ERIC Number: ED222154
Record Type: RIE
Publication Date: 1982
Persistent Ratee Contaminants in Performance Appraisal.
Van Fleet, David D.; Chamberlain, Howard
The hypothesis that conventional approaches to evaluating contaminants in performance appraisal overlook important individual ratee effects was examined. A rating form was developed that consisted of the following dimensions and behaviors: warmth; guided discourse or indirect teaching methods; control of subject matter; enthusiasm and reinforcing; organizing and managing; presenting and explaining; evaluating; and advising and counseling. Administration of the form to evaluate 23 instructors resulted in approximately 1,500 observations per semester. The reliability of the form and its factor stability were assessed, and possible contaminants were checked to assure that the evaluations were more likely to result from the instructor's performance than from student or course factors. It was found that 8.6 percent of the instructor ratees had persistent and significant contaminants associated with their evaluations; a looser definition of "persistent" pushes the figure to 34.7 percent. It is suggested that the evaluations may not be assessing performance accurately because of ratee contaminants, including expected grade in the course, the time at which the course begins, the time and effort required of the student, and the student's major. These contaminants occurred in spite of the fact that the instrument was found to have face validity, factor stability, and internal consistency. It is proposed that adjustments could be made on an individual basis and only for those contaminants that are persistent for each instructor. However, what is needed is a practical decision rule that would permit users of such evaluations to make necessary adjustments in the appraisals to correct for such persistent effects. Interactions among contaminants should also be addressed. (SW)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A
Note: Paper presented at the Annual Meeting of the American Psychological Association (Washington, DC, August 1982).