When the university starts sending out teaching evaluation …

When I say "anyone": needless to say, the people who are benefitting from the bias (like me) are the ones who should helping to correct it. Men in math, this is your job! Of course, it should also be dealt with at the institutional level, not just ad hoc.
OK, on to my email:
2/16

"You may have received automated reminders about course evals this fall. I encourage you to fill the evals out. I'd be particularly grateful for written feedback about what worked for you in the class, what was difficult, & how you ultimately spent your time for this class.

3/16

However, I don't feel comfortable just sending you an email saying: "please take the time to evaluate me". I do think student evaluations of teachers can be valuable: I have made changes to my teaching style as a direct result of comments from student teaching evaluations.
4/16

But teaching evaluations have a weakness: they are not an unbiased estimator of teaching quality. There is strong evidence that teaching evals tend to favour men over women, and that teaching evals tend to favour white instructors over non-white instructors.
5/16

Here is some specific information which may be useful if you wish to learn more about the biases described above:
6/16

Mengel, Sauermann, Zölitz - Gender bias in teaching evaluations. This one directly addresses mathematics courses, which I think is useful. The short version: female instructors receive lower evals than they deserve, mostly because of male students.
https://t.co/znL2HgtTSz
7/16

Boring - Reducing discrimination through norms or information. This one is a study which basically says that sending a message like this is a good idea!

https://t.co/eK6oD3I7sI

The conclusions of the study are as follows:
8/16

a) Simply reminding people not to be biased when filling out their teaching evaluations seems not to have an effect.
b) If as well as the reminder, you inform people that people that bias really does exist, in their exact setting, then does help reduce the resulting bias.
9/16

Chávez and Mitchell - Exploring Bias in Student Evaluations: Gender, Race, and Ethnicity. An experiment in the setting of an online course, which made it possible to compare variables in a much more controlled way.
https://t.co/3XVaLofLoe
10/16

.@mcgillu's report on gender equity in course evaluations (not peer-reviewed) shows statistically significant differences in course evaluations depending on the instructor's gender.

https://t.co/TO2HCgv4bT
11/16

This message feels particularly important as none of this is mentioned in McGill's guidelines for interpreting course evaluations (https://t.co/WSmNX8jCBn)
12/16

I hope you will all take seriously both the existence of this problem, and the idea that by paying attention, and having information about where and when this happens, you can actually help improve the situation. "
End email.
13/16

Then I include a P.S. with a few other references for any of my students who want to learn more:

* Carolina Women's centre - Teaching evaluations and bias
https://t.co/fGK3jTZZNg
14/16

* Bias in Student Evaluations of Minority Faculty: A Selected Bibliography

https://t.co/cKvcgP4Vt0
15/16

* Huston - Race and Gender Bias in Higher Education: Could Faculty Course Evaluations Impede Further Progress Toward Parity?

https://t.co/Vb76Q7TTZl
16/16