I’m interested in trying to quantify potential biases in academia, particularly those that are gender-based. One place these biases could appear is in comparing the proportion of men to women in the audience at an event (class, lab meeting, conference, etc.) to the proportion of men to women who interact in some way with the speaker (the professor of a class, the person presenting at a lab meeting, the speaker at a conference, etc.). I believe this is an interesting area to look at because there are potentially huge ramifications to someone’s career in academia if their participation in an event is stifled because of their gender or status, or the gender or status of the speaker.
I started counting the men and women in my undergraduate classes and tracking who participated and who did not. Once I developed a system, I started tracking these numbers at other events. My results show that men interact with the speaker at a much higher rate than women do — even when men make up less than half the audience, they are the ones participating way more than 50% of the time. Women’s participation rates are higher when the speaker is a woman compared to a man, though they still don’t participate equal to their proportion in the audience. Finally, men faculty members are the worst “sinners”: in my data, they make up about 10% of the audience but nearly half the interactions with the speaker.
I am developing a Shiny app to share my results and data. This will be coming in the next few weeks.