Do Student Evaluations Make Sense

Academics have long debated the validity of using student course evaluations as accurate measures of teaching performance.

Some argue that results can be skewed by low student response rates, students’ previous experience with the material, or the likelihood that students will judge professors who are teaching difficult courses more harshly than those teaching easier subjects. A recent paper by two educators from the University of California, Berkeley, make the argument that student evaluations, at least as the primary basis for promotion and tenure decisions, should be abandoned.

Authors Philip B. Stark, professor and chair of the statistics department, and Richard Freishtat, senior consultant in the university’s Center for Teaching and Learning, argue that qualitative, rather than quantitative, measures of teaching performance would be more effective ways to evaluate quality of teaching. These include in-class observations, in which faculty sit in on each other’s lectures and offer detailed written feedback, as well as the creation and evaluation of teaching portfolios. Such measures were piloted in UC Berkeley’s statistics department in spring 2013.

The authors write that they believe these approaches “better reflect whether faculty are dedicated teachers” and “comprise a much fairer assessment” of teaching performance.

> “An Evaluation of Course Evaluations” is available at www.stat.berkeley.edu/~stark/Preprints/evaluations14.pdf.