Analysis finds gender bias in peer-reviewer picks
Both men and women ask too few women to review scientific papers
Scientists often pride themselves on avoiding bias. But when asked to pick scientific reviewers, scientists tend to suggest too few women, a new study shows.
racorn/Shutterstock
Gender bias works in subtle ways, even in the scientific process. The latest illustration of that: Scientists recommend women less often than men as reviewers for scientific papers, a new analysis shows. That seemingly minor oversight is yet another missed opportunity for women that might end up having an impact on hiring, promotions and more.
Peer review is one of the bricks in the foundation supporting science. A researcher’s results don’t get published in a journal until they successfully pass through a gauntlet of scientific peers, who scrutinize the paper for faulty findings, gaps in logic or less-than-meticulous methods. The scientist submitting the paper gets to suggest names for those potential reviewers. Scientific journal editors may contact some of the recommended scientists, and then reach out to a few more.
But peer review isn’t just about the paper (and scientist) being examined. Being the one doing the reviewing “has a number of really positive benefits,” says Brooks Hanson, an earth scientist and director of publications at the American Geophysical Union in Washington, D.C. “You read papers differently as a reviewer than you do as a reader or author. You look at issues differently. It’s a learning experience in how to write papers and how to present research.”
Serving as a peer reviewer can also be a networking tool for scientific collaborations, as reviewers seek out authors whose work they admired. And of course, scientists put the journals they review for on their resumes when they apply for faculty positions, research grants and awards.
But Hanson didn’t really think of looking at who got invited to be peer reviewers until Marcia McNutt, a geophysicist and then editor in chief of Science, stopped by his office. McNutt — now president of the National Academy of Sciences in Washington, D.C. — was organizing a conference on gender bias, and asked if Hanson might be able to get any data. The American Geophysical Union is a membership organization for scientists in Earth and space sciences, and it publishes 20 scientific journals. If Hanson could associate the gender and age of the members with the papers they authored and edited, he might be able to see if there was gender bias in who got picked for peer review.
“We were skeptical at first,” he says. Hanson knew AGU had data for who reviewed and submitted papers to the journals, as well as data for AGU members. But he was unsure how well the two datasets would match up. To find out, he turned to Jory Lerback, who was then a data analyst with AGU. Merging the datasets of all the scientific articles submitted to AGU journals from 2012 to 2015 gave Lerback and Hanson a total of 24,368 paper authors and 62,552 peer reviews performed by 14,919 reviewers. For those papers, they had 97,083 author-named reviewer suggestions, and 118,873 editor-named reviewer requests.
While women were authors less often and submitted fewer papers, their papers were accepted 61 percent of the time. Male authors had papers accepted 57 percent of the time. “We were surprised by that,” Hanson says. While one reviewer of Hanson and Lerback’s analysis (a peer-reviewer on a paper about peer review surely must feel the irony) suggested that women might be getting higher acceptance rates due to reverse discrimination, Hanson disagrees. He wonders if “[women] are putting more care into the submission of the papers,” he says. Men might be more inclined to take more risks, possibly submitting papers to a journal that is beyond their work’s reach. Women, he says, might be more cautious.
When suggesting reviewers, male authors suggested female reviewers only 15 percent of the time, and male editors suggested women just 17 percent of the time. Female authors suggested women reviewers 21 percent of the time, and female editors suggested them 22 percent of the time. The result? Women made up only 20 percent of reviewers, even though 28 percent of AGU members are women and 27 percent of the published first authors in the study were women.
The Earth and space sciences have historically had far fewer women than men, but the numbers are improving for younger scientists. “We thought maybe people were just asking older people to review papers,” says Lerback. But when Lerback and Hanson broke out the results by age, they saw the gender discrepancy knew no age bracket. “It makes us more confident that this is a real issue and not just an age-related phenomenon,” Lerback says. She and Hanson report their analysis in a comment piece in the January 26 Nature.
“It’s incredibly compelling data and one of those ‘why hasn’t someone done this already’ studies,” says Jessi Smith, a social psychologist at Montana State University in Bozeman. While it’s disheartening to see yet another area of bias against women, she notes, “this is good that people are asking these questions and taking a hard look. We can’t solve it if we don’t know about it.”
The time when authors are asked to suggest reviewers might play a role, Smith notes. When scientific papers are submitted online, the field for “suggested reviewers” is often last. “That question for ‘who do you want’ always surprises me,” admits Smith. It’s a very small thing, but she notes that when people are tired and in a hurry, “that’s when these biases take place.”
Knowing that the bias exists is part of the battle, Hanson says, but it’s also time for things to change. He’s pushing for more diversity in the journals’ editorial pool — for gender, nationality and underrepresented minorities. It’s also helpful to remember that suggesting people for peer review isn’t just about the author’s work. “Even small biases, if they happen repeatedly, can have career-related effects,” he says.
Peer-reviewer selection is something that hasn’t gotten a lot of scientific attention. “I think getting a dialog going for everyone to be aware of opportunity gaps, [asking] ‘what can I do to make sure I’m not contributing to this’… I think that will be important to keep editors and authors aware and accountable,” Lerbeck says. She’s now experiencing the peer-review life first hand: She’s now a graduate student at the University of Utah in Salt Lake City. While there, she’s starting a group to discuss bias — to make sure that the conversations keep going.
Follow me on Twitter: @scicurious
Editor’s note: This story was updated February 8, 2017, to clarify Lerback’s role in the research.
