Predictors of depression in children with high-functioning autism spectrum disorders: The relationship between self-perceived social competence, intellectual ability, and depressive symptomology
MetadataShow full item record
The present study examined the relationships between cognitive ability, age, self-perceived social competency, and depression in a sample of children with high-functioning autism spectrum disorders (HFASDs) and a sample of typically-developing children. Previous research and tentative theory suggested that depressive symptoms show a high co-morbidity with HFASDs; that reported levels of depression in HFASDs may vary as a function of child vs. parent report; and that relationships among cognitive ability, age, self-perceived social competency, and depression variables would meaningfully differ across the two groups. Hypotheses, based on these assumptions, were tested using a sample of 30 children with HFASDs and 30 typically-developing children. Participants were matched by age, gender, and ethnicity across groups. Demographic differences between the groups were non-significant for age, gender, ethnicity and years of parent education, but significant for IQ. The between-group IQ difference was statistically controlled in all major analyses. BASC-2 parent ratings yielded significantly lower Social Skills scores and significantly higher Withdrawal scores for the HFASD group. Correlations among self-reported Interpersonal Relations and Social Stress scores from the BASC-2 with age and IQ, within each group, yielded only mixed support for the predicted relationships. Consistent with expectations, correlations between self-perceived social competency measures from the BASC-2 and self-reported depression scores from the BASC-2 and CDI yielded generally significant and consistent relationships across both groups. As predicted, a significant interaction was found between rater (parent vs. child) and group (HFASD vs. typically-developing) for perceived child depression level. Two explanatory hierarchical regression models were tested with one using BASC-2 SRP Interpersonal Relations as the criterion and the other using BASC-2 SRP Depression. The three-step hierarchical model for Interpersonal Relations accounted for 21.5% of the variance in the criterion and each predictor (i.e., diagnosis, IQ, and age) obtained statistical significance on its entry step. The four-step depression model accounted for 22.8% of the variance in the criterion, and each predictor (i.e., diagnosis, IQ, age, and BASC-2 SRP Interpersonal Relations), with the exception of age, obtained statistical significance on its entry step. Implications of all results are discussed and recommendations made for further research.