On 12 April, Hillary Rodham Clinton declared her second candidacy for president, this time as the Democratic Party favorite, supported by a large majority of Democratic-leaning voters. The next day, a scientific study declared female candidates front-runners in another career competition that many have long believed is stacked in favor of men: the race for assistant professorships on university science faculties.
Both events attracted intense media attention. The peer-reviewed article, “National hiring experiments reveal 2:1 faculty preference for women on STEM [science, technology, engineering, and mathematics] tenure track,” by psychologists Wendy Williams and Stephen Ceci of Cornell University, got so many mentions from news reporters, bloggers, Facebookers, and Tweeters that it ranks “amongst the highest ever scored” for the prestigious journal Proceedings of the National Academy of Sciences (PNAS)—ranked No. “14 out of 33,942” as of 29 April—according to altmetrics, which tracks social media mentions of scientific papers. Some of that attention has been positive—but much of it has been not only negative but also unpleasantly personal. “We’ve had such amazing attacks. They really, really hate us,” Williams tells Science Careers by phone. (See Box)
It’s tempting to blame gender when you don't get a job and you’re a woman. It’s easier … than to admit that the entire premise of what you've done for the past 7 years of your life was flawed at the root.
Clinton’s announcement was a big story partly because the prospect of a woman president threatens ancient notions of gender still cherished by many, especially conservatives. But why should a scholarly report on experiments that break little new scientific ground have such an outsized impact? It’s probably because Williams and Ceci’s findings also question strongly held ideas about gender. In this case, though, the people complaining view themselves as progressive advocates for women’s advancement.
Making sense of the controversy surrounding the paper requires clarifying what it does and doesn’t say. The paper reports on a series of experiments that presented more than 800 male and female biology, engineering, psychology, and economics faculty members at 371 universities with various versions of the credentials of fictitious applicants for assistant professorships. The identical records, which indicated outstanding achievement, were sometimes attributed to a woman and sometimes to a man. In every field but economics, where the data indicated no gender preference, the respondents strongly preferred the purportedly female candidates.
The paper does not argue that no gender bias exists in academic science. It only examines whether gender bias influences faculty members at the point of choosing among highly qualified candidates, Williams and Ceci emphasize in the interview. “No one is saying that there are no differences in the experiences of women and men in the academy,” Williams says. “We’ve never suggested or said that there are no problems at any … point” other than hiring. In fact, the pair will soon publish findings about other aspects of life as a graduate student and a new assistant professor. And there, “we do find differences,” Williams says.
The importance of being outstanding
“The [current] Williams and Ceci article presents evidence of a preference for a female candidate over a male when performance and lifestyle are identical. In other words, they do not find that employers would discriminate against women solely because of their gender,” economist Ernesto Reuben of Columbia University writes in an e-mail to Science Careers. “This is an important finding.”
In March, Reuben and co-authors published a hiring study of their own in PNAS. They found that “both male and female subjects are twice more likely to hire a man than a woman” to do “an arithmetic task” when gender is all that is known about an applicant. The male advantage diminishes, though, when the prospective employer learns that women have proven equally capable of doing the task. (Men, the authors note, “are more likely to boast about their performance, whereas women tend to underestimate it.”)
“Employers, especially [those] with strong implicit stereotypes about women and mathematics … tended not to take this [differential boasting] into account,” the authors continue. Information about performance from an “objective” source, however, strongly reduced discrimination. So, even if an employer holds implicit gender stereotypes, bias does not necessarily control hiring decisions when objective information about performance is available, Reuben’s study suggests. Williams and Ceci’s study provided objective evidence of high accomplishment and “excludes [faculty hiring] discrimination based on [gender] as a major culprit in the lack of women in STEM fields,” Reuben writes in the e-mail.
A large body of other research also finds a female advantage in faculty hiring, Ceci notes. “We looked at eight [faculty] hiring audits [examining] who applies and who gets hired and … all eight … showed that although fewer women than men apply for STEM jobs, those who do apply are hired at a higher rate,” he says in the interview. “In real-world hiring of tenure-track applicants in STEM fields in the United States and Canada … [t]here is a female advantage in all large-scale studies dating back to the 1980s,” he and Williams write in an online appendix to their article. Generally, however, that “female advantage is less than the 2-to-1 finding reported here.” And, they add, “actual hiring analyses may underestimate the strength of female preference” because they count only who took a job, not the number of offers that applicants received.
A meta-analysis of 136 studies of gender in hiring published in January in the Journal of Applied Psychology also failed to find bias when “information clearly indicated high competence of those being evaluated.” This is relevant because the applicants evaluated in the Williams and Ceci study were exceptionally competent, as are all the candidates who make it to a faculty job short list these days. Given the vast oversupply of Ph.D.s, faculty hiring has “been a buyer’s market for years,” Ceci says. A search he recently chaired, for example, attracted 267 applicants, and “any of 30 or more would have been outstanding.” “Every search yields several dozen applicants who are all extraordinary, and you don’t even know how to choose among them they’re so good,” Williams adds.
A large body of data thus converges on the conclusion that bias against women at such rarefied levels of accomplishment is extremely unlikely, they conclude. Indeed, observes Columbia’s Reuben, Williams and Ceci’s finding that women received higher ratings than men “suggests that evaluators understand that there exists discrimination against women in academia and therefore give more credit to a woman who is judged as exceptional by her peers than to a man with the same narrative.”
Women are “hired more often because a lot of us faculty have internalized the value of gender diversity, and they actually want more women when … all other things [are] equal,” Ceci explains. “Ideas about women have changed,” in academic hiring, Williams adds. “It’s just like the evolution of ideas about gay marriage. … I don’t see what’s so threatening about acknowledging that in some areas ideas have changed, and in other areas there still are problems.”
“The outpouring of vitriol [shows, however,] that there are a lot of people who don’t want to acknowledge the data,” she continues. “It’s not like they’re willing to acknowledge it and discuss it. … They try to say that we had an agenda or that the entire method at its root was completely flawed.” Many of their detractors, she adds, falsely characterize their work and its conclusions and then attack “straw men.” This opposition, she believes, is emotional rather than scholarly.
The venom may arise because some people are deeply invested in a narrative of persistent and pervasive bias against women in academic science, perhaps because of ideological beliefs in the ubiquity of gender discrimination. Some people may be attempting to salve their disappointment at a job market that keeps thousands of well-qualified people from having the careers they want and worked for. “It’s tempting to blame gender when you don't get a job and you’re a woman,” Williams says. “It’s easier … than to admit that the entire premise of what you've done for the past 7 years of your life was flawed at the root.”
In any case, as we’ve mentioned before, that women are twice as likely as men to be hired onto science faculties does not mean that many of them will actually get jobs. As Williams, Ceci, and co-authors noted in earlier work, one 13-year study of “a large state university” found that 2-to-1 female advantage resulted in the hiring of 4.28% of female applicants as opposed to 2.03% of males.
Some detractors may be protecting established careers, such as scholars and writers who built their reputations on describing discrimination that Williams and Ceci’s data suggest has faded at the point of hiring. Some may be part of universities’ substantial antibias administrative and training apparatus, which, like every part of today’s financially stressed institutions, is under pressure.
Williams says she and Ceci will soon publish data showing “some challenges for women in some areas” of academic science. “People will embrace that,” she predicts. In the meantime, people “acting as though the [current] study is worthless because it came out in a way that you personally don’t like—that’s not science,” she says. “That’s religion.” Or, as Hillary Clinton might put it, that’s gender politics.
*Attacking the messengers
Nonscholarly reactions to Williams and Ceci began with the publication of a paper and essay declaring that “female PhD applicants fare at least as well as their male counterparts in math-intensive fields.” Many of the comments clustered under the Twitter hashtag #StillaProblem. The attacks escalated with the publication of the current paper, many through the hashtag #GaslightingDuo.
As used colloquially, “gaslighting” describes a situation in which “false information is presented to the victim, making them doubt their own memory, perception and quite often, their sanity,” according to the Urban Dictionary. The term derives from the great 1944 film thriller Gaslight, in which suave, nefarious con artist Gregory Anton (played by Charles Boyer) skillfully manipulates emotion and technology to try to swindle naïve, love-struck heiress Paula (played by Ingrid Bergman) out of her fortune. In the end, a staunch and courageous admirer, Brian Cameron (played by Joseph Cotten), gathers evidence to unmask the evildoer and rescue the lady in distress. The analogy between the film and the peer-reviewed and extensively documented research appears to be intended to accuse Williams and Ceci of conscious and malicious distortion.
A recent book, Galileo's Middle Finger: Heretics, Activists, and the Search for Justice in Science, and other writings by Northwestern University historian of science and medicine Alice Dreger, examine the phenomenon of ad hominem against scientists whose data violates critics’ ideological agendas. As she notes, topics touching on sex or gender often set off the anger. “While it is true that you can also tick people off by making challenging scientific claims about alien abductions, chronic Lyme disease, or race—especially race—the surest route to having your reputation threatened is to contradict some deeply held belief about sex,” she writes.
Williams and Ceci’s ordeal does, however, resemble Gaslight in at least one encouraging respect. Like Cotten’s character in the film, who carefully gathered evidence to foil the Boyer character’s sinister plot, other researchers have volunteered data in support of the duo’s conclusion. For example, William H. Casey, professor of chemistry at the University of California (UC), Davis, and former chair of the UC Davis Committee on Academic Personnel, which, in his words, “handles all hiring, merit advancements, and promotions (tenure)” at the university, sent Williams and Ceci calculations based on university information on 5 years of hiring at UC Davis. (“WF,” by the way, stands for “workforce availability”—roughly, the representation of women in a particular field.) Casey wrote to Williams and Ceci that these figures are “perfectly consistent with the hypotheses of your last papers. For Math and Physical Sciences, for example, women were 15.2% of the applicant pool, 31.4% of the short list invited for an interview, and 26.7% of the faculty actually hired.”