Mining Free-Text Course Evaluation Responses: Lower Perceived Importance of the Instructor in Online Classes

• In Event: Friday Roundtable Session 9:45 am
  In Roundtable Session: Innovation in Statistical Methods (Table 31)

Abstract

This study explores what can be learned from student evaluations of teaching (SETs) at [institution]. Following previous literature, topic models were fit to the free-text responses to two questions from the SET survey. Subsequent regression models revealed differences between online and in-person classes in the topics students were most likely to mention; notably, students in online classes were less likely to mention the instructor. This pattern is consistent with other evidence that students perceive the instructor to be less important in online classes, a finding with important implications for how institutions manage teaching assignments. This study illustrates how free-text SET responses can be mined for insights that inform institutional decision-making.

Author

• Abby Kaplan, Salt Lake Community College