Date Awarded


Document Type


Degree Name

Doctor of Education (Ed.D.)




P. Michael Politano


This study investigated the behavioral parameters of depression in children and adolescents. Demographic information, DSM-III psychiatric diagnosis, scores on the Children's Depression Inventory (CDI), and admitting problem (POR) behaviors were obtained from an archival data pool collected at four different child psychiatric hospitals on 630 inpatient subjects 5 to 20 years of age.;Subjects were assigned to one of four research groups defined by psychiatric diagnosis (DSM-III) and/or by scores on the CDI, i.e., CDI only groups (high vs. low scorers), DSM-III only groups (depressed vs. conduct-disordered), DSM-III + CDI groups (depressed high scorers vs. conduct-disordered low scorers), and DSM-III only (depressed children vs. depressed adolescents). Forty POR behaviors previously identified as being symptoms or associated features of childhood depression were used in a discriminant function analysis for each research question.;There was no difference between groups on POR behaviors when separated only by CDI scores. Results supported some behavioral difference between psychiatrically diagnosed depressed and conduct-disordered youth. Sadness appeared to be the most powerful discriminating variable for predicting depression as separate from conduct disorder when depression was defined by DSM-III criteria alone and in conjunction with CDI cutoff scores of 11 or higher. Aggression was significant in predicting conduct disorder for the DSM-III only groups. Poor self-concept and school underachievement were also indicated as behavior variables having discriminating power when depression was defined by DSM-III only. Age differences in POR behaviors for DSM-III depressed children and adolescents were not significant.;Results were discussed in terms of trends in the data that may be useful for further investigative efforts. The presence of a number of behaviors in DSM-III and DSM-III plus CDI depressed groups supported findings of previous psychometric studies; support was also found for several overlapping behaviors between depressed and conduct-disordered groups as reported in the literature. It was concluded that further behavioral study is needed with depressed children and adolescents to confirm findings from previous psychiatrically and psychometrically based studies. Findings would ultimately be useful in formulating developmental refinements in the assessment criteria for childhood depression and aiding in the differential diagnosis of depression and conduct disorder.



© The Author