Personal characteristics may affect evaluation of teaching: A prospective study in medical students

Dear Sir

Student Evaluation of Teaching (SET) is widely used for assessing the quality of teaching process and instructor performance but the validity of ratings is controversial (Irby et al., 1987). We hypothesized that specific personal characteristics of medical students may affect their SET ratings. The study was prospectively performed in students attending Internal Medicine (7th and 8th semester of the 6-year programme) at the Athens University Medical School. During the 7th semester, students come in touch with clinical medicine for the first time in their career (156 teaching hours, 50% in small groups in the wards of the Internal Medicine Department). During the 8th semester (126 teaching hours, 50% in wards) students have previous experience of the teaching process and the subject of Internal Medicine. Examinations take place at the end of each semester; therefore, the obtained grade for the 7th semester is known at end of the 8th semester. Participating students filled in anonymous questionnaires at the end of each semester. In addition to SET, visual analogue scales regarding their initial expectations from the teaching process, interest on the educational subject, self-perception of ‘student quality’ and ambition as a future professional physician were obtained, whereas the 7th semester's exam grade was recorded.

Analysis of 446 questionnaires (comparable results across the three academic years with median participation rate of 91%) showed that a higher initial interest in the subject of teaching was consistently associated with higher SET ratings (p < 0.0005). A trend for granting lower SET ratings by students with lower initial expectations and worst examination performance in the previous semester (p : 0.080 and p : 0.069, respectively) was found, suggesting that more ‘indifferent’ towards their education students evaluate the educational process in a ‘stricter’ manner. In contrast, gender, individual ambition status and the degree of self-perception of quality as a student did not influence SET ratings.

Taken together, these data suggest a correlation between effective student learning and better evaluation of teaching. Higher exam grades imply increased student learning, as a result of effective teaching during the semester. Thus, increased student learning leads to student satisfaction, which in turn leads to higher SET ratings.