Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Frank M. Gresham


The similarities and differences of the response classes of depression and anxiety were investigated with a sample of pre-adolescents and adolescents. The subjects were 412 pre-adolescents and adolescents from a local public school and a local psychiatric hospital. All of the participants were administered a battery of self-rating on two consecutive days. Two self-ratings of depression and two anxiety measures were utilized in order to adequately sample the content domains. Additionally, self-ratings were used to assess nine related response classes (i.e, fears, social skills, major life events, activity, hopelessness, depressed mood, anxious mood, parent-adolescent conflict, and dysfunctional thoughts). All assessment measures were evaluated in terms of their psychometric properties e.g., internal consistency, homogeneity, test-retest reliabilities, and criterion-related validity. The findings from these analyses indicated that all but one of the measures exhibited high internal consistency and homogeneity. Moreover, those measures for which test-retest reliability was assessed exhibited acceptable temporal stability. Regarding the criterion-related validity, as expected, the total scores from the depression and anxiety measures were moderately correlated. Additionally, only in a few instances did the demographic variables correlate significantly with one or more of the independent or dependent variables. Regarding the first primary analysis, the findings from the revised 2 (gender) x 2 (Trait; depression and anxiety present, depression and anxiety absent) extreme groups MANOVA indicated that the two traits could be significantly differentiated in the predicted direction, however, using this approach two Traits (anxiety present, depression present) were excluded due to zero cell sizes. The second primary analysis, cluster analysis, indicated that eight clusters could be reliably replicated across samples. These cluster(s) appear to represent each of the following response classes: depression and anxiety present, depression present, anxiety present, depression and anxiety absent. In summary, based on the findings from this study it appears that depression and anxiety can be significantly differentiated when a multivariate approach such as cluster analysis is used in an appropriate manner.