Who’s Asking? at AFS 148
August 25, 2018
A few weeks ago, my partner introduced me to Cathy Deng’s online tool, “Who’s Talking?”, which allows you to record the proportion of a conversation in which a “Dude” versus “Not a Dude” is speaking. I immediately started thinking about scenarios where it could be used to demonstrate some of the underlying problems that I see first hand in science. Last week (August 20 - 24) I attended the American Fisheries Society’s (AFS) 148th Annual Meeting in Atlantic City, New Jersey. I know that AFS is working hard to increase diversity, inclusion, and equity in the Society, and has a Equal Opportunities Section that I’m looking forward to getting more involved with. I decided that a great first-time use of “Who’s Talking” would be to monitor the five minute question periods following scientific talks at AFS, and see if any useful information could be collected about who is dominating the conversation from a gender perspective.
Quoting a 2017 study by 1Hinsley et al., “…these methods treat gender as a binary dichotomy, which is not the case. In addition, the visual identification of gender assumes that people identify as the gender they appear. This will not always be true…” I apologize for the exclusion and acknowledge the limitations – I hope that others will share ideas about how to make this kind of informal observational study more inclusive.
How were data collected?
I informally collected information during the question period of talks presented at AFS. In each instance, I recorded the number of males and females in the audience (excluding myself and the moderator), the gender of the presenter, the gender of the person asking each question, and the duration of the question using “Who’s Talking?”.
What data were collected?
-
A total of 79 questions were asked at 26 talks and the data and code can be found on GitHub.
-
Of a total 864 audience members, 343 (39.7 %) were women, with women representing an average of 39.1 % of the audience in each talk.
Is there is a difference in the number of questions asked by men and women?
-
Of a total 79 questions, 17 questions (21.5 %) were asked by women, with women asking an average of 19.4 % of the questions in each talk. This translates to men asking 3.6 questions for every one question asked by women.
-
In 69.2 % of talks, women asked no questions. In these instances, men asked one to five questions per talk.
-
Across talks, men asked significantly more (2.38 +/- 1.44) questions per talk than women (0.65 +/- 1.16) (Wilcoxon test, p < 0.01).
Is there a difference in the duration of questions asked by men and women?
-
Of a total of 934 seconds of questions, 208 seconds (22.3 %) were asked by women.
-
Across questions, there was no significant difference in the duration of questions asked by men (11.7 +/- 7.73 seconds) and women (12.2 +/- 6.61 seconds) (Wilcoxon test, p = 0.616).
Does it matter who asks the first question?
-
Women asked the first question in five (19.2 %) talks while men asked the first question in 21 (80.7 %) talks.
-
100 % of the talks where women asked the first question had women asking proportionally as many or more questions as men; this was the case in only 9.5 % of talks where men asked the first question.
Final Thoughts
This is far from a perfectly scientific study; I’m sure that we can all think of confounding factors! For example, I collected this data while attending talks that I was specifically interested in, including anything related to data science, sharks, and community outreach. Maybe attending topics from other domains (genomics, physiology, etc.) would have had completely different results! While I never intended for this to become anything more than an interesting conversation piece, I would like to challenge women to continue asking questions at conferences and seminars - and, specifically, to make an effort to ask the first question following a talk. If you have any thoughts about who is dominating conversations in your environment, I would highly recommend trying out the “Who’s Talking?” app!