The effect of right or left placement of the positive response on Likert-type scales used by medical students for rating instruction

Purpose. To determine whether students' ratings of instruction demonstrate a primacy effect, to determine whether the primacy effect relates to the number of response options, and to assess whether the primacy effect relates to how favorably the instructional activity is rated. Method. In 1995-96 six different forms of a 13-item course-evaluation questionnaire (with Likert type items) were used to evaluate two second-year courses, Respiratory and Hepatic, at the University of Wisconsin Medical School; the first was one of the most highly rated courses at Wisconsin, and the second was less highly rated. The forms differed by whether they contained five, six, or seven response options, and whether the ''strongly agree'' rating was on the left side or the right side of the page. The second-year class that participated in the study comprised 140 students. The six different forms of the course evaluation questionnaires were randomly distributed to the students in equal numbers. Results were analyzed with a number of statistical methods. Results. Completed questionnaires were obtained from 132 students (94%) in Respiratory and from 119 students (85%) in Hepatic. Overall, the forms with the positive rating on the left side had more positive ratings and less variance. For Respiratory, primacy affected the response variance. For Hepatic, primacy affected the response means. Conclusion. Faculty-evaluation systems are increasingly using students' ratings for making important decisions regarding salaries, teaching assignments, tenure, etc. The evidence that the primacy effect influences such ratings highlights the need to standardize as much as possible how such ratings are obtained.