Gender bias in student evaluations of teaching : do debiasing campaigns work?

This paper presents the results of a field experiment aimed at reducing the gender bias in teaching evaluations at a higher education institution. In the intervention, before they completed the teaching evaluation questionnaire, students were individually randomized in three groups. One third received an email invitation to watch a video that informed them of the existence of implicit bias (treatment 1). Another third of the students received an email invitation to watch a video with an explicit message that made them aware of the presence of gender bias in teaching evaluations (treatment 2). This second video also mentioned the fact that the academic literature has shown that this form of discrimination often originates with male students. At the end of both videos, all the students treated were asked to avoid displaying prejudice when they completed the questionnaire. The final third of students was assigned to the control group and did not receive any message. The results indicate that the video on implicit bias served to reduce the score gap between male and female lecturers. However, the video on gender bias had an unintended consequence: male students in the treatment group awarded their female teachers even lower scores than did the control group - confirming the risk of backlash or reactance in this kind of debiasing campaign. Such an effect was found to be particularly strong in female-dominated academic contexts.