Defense Date


Document Type


Degree Name

Doctor of Philosophy



First Advisor

Stanley R. Strong


This study tested the general hypothesis that a client's compliance or defiance of a therapeutic directive could be accurately predicted by manipulating the variables of incongruence and dependency within the client-counselor relationship. An attempt to control the relationship variables was made by using a no-choice, paradoxical directive to increase levels of relationship incongruence. The manipulation of the client's perception of their counselor's level of experience and expertness was aimed at controlling the dependency variables. The hypothesis that clients would report improvement of their symptom, following the delivery of a paradoxical directive, was also investigated.

The subjects of the study were 30 undergraduate students at Virginia Commonwealth University. All of the subjects reported to experiencing problems with Procrastination and wished to change this behavior. Subjects were randomly assigned to two treatment groups and a no-treatment control condition. In the treatment conditions, students received two interviews with counselors who were reported as being either expert or inexperienced. Each subject was given exactly the same paradoxical directive regardless of the experience level of the counselor. During the second interview, subjects were asked by their counselors and a confederate peer if they had completed the paradoxical assignment. Subjects responses were recorded and coded. All 30 subjects completed a weekly procrastination measure. In addition, treatment group subjects completed questionnaires regarding their perception of their counselors. All subjects completed inventories measuring their orientation to the change process. Subjects of all three conditions reported to significantly decreasing (p<.001) their levels of procrastination over time. There was no significant difference reported between the three groups.

Statistical analysis revealed that subjects' response types could not be accurately predicted at the p<.05 level. An analysis of the available data suggests that the subjects did not differentially perceive the counselors as expert or inexperienced, therefore, one of the experimental variables may not have been successfully manipulated. Further, the data indicates that the current primary hypotheses need to be revised and reevaluated.


Scanned, with permission from the author, from the original print version, which resides in University Archives.


© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission


Included in

Psychology Commons