May 24, 2005
Student Evaluations
Posted by Gordon Smith

Brayden King points to this paper on the relationship between student evaluations, perceived quality, easiness, and sexiness. This is from the abstract: "half of the variation in student opinion survey scores used by universities for promotion, tenure, and teaching award decisions may be due to the easiness of the course and the sexiness of the professor."

In addition to making me wonder why I didn't become a sociologist, this finding caused me to ask how the authors obtained a "sexiness" measure. Prior studies support this finding, attributing the higher evaluations to a "Halo Effect," but I am happy to say that this dimension has never appeared on a student evaluation for one of my courses (though a student once opined that I am a good dancer). Sure enough, students in this instance were not asked to rate the sexiness of their professors. Instead, the authors downloaded data from and "calculated an Average Sexiness score for each faculty member by dividing the number of total posts for a professor by the number of posts noting that the professor is sexy." Oh, and if you are wondering about methodological problems, the authors have an entire helpful sentence: "Since student posts at are voluntary, our sample data are self-selected."

What kind of students rate their professors on Did the authors consider the fact that professor can rate themselves? Are the students who rate professors on that site more likely than other students to be interested in whether the professor is sexy? Or an easy grader? Or might they believe that this is the sort of information that would be most useful or sought after by people reading I'll stop, but I trust you get the idea.

A number of prior studies find a correlation between easy grading and high evaluations, though this connection appears to be disputed in the literature. My intuition is that the connection is not so strong in law schools, especially where curves are mandatory and strict.

In my experience, the biggest factor affecting student evaluations has been whether I grade in-term assignments. If the evaluations are filled out before the students receive any grades, they are inevitably higher than if the students have had a mid-term. As it turns out, my experience is supported by the research. (See ENGDAHL, RICHARD, ROBERT KEATING & JOHN PERRACHIONE (1993). Effects of Grade Feedback on Student Evaluation of Instruction. Journal of Management Education, 17(2), 174-184.)

Maybe I am blind, but I think most students want professors who are fair and competent. My advice to new professors: assuming a baseline of fairness and competence, the best way to raise your evaluations is to learn your students' names and use those names, not only in class, but in the hallways and student lounge. In other words, take a sincere interest in your students. (I should note that I have sometimes fallen short of my aspiration to know all of my students by their first names, and when that happens, it makes me feel as if I have not fully done my job as a teacher.)

If that doesn't work, you can try a trick used by a former colleague: on evaluation day, she baked cookies for her class. No, I am not kidding.

Juan Non-Volokh points to this literature review, which emphasizes the correlation between easiness and high evaluations.

Thanks to Kaimi at Prawfsblawg for the tip to Brayden. (Kaimi, go to the gym, but not to improve your evaluations!)

UPDATE: Paul Caron has some views from the perspective of a tax professor, including links to last year's  Law Prof Hunk Contest.

Law Schools/Lawyering | Bookmark

TrackBacks (0)

TrackBack URL for this entry:

Links to weblogs that reference Student Evaluations:

Recent Comments
Popular Threads
Search The Glom
The Glom on Twitter
Archives by Topic
Archives by Date
April 2018
Sun Mon Tue Wed Thu Fri Sat
1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 18 19 20 21
22 23 24 25 26 27 28
29 30          
Miscellaneous Links