Like most faculty I take teaching evaluations very seriously. Every year I read mine, and rethink assignments and readings based on student feedback. As at most universities, teaching evaluations in my department are also a key instrument to measure faculty performance for promotion and tenure decisions. But what if teaching evaluations are inherently biased?
One of my colleagues recently shared a post at the LSE Impact Blog, which discussed in detail evidence that female instructors rated lower on teaching evaluations. In one particular case, the students were taking classes with a common final exam, so there was a means to evaluate how effective instructors were in teaching the learning outcomes. The bottom line was that female instructors tended to measure more poorly than their male peers on course evaluations because of bias. I think that this particular blog post, and the articles that it refers to, all merit reading and careful discussion.
At a recent Portland State University workshop that I attended the speaker argued that when evaluating teaching it is critical to draw on multiple forms of evidence. Based on the information in this blog, departments across academia should perhaps consider creating a richer set of metrics to judge how effective faculty are at teaching, rather than relying on the time-honored course evaluations. We could, for example, ask an outside person to evaluate the course shells in the learning management system; look at student success data to show how students perform in later classes; evaluate the assignments that students have completed; and measure the number of new courses that they’ve created. This will entail more work. But there are reasons for concern that our existing system is flawed, and that this may be undermining women’s success.
Shawn Smallman, 2017