Effect of administering student evaluations of teaching after final exams in veterinary medical schools

In higher education, instructors and administrators use student evaluations of teaching (SETs) as formative and summative assessments of instruction; thus, they need adequate response rates. Veterinary students are often requested to complete over 30 SETs each semester, and response rate is shown to decline as the number of SETs increases. Nonresponse threatens the validity and reliability of results. Allowing students to complete SETs after final exams was postulated to help increase response; however, students’ knowledge of their final course grade has been previously shown to negatively influence SET scores. A possible rationale for this influence is attributional bias, in which people attribute success to themselves and failure to someone else.

This mixed-methods case study explored how administering SETs to veterinary students after final exams (versus before) affected numerical scores on SET items, completion rate, and volume and substance of comments. Participants (n = 262) were randomly assigned to before or after finals groups, and 171 students completed 2,926 SETs and 1,149 open-ended comments. Attributional theory guided the analyses.

Students were more likely to complete evaluations before finals (versus after), and first-year students completed more SETs than third-year students. Compared to the prior year, in which SETs were administered before finals, students in the study year completed 31% fewer SETs. Timing of SET delivery did not significantly affect the number of students providing comments or comment length. Timing also did not affect SET item scores, but third-year students rated instructors higher than first-year students on five of 10 items. Students’ expected grade was positively correlated with all 10 SET items, suggesting attributional bias might be occurring, but not due to timing.

Three themes emerged from student comments, regardless of timing: classroom climate, achievement striving and goal attainment, and operational deliverables. Students provided insights into the essence of their learning, illustrating how instructors facilitated or impeded achievement of student goals and providing their perceptions of the curriculum. Regardless of the focus of student comments, they were more positive than critical before and after final exams.

Alternatives to increase SET response rates and to refine and improve programs to assess teaching are provided.