THE EFFECTS OF TEACHER SEX AND STUDENT SEX ON STUDENT RATINGS OF COLLEGE INSTRUCTORS: A FACTOR ANALYTIC INVESTIGATION

Investigations of possible correlates of student evaluations have been conducted to determine whether extraneous variables influence, and hence, invalidate, student ratings. Sex of student and sex of instructor have been studied in this regard. Although a number of investigators found that student ratings were not affected either by student sex or instructor sex, other results have indicated that sex of student and of instructor may affect some aspects of student evaluations of their instructors. For example, in some studies, women tended to rate their instructors higher than did men in three general areas: competency, structure, and personality. In other studies women instructors were seen as more student oriented than men, while some investigators have reported that male instructors were rated better in presenting material. The purpose of the present study was to attempt to clarify the effects of teacher and student sex on student ratings. The research was factor-analytic in nature and used as data student ratings obtained from the Southern Illinois University (SIU) Instructional Improvement Questionnaire (IIQ). A previous factor analysis of the Instructor Evaluation section of the IIQ yielded three subscales: (1) Student Orientation, (2) Assignments and Grading, and (3) Presentation of Material (Pohlmann, 1976). Student ratings on the IIQ were obtained from SIU for the Spring 1973 and Winter 1974 semesters. Complete records were obtained for 4,585 male and 1,371 female undergraduates. The first 19 items of the Instructor Evaluation section of the IIQ were subjected to a principal axes factor analysis with rotation by the direct oblimin method. Four factors were extracted; three similar to Pohlmann\u27s (1976) plus an additional factor which seemed to reflect classroom behavior on the part of instructors. All further factor analyses were rotated to maximal congruence with this entire sample solution. Most factor analyses in the present study were performed using data from a sample of courses taught by 30 male and 30 female instructors matched by department and course level (N of students in the sample = 1,168). The matched sample was used in order to minimize possible interactions between sex and subject matter or course level and to reduce the large difference in the numbers of students taught by male and female teachers. After oblique procrustes transformations were performed to maximize congruence with the target matrix, the factor patterns obtained for male and female students were found to be congruent. Factor scores for each factor were estimated for each group, and mean differences were not significant between the profiles of male and female students. Correlations between the four obtained factors and item 20 of the Instructor Evaluation section of the IIQ ( In general, taught the class effectively, used as an overall measure of teaching effectiveness) were computed and were not found to differ significantly for male and female students. Similar results were obtained when the same analyses were performed comparing male and female instructors. The results of the analyses did not lend support to predictions of differential rating patterns by sex of student and/or sex of instructor. However, methodological problems relating to the factor solution and the procedure used for matching instructors limit the strength of conclusions that might be drawn from the study. It was suggested that if differences in teaching effectiveness exist between male and female teachers, the rating scales currently in use might not be sensitive to these differences. Similarly, if female and male students perceive different behaviors to be components of teaching effectiveness, perhaps these behaviors are not contained in commonly-used rating scales. Future research should be designed with these possibilities in mind