Women are underrepresented in science, tech, engineering, and math (STEM), and two years ago a study offered a counterintuitive explanation as to why. The authors pointed out that countries with more gender equality, like Finland, tended to have fewer women earning degrees in those fields.
But more women studied science and tech in countries with less gender-progressive policies, such as Algeria, reported the researchers, who called this phenomenon the “gender-equality paradox” in STEM education.
The 2018 finding drew widespread attention from mainstream media outlets, like the Atlantic and Ars Technica, as well as from conservative think tank the American Enterprise Institute and Jordan Peterson, the controversial psychologist most famous for his YouTube videos addressing what he’s called the “crisis of masculinity.” Peterson and others cited the study to argue that, free from societal constraints, women choose to stay away from technical fields — a choice they make because of an innate lack of interest, not because of the patriarchy.
But outside researchers questioned that conclusion after they tried, and failed, to replicate the original study. Sarah Richardson, a science historian at Harvard University, told BuzzFeed News that the study authors used a “very selective set of data” to produce a “contrived and distorted picture of the global distribution of women in STEM achievement.”
In December 2019, a lengthy 1,113-word correction was added to the paper, clarifying how the researchers had arrived at their conclusions and correcting several sentences and misleading figures. In a separate article and series of blog posts on Tuesday, Richardson and her colleagues at Harvard’s GenderSci Lab laid out what they saw as the significant problems with the study’s methodology, including the researchers’ calculations for determining the percentage of women STEM graduates and the metrics they used to assess gender equity in each country.
And they called into question the study’s fundamental premise: that the correlation the authors apparently found between national gender equity and women in STEM means the former directly affects the latter.
“When we looked under the surface, this appears to be a case of massaging one’s data — selecting for different countries, particular gender measures, particular women-in-STEM measures — to produce the narrative that you want to see,” Richardson said.
“In the end, we do not think that there is a ‘gender-equality paradox.’”
But one of the study’s authors, Gijsbert Stoet of the University of Essex, stands behind the correlation they found and argued that it remains even when using Richardson’s preferred calculations. The problem with the critique, he said by email, “is that they cannot explain the phenomenon we reported.”
When the study came out in February 2018 in the journal Psychological Science, it provided fodder for the likes of YouTube channel Independent Man. “It’s not that women don’t have the aptitude to take STEM subjects in the more egalitarian societies,” explained one clip. “They just choose not to.”
This clip, which has been viewed more than 89,000 times, lives alongside videos like “Debunking the Black Lives Matter Narrative” and “Toxic Femininity.”
In an interview at the time, Peterson mentioned “a great paper” showing that “as societies become more egalitarian, the enrollment gap between men and women in STEM fields increases.” He added, “And what do the feminists say about that? ‘Pseudoscience.’”
And a member of the American Enterprise Institute cited it to argue that underrepresentation of women in STEM “may actually be the result of the great advances in female empowerment, progress, and advancement that have taken place in recent decades, and not the result of systematic gender discrimination.”
In the paper, a pair of psychologists — Stoet and David Geary of the University of Missouri — found that across most countries, girls are as good as boys, and often better, at math and science. But in countries with greater gender equality like Norway and Finland, women make up less than 25% of college graduates in STEM fields. In and of itself, this gender gap isn’t news. But the researchers theorized that because these countries tend to be richer, women have the financial freedom to pursue their natural interests — which drives them more toward the humanities.
In contrast, in countries with historically less gender equality, such as Algeria and Turkey, women make up much higher percentages of STEM degree-holders, according to their analysis. Because economic opportunities tend to be fewer there, those conditions “may make relatively high-paying STEM occupations more attractive” to women, Stoet and Geary wrote.
But Richardson thought a lot of these numbers seemed off. So she and a team tried to recreate the analysis with the publicly available data it was based on, including college graduation data from UNESCO.
The researchers had reported, for instance, that “the percentage of women among STEM graduates” in Algeria was 40.7%. But Richardson found that in 2015, UNESCO reported a total of 89,887 STEM graduates in Algeria, and 48,135 of them — or 53.6% — were women.
So where did 40.7% come from?
Eventually, Richardson’s team would learn that Stoet and Geary had added different sets of numbers: the percentage of STEM graduates among women (in Algeria’s case, 26.66%) and the percentage of STEM graduates among men (38.89%). That added up to a total of 65.55%. Then they divided the percent of women STEM graduates by the total, producing a rate of 40.7%.
“What they had done is create their own ratio of those two, which has never been validated or used in STEM research,” Richardson said.
That metric was not explained in the paper. In the recently issued correction, the authors went into detail on the math they’d come up with.
After Richardson and her colleagues recalculated each country’s figures, they found that overall, the study underestimated the number of women STEM graduates worldwide by about 8%.
That wasn’t the only problem. Even after Richardson’s team learned about the study authors’ method of calculating the ratio, they still couldn’t replicate all of their results. Richardson also took issue with the metric used to assess each country’s level of gender equity and the fact that the study did not examine trends over a long period of time.
Richardson’s team found that there are large variations in the gender gaps between STEM graduates among countries, no matter how they are measured. “These variations do not conform to simple patterns,” Richardson and graduate student Joseph Bruch wrote in a blog post, adding that gender inequalities are “not easily represented along a single dimension and with a single measure, as Stoet and Geary attempt to do.”
Maria Charles, a sociologist at the University of California, Santa Barbara, who studies gender and STEM education and was not involved with Richardson’s analysis, told BuzzFeed News by email, “There is no evidence that gender stereotypes and unconscious gender biases are less pronounced in advanced industrial societies — even in societies where women are well represented in universities, labor markets, and polities.”
In a letter responding to Richardson’s allegations, Stoet and Geary said they had chosen their metric to reflect a woman’s likelihood of completing a STEM degree compared to a man’s. They also said that despite the specific approach to calculations they’d taken, the overall correlation that they had found between nations’ gender-equity levels and the number of women in STEM remained the same.
Richardson said she first emailed Stoet with questions about the source of his numbers in December 2018. He replied and then stopped writing back, she said, at which point she contacted the editors at Psychological Science.
Asked whether the paper should have been retracted instead of corrected, the editors, Tim Pleskac and Steve Lindsay, said by email: “In our view, retraction is appropriate when the reported results have been convincingly shown to be fundamentally in error. In our view, the Stoet and Geary article, post-Corrigendum, was not fundamentally in error.”
Richardson said that the messiness underlying the findings reinforces that there is no one factor that determines whether women pursue or succeed in science and technology.
“Cultural patterns around women’s achievement in and preferences for STEM are incredibly complex and incredibly diverse across the globe,” she said.