Comparing instructor self-perception versus student perceptions using the same teaching evaluation instrument: a study of computer science courses in an urban master's degree program

Abstract

This study compares instructor self-perceptions with student perceptions of teaching quality using the same 16-item evaluation instrument. Three hypotheses were investigated: (1) Instructors' self-evaluations will be higher than those of their respective students; (2) The more similar student-instructor perceptions, the more likely instructors will receive a higher score compared to when student-instructor perceptions are more divergent; and (3) Students taking a course as a major requirement will be more critical of the instructor than those students taking the course as distribution requirement or an elective. A total of 1,524 individuals (1,452 graduate students and 72 instructors) in a part-time evening computer science program participated in the study of 79 courses over the spring and fall semesters of 1996. Overall, instructors generally perceived themselves more positively than their students, although statistically significant differences were observed for only three relevant items (involving grading fairness, presentation clarity and instructor enthusiasm) of the 16 items evaluated. Instructors whose perceptions were more similar to their students were generally rated higher than those instructors whose perceptions were more divergent from their students; however, the difference was not significant. Finally, contrary to the third hypothesis, the reason for taking a course did not have a significant effect on student ratings of the instructor.