Analysis of Course Evaluations in an Engineering Institute Using a Hierarchical Linear Model

Despite the wide usage of course evaluations as tools to assess and improve the quality of college teaching by higher education institutions, there are ongoing discussion on the reliability and validity of the evaluation results. This study aims to provide empirical insights into this discussion by analyzing the course evaluations of a small-sized engineering college in South Korea using a hierarchical linear model. Through a priori analysis, individual student-level and course-level factors related to students’ course satisfaction were incorporated into the model. Firstly, the unconditional model results of the hierarchical linear model revealed that the variance due to differences between courses accounted for only 6% of the total variance. Among student-level variables, students' gender, class attitudes, and perceived difficulty of the course were found to be significantly associated with course satisfaction. Contrary to expectations, courses perceived as difficult showed higher satisfaction levels. This study concludes with recommendations for the desirable utilization of course evaluation results.