Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)




This study examined the relationship between outcome in therapy and the variables of method of group therapy employed in interaction with a patient's personality. Four therapy assignments were theoretically and empirically matched with particular groups of patients: Social-Emotional groups for mixed reaction patients, Expressive-Interpretive groups for non-psychotic patients, Emotional-Relationship groups for withdrawn patients, and Social-Management groups for aggressive patients. Subjects in the study were also classified according to time involved in therapy. These categories are: Rejectors, Retreaters, Withdrawers, Walkouts, Self-Terminators, and Finishers. Test data were pre-therapy scores on the Interpersonal Check List (ICL) and Thematic Apperception Test (TAT) for all subjects and post-therapy scores for those who completed therapy (Finishers). Using 147 subjects previously assigned to profiles by a linear typal analysis, analysis of variance examining sex as a contributing factor to correct or incorrect assignment yielded no significant differences between males and females. Pre-therapy scores for 185 subjects were subjected to a cluster analysis using the Overall and Klett procedure of linear typal analysis. This analysis yielded pure type profiles in which the subjects clustered according to combinations of the variables examined. Interpretations of the variable combinations in the profiles revealed personality constellations consistent with personality dynamics of the four original therapy assignments. To determine effectiveness of these therapy methods post-therapy scores of the 58 subjects who "finished" therapy were used to determine change scores on the ICL and TAT. A criterion for improvement in a therapy method on the ICL and TAT was developed and subjected to multivariate and univariate analyses of variance to determine effectiveness of correct assignment to a therapy method. No significant differences were found between patients correctly and incorrectly assigned to a particular method. Various comparisons revealed no significant differences between patients correctly assigned to a method and patients with characteristics common to other therapy assigned patients also assigned to that same method. Investigation of the ICL self dimensions as a means of examining change and determining improvement yielded no conclusive findings.