who had received counselling had higher self-image scores. No sex differences were found in any of the analyses. The results of the present study indicate that more work should focus on validation of the internal structure of the OSIQ, and that further examination of its abilities to differentiate between clinical and "normal" subjects and between males and females is necessary.
|
|---|