“Do the best professors get the worst ratings?” asks Nate Kornell Ph. D in Psychology Today. He cites two studies. One study compares groups of students watching a one minute video on cat genetics.
- “In the fluent speaker video, the speaker stood upright, maintained eye contact with the camera, and spoke fluidly without notes.
- In the disfluent speaker video, the speaker stood behind the desk and leaned forward to read the information from notes. She did not maintain eye contact and she read haltingly.”
The first speaker earned higher ratings for effectiveness and the amount they learned. However, both groups did equally well on the test. Higher ratings did not correspond with teaching effectiveness.
The other study included 10,000 cadets at the Air Force Academy. They were randomly assigned introductory calculus classes with the same syllabus and standardized exam and they took advanced classes with standardized exams. Less experienced professors got the highest ratings and their students got the highest scores in the introductory classes. But, experienced professors’ students earned higher scores in advanced classes.
Nate Kornell: “To summarize the findings: because they didn’t teach to the test, the professors who instilled the deepest learning in their students came out looking the worst in terms of student evaluations and initial exam performance. To me, these results were staggering, and I don’t say that lightly.”
Student evaluations have questionable value for evaluating teaching. This aligns with my observations. Even so, they can provide a valuable window into student perceptions that teachers can use to enhance classes.
Standardized exams have questionable value for evaluating teaching, too. They can measure alignment of teaching/learning to a test, rather than depth of understanding a student develops during a class. This implies the foolishness of increasing the importance of standardized testing in teacher evaluation if we value students’ depth of understanding of a subject.
(Thanks to Mark Guzdial for pointing out this study in his Computing Education blog “Learning for today versus learning for tomorrow: Teaching evaluations.”)