When Beryl Cummings asked her first-ever question in the auditorium at a genetics conference, she chose a topic she knew a lot about, formulated her question as meticulously as she could, and addressed her query to a female presenter.
In science, questions matter a lot, said Cummings, who was then working on her doctorate in computational genomics at Harvard. But as a young female scientist speaking up in a public forum, she said, the stakes just felt a little higher.
It’s a point of deepening consensus: For women keen on careers in science, technology, engineering and math, getting into the room may just be half the battle.
A bit over half of all doctoral degrees in the biomedical sciences are now earned by women. But if their growing numbers in STEM fields are to translate into meaningful scientific contributions, these women need their voices to be heard — in classrooms, at meetings and on conference daises.
New research offers some surprising insight into an often overlooked variable in this equation. For a female scientist’s voice to be heard, she must first decide to use it. That can mean standing up at a meeting and asking a question. And that’s not happening as often as one might expect.
The study was published last month in the American Journal of Human Genetics. It culled data from several years’ worth of genetics conferences and found that men were overrepresented among the questioners at scientific meetings and symposiums. Women, on the other hand, were coming up short.
The study authors say there may be many explanations for women’s reticence to speak up at professional conferences. But they also suggest that more documentation of the gender gap, and a wider awareness of its existence, would help rectify the imbalance.
After all, the problem isn’t limited to the genetics field. Previous work has shown that women are underrepresented as speakers in a wide range of scientific disciplines, including microbiology, virology and evolutionary biology.
“We’re missing out” when the voices of women and minorities are not fully represented, Dr. Francis S. Collins, director of the National Institutes of Health, said in an interview.
Collins himself has recently called attention to the lack of diversity — gender and otherwise — among speakers at scientific meetings.
“I want to send a clear message of concern,” he wrote in public statement. “It is time to end the tradition in science of all-male speaking panels, sometimes wryly referred to as ‘manels.’”
Declaring that “it is not enough to give lip service to equality,” Collins put conference organizers on notice: To secure his attendance, they’ll have to furnish evidence that “scientists of all backgrounds” have been considered and invited to present, speak and lead sessions.
“And I’m not going to accept the idea there aren’t enough women in this field to fill those spaces,” Collins said in an interview. “It’s simply not true.
The new research focusing on genetics conferences offers a nuanced profile of women’s representation in one of science’s most cutting-edge fields.
About 45% of the roughly 8,000 geneticist who belong to the American Society of Human Genetics are women, and the work presented at annual conferences suggests that female geneticists are contributing heavily to their field. In nine of 14 sub-fields of genetics, the proportion of papers presented by women exceeded their overall representation in the society, the study authors found.
But the pattern was different in the question-and-answer portion of the proceedings, where professional reputations are shaped and research agendas are advanced. The analysis of recordings made at four of the society’s annual meetings revealed that men asked 65% of the questions posed to speakers.
Even in sessions where a majority of the audience members were women, men dominated the questioning.
“When women are 70% of a room, they still asked only about 40% of the questions,” said Natalie Telis, who led the study with Emily Glassberg while both were pursuing doctorates at Stanford University.
Now working in the private sector, Telis reckons that an audience would need to be 80% to 90% women to assure that the question-asking would be split evenly between men and women.
Collins said those measurements offer a robust confirmation of his own experience.
“I’ve gone to lots of scientific meetings and I can confirm that would have been my observation as well: Men are likely to put themselves forward, demonstrating their competence and their willingness to self-promote,” he said. “Women are less likely to do that.”
Telis said that conference attendees who publicly challenge, contextualize or simply amplify another’s comments speak volumes about their sense of security among colleagues. It is perhaps no surprise, then, that male questioners tended to direct their queries to men while female questioners were more likely to direct theirs to women, as Cummings did.
That particular finding suggests that the divide between men and women is often as much about implicit preferences for “people like me” and biases against “others” than it is about gender discrimination per se, Telis said. It also raises the possibility that making people more aware of those preferences could be a powerful way to counter those subtle biases.
Consider this: At the opening session of the 2017 meeting of the American Society of Human Genetics, Telis presented some of her preliminary findings and touched off a broad discussion focusing on women’s participation in question-and-answer sessions.
In its wake, she and her co-authors detected a measurable difference. The proportion of questions asked by women following invited and plenary talks was similar to that seen in prior years. But the proportion of talks that were followed by zero questions from women fell from 51% to 30%.
Cummings, whose first-ever conference question was posed during that meeting, said that the new data certainly reflects her own experience.
As she considered whether to stand and ask a question, she felt the weight of the audience’s potential judgment. Over the years, she’d heard a few male colleagues ask questions that were silly or misinformed. But such a misstep didn’t feel like an option for her. This was no occasion for bumbling or for wandering into unknown territory, she concluded.
Now she has some data to back up those impressions, and so do colleagues who may have been skeptical of the barriers that many women perceive.
“People talk about this all the time,” said Cummings, who is now a post-doctoral fellow at the Broad Institute in Cambridge, Mass. “But this is actually data. And that really speaks to scientists in the language they understand.”