Economics is a male-dominated profession in post-secondary education and in industry. In the last few years, some economists have been challenging that norm and calling out institutionalized practices and conditions that discourage more diversity in their profession. Both the Canadian Economics Association and the American Economics Association have undertaken surveys of their membership to identify the representation of different demographic groups, and to hear from members of those groups about their academic or workplace experiences. Now, a group of 101 economists has released the results of a study that looks at gender-related behaviour in a significant part of academic work: the research seminar.

At these seminars, a researcher presents a summary of work that they are currently undertaking, or a project that they have completed. The researcher gets a set amount of presentation time, and then there is time allocated for questions, comments, or feedback from the audience. This format is used in research presentations at universities, where colleagues share their work with each other, but is also part of job interviews, with the job candidate presenting their research, and academic conferences.

I’ve sat through a lot of these seminars in my own academic discipline, and also presented my own work in this format. The first research presentation I did was at an academic conference where I presented a summary of my master’s thesis. I had already presented my thesis at my university to a panel of professors, as part of the requirements to complete my degree, but that presentation was in a familiar setting with people that I already knew – plus I got a lot of my friends to attend, so I would have friendly faces in the audience. Being totally unfamiliar with how academic conferences worked, and not knowing anyone at this bigger event, I was completely and utterly terrified, because I had no idea what kinds of questions the audience members would ask.

As it turned out the audience was very kind, and this was a friendly and welcoming conference. But since then I have witnessed some truly terrible behaviour from seminar attendees, like asking long-winded “questions” that were simply excuses to promote their own research, or dominating the entire question period with increasingly hostile questions to the presenter. Questions and comments can be very useful to the presenter in suggesting possibilities for additional dimensions to the research. But there are also many different ways to approach any research question, and too often a question is really the questioner challenging the presenter as to why the presenter didn’t do the research the way the questioner would have done it. A research seminar, like a lot of processes in the academic world, runs on unspoken norms of mutually respectful behaviour – and it’s very easy for those with their own agendas to ignore those norms, with very negative effects.

Many academics would agree, anecdotally, that women are far more likely to be the targets of negative behaviour in seminars: behaviour like being interrupted, being talked over, or subjected to aggressive or negative questioning. The researchers decided to investigate whether this anecdotal assumption was true, because, in their words, “the[re is a] distinctively aggressive culture of economics seminars, and [a] continuing under-representation of women among the senior ranks of the economics profession” (p. 2).

During 2019, the 101 economists involved in this research project collected data at 468 university research seminars, including 178 presentations by job candidates. The data covered 33 different academic institutions. The researchers collected information such as the demographic characteristics of the presenter and the audience members (gender, faculty member or student, senior or junior faculty member), the length of the seminar, the topic of the presentation, whether the seminar was a presentation by a job candidate, who asked questions of the presenter, the number and type of questions asked, the “tone” of each question, and whether there were any explicitly stated “rules” for the session, such as requesting audience members not to ask questions during the presentation itself.

As you can imagine, this large amount of detailed data produced many interesting results. But here are some of the more startling ones.

• There are 4.5 times as many questions from men as from women during regular seminars—and more than 7 times during job market talks—despite men only outnumbering women 2 to 1 in attendance.
• Only 8 percent of seminars had any rules in place, such as not asking any questions during the first 10-15 minutes or only asking clarifying questions initially.
• Among rated interactions [between the presenter and a questioner], 70 percent were coded as supportive, 15 percent as patronizing, 9 percent as disruptive, 3 percent as demeaning and 3 percent as hostile. [It’s important to note that the majority of interactions were not assigned codes, although the data collectors were given examples of each category of interaction.]

Data were also collected at a summer seminar series, by another set of researchers. These data covered 433 presentations. At these seminars, the scheduled session time was much shorter than at the university seminars, so there were stricter rules around audience questions. Because the seminar attendees were aware that data were being collected, the researchers didn’t collect data on the tone of the questions – but they did find that, as in the other data set, women were asked more questions than men. Additionally, “the norm that questions should be held until later is 23 percentage points more likely to be breached when a woman ispresenting in a talk [on macro-economics] compared to a man” (p. 23).

The researchers conclude that there are major differences in how male and female presenters are treated by audiences at seminars. While they note that it is impossible to tell whether this treatment translates into long-term differences in career patterns, they also point out that “many of us have heard stories of friends and colleagues whose bad experiences in seminars have led them to re-evaluate whether a career in economics is really the best choice for them” (p. 24).

This is a very important point, illustrating what researchers in other fields have labeled “micro-aggressions”. People may not necessarily be discouraged from participating in, or staying in, an occupation because of a few bad experiences. What may be driving them out, or keeping them out, is continually experiencing behaviour that implicitly suggests they shouldn’t be there or that they are producing poor-quality work. The researchers add that “one characterization of the emerging literature on gender biases within the economics profession is that every rock we look under reveals yet another way in which existing institutions are biased against women” (p. 24). This suggests that these implicitly discriminatory behaviours can be very difficult to combat because they are so widespread and generally not investigated, or even recognized for what they are.

The entire study is well worth reading – it’s very accessibly written, even for readers who may not be familiar with detailed statistical analyses. In addition to the findings I’ve mentioned, there are others that suggest potentially very intriguing directions for future research, such as different norms for behaviour in seminars around specific topics within a single academic discipline; whether audiences respond differently to junior and senior faculty members’ seminar presentations; and whether audiences respond differently to presentations by faculty members from elite andfrom non-elite institutions.

But one of the biggest contributions of this study is that it can be a blueprint for similar studies in other academic fields. I would love to see a study like this carried out in business schools; women are under-represented among business school faculty, even though almost half of business students are women, and there are very few women faculty members in some business disciplines. Many factors are contributing to that situation, but investigating the type and quality of interactions in formal academic presentations may be one way to start uncovering hidden factors that could be having profoundly negative effects.