Why Good Teaching Evaluations May Reward Bad Teaching

In this article, I address the paradox that university grade point averages have increased for decades, whereas the time students invest in their studies has decreased. I argue that one major contributor to this paradox is grading leniency, encouraged by the practice of university administrators to base important personnel decisions on student evaluations of teaching. Grading leniency creates strong incentives for instructors to teach in ways that would result in good student evaluations. Because many instructors believe that the average student prefers courses that are entertaining, require little work, and result in high grades, they feel under pressure to conform to those expectations. Evidence is presented that the positive association between student grades and their evaluation of teaching reflects a bias rather than teaching effectiveness. If good teaching evaluations reflected improved student learning due to effective teaching, they should be positively related to the grades received in subsequent courses that build on knowledge gained in the previous course. Findings that teaching evaluations of concurrent courses, though positively correlated with concurrent grades, are negatively related to student performance in subsequent courses are more consistent with the assumption that concurrent evaluations are the result of lenient grading rather than effective teaching. Policy implications are discussed.