Degree Name

EdD (Doctor of Education)


Educational Leadership

Date of Award

August 1981


The problem of this study was to determine if differences existed in the supervisors' perceptions of the importance of specified supervisory competencies. A list of thirty-six competencies which had been developed and validated by Ben M. Harris was adopted. Competencies were defined as any combination of knowledge and skill that is adequate for accomplishing some specified outcome. Included in the study were supervisors at the state department level in nine Southeastern states which were as follows: Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, South Carolina, Tennessee, and Virginia. Forty supervisors at the state department level were randomly selected from each state. Thirty-six null hypotheses were formulated to be tested at the .05 level of significance. Each hypothesis concerned a specific competency. Competencies were grouped according to task area. The analysis of variance was used as the first step in data analysis. This yielded an F ratio which indicated whether or not a significant difference existed. If a significant difference was revealed a follow-up test was conducted to determine where specific differences lay. The Newman-Keuls procedure was used for this purpose. Significant differences were revealed in only eight of the thirty-six hypotheses tested which were concerned with the following competencies: (A-3) Developing and adapting curricula; (C-2) Recruiting and selecting personnel; (C-3) Assigning personnel; (F-3) Designing in-service training sessions; (F-4) Conducting in-service training sessions; (G-1) Informing the public; (H-1) Developing educational specifications; (I-4) Analyzing and interpreting data. Thus, the null hypothesis was rejected for hypotheses 3, 8, 9, 19, 20, 27, 30, and 36. Major conclusions indicated that generally supervisors from the nine states did not differ significantly. This was not consistent with the diversity of roles and perceptions of supervisors as proclaimed by the literature. Even when significant differences existed specific differences between states were minimal. The F probability in seventeen competencies exceeded the 0.2500 level which indicated little difference and possibly some correlation existed. Recommendations included future research in supervision, clarification of supervisory roles and job descriptions, and implications for universities with graduate programs in supervision.

Document Type

Dissertation - unrestricted