Increasing the Significance of Course Evaluations in Large-Enrollment Geoscience Classes

Abstract

The goal of this study was to identify student evaluation of teaching (SET) questions that could be biased against large-enrollment geoscience classes. SET questionnaires were collected from twenty geoscience departments; individual questions were examined for potential bias by examining differences in SET responses for classes of differing size and discipline using data from the Department of Geography and Earth Sciences at UNC Charlotte.At UNC Charlotte, lower-level courses receive 6% lower evaluation scores than upper-level courses and overall ratings by students in larger classes were 12% lower than those in classes with fewer than 75 students. Comparisons between instructors teaching different disciplines should be reviewed with discretion. When introductory geography and earth science classes are compared, geography instructors scored 20% higher than earth science instructors (n = 40 sections for geography, n = 38 sections for earth sciences).Student evaluations should also be viewed in the context of the difficulty of the professor. At UNC Charlotte, evaluations are plotted against the mean grade-point average assigned by the professor. These plots are both summative when considered by the Reappointment, Promotion, and Tenure (RPT) committee, and formative when used by the instructor to sense how difficult and/or effective he or she is relative to other instructors.