Researchers Find Flaws in Popular Theory on Women’s Math Performance

This post contributed by Celia Smith, ESA Education Programs Coordinator


In science, neat and tidy explanations rarely tell the whole story, and that is exactly what researchers at the University of Missouri have found about stereotype threat theory in their paper on the subject, currently in press at the Review of General Psychology.

Though it may sound like psychological jargon, stereotype threat is a popular theory with policymakers and the media and is also expressed more idiomatically as the ‘self-fulfilling prophecy.’ Since the theory was first described in a 1999 a Journal of Experimental Social Psychology paper, one of its most popular applications has been to explain why women have lower rates of achievement in science, technology, engineering and mathematics (STEM) than men. Supposedly, girls grow up believing that boys are better at math, and belief in this stereotype hampers their performance in math and related fields of science.

Stereotype threat has been widely accepted as a simple and intuitive explanation for the relative lack of high-achieving women in STEM that places blame on social biases rather than flaws in the education system or academia. However, University of Missouri psychology professor David Geary and Gijsbert Stoet of the University of Leeds found that many replications of the original stereotype threat study contained serious flaws in statistical analyses and scientific methodology. Some studies even lacked a control group, meaning they did not compare the experimental effects of stereotyping on women with those on men.

In addition to exposing serious holes in a popular theory, Geary and Stoet’s research highlights a common challenge in problem solving: asking the right questions. U.S. Department of Labor Statistics data show that in 2009, women comprised 29 percent of all environmental scientists and geoscientists, 25 percent of all computer scientists and mathematicians, and just 7 percent of mechanical engineers, indicating a ‘gender gap’ in STEM fields. However, data from the National Center for Education Statistics also show that girls and boys generally leave high school equally well-prepared to study STEM at higher levels. In fact, from 1990-2005, girls earned consistently higher grade point averages than boys in all math and science subjects combined.

These statistics suggest that it is important to distinguish between grades and career success when measuring ‘achievement’ in math and science. The authors of the original stereotype threat study tried to explain the gender gap by suggesting that women performed more poorly on difficult math tests when introduced to a negative stereotype; Geary and Stoet’s research disputes this theory, yet the fact remains that there are still proportionately fewer women with established careers in many STEM fields today. In other words, stereotype threat may have an insignificant effect on women’s higher math performance, but biased social expectations may still influence their career choices.

If this is the case, we may need to use some of the resources currently devoted to reducing stereotype threat to study the personal, financial, and professional incentives and disincentives for women in STEM fields, and identify what is discouraging women from pursuing these careers further despite having the skills to do so (an approach exemplified in this National Research Council report from 2010). Interventions in these areas, either at the collegiate level or in the workplace, are unlikely to be simple or straightforward…but that may be a good indication that we are getting to the bottom of a pervasive societal inequality.

Watch a video of author Gijsbert Stoet of the University of Leeds explaining his latest research on stereotype threat:


Author: Nadine Lymn

ESA Director of Public Affairs

Share This Post On


  1. While it’s fine to highlight research that purports (I’m trying to be generous here) to scientifically dissect hot button debates like “women are biologically bad at math,” this video is pretty terrible one-sided propaganda. Shame on ESA for posting it.

  2. I have a problem with the methodology of citing rare awards as a measure of general success within a field. (:

    But seriously, Stoet and Geary’s challenge to stereotype threat and the methodology of other researchers may be entirely legit–it’s just hard to evaluate or discuss since the paper hasn’t been published yet. There are other interesting ideas about the gender gap hitting the popular press. I’ve seen different types of cultural influences mentioned (and sometimes blended into the concept of stereotype threat) in coverage of a couple big science stories this year. So it isn’t clear to me that an over-emphasis on stereotype threat is a big problem (I’m willing to entertain it, but I’m not persuaded by the video).

    Geary’s previous work and his popular books on evolutionary psychology give me the impression that he’s sympathetic to the Laurence Summers position. The biological hypothesis makes women scientists anxious because of a history of popular belief in women’s intellectual inferiority, and a sense that the general public still leans toward that belief, at least for mathy fields. And yeah, I can’t approach that topic with anything like pure “scientific objectivity,” but I don’t think that anyone can.

  3. If this paper turns out to be legit after being published and vetted, this would be an interesting finding and go to show that more data is needed to be collected on the topic.

    The general public is, naturally, eager for clear research findings with straight answers for problems, and researchers may find themselves in the position to be pressured to provide such solutions. This is a problem with the publication process. Journals are pressured to publish “sexy” (excitingly appealing) articles. Researchers, in turn, are pressured to couch their findings in a “sexy story” that will get them published. Although I agree that published research findings should be couched in the authors’ interpretation of their significance, the sad truth is that the pressure to “publish or perish” can unfortunately lead to loss of objectivity. I am not making claims that the authors of this new paper or the previous papers have lost their objectivity.

    Re: Patricia Logan-Greene’s comment: I don’t see how the video is “terribly one-sided” if the message is, “hey wait a minute, we may not have the answer to this question yet, don’t stop looking.” Do you expect them to devote equal time to the opposite of their point (“the answer was found”)? Perhaps after this has been in press, they can link to a rebuttal video.

    Also, I thought science is about being open-minded to changing your beliefs with the gaining of new information. The authors mentioned in the video provide a new perspective (based on findings) on an important issue. Their message is no more (but possibly less) propaganda than the vast majority of research publications on any topic.

  4. Seriously? This video doesn’t strike you as outlandishly stupid when it says, “But when it comes to exceptional achievement, men clearly outnumber women. For example, the Fields Medal, one of the top awards for achievements in mathematics, has only been awarded to men.”

    I honestly don’t know how anyone can argue on behalf of a video that includes something like that. Especially when it goes on to say “Only science can help us find out why is the case!” The answer to the real question in there (why aren’t females receiving these tops awards) is so screamingly obvious that only a man hellbent on maintaining male domination in science, mathematics, and awards determination would pose it.

    Ahem. Perhaps that last sentence was a bit over the line. But again: seriously?!?!

    And, as to your last point, this particular science is NOT about being open to changing one’s beliefs. This is about finding ways to reinforce and/or validate beliefs that science has shown to be false and sexist (racist, elitist, whatever). Since their article still hasn’t been released, I can’t speak to its quality; however, I feel wholly confident after watching this travesty of a “scientific” video that these folks are light years from objectivity in their research.

    Perhaps they just like to be out of step with the zeitgeist in order to garner attention. And by zeitgeist I mean the inexorable decline of male hegemony of STEM.

    But we at least agree that it’s propaganda!

  5. I think YouTube says that the paper has been published by now.

    Patricia: If you think the answer is so “screamingly obvious”, it is unlikely that researchers would seriously debate this issue, unless you believe that there is a mass conspiracy against all women.

  6. >> Patricia writes “But we at least agree that it’s propaganda!”

    Nobody agrees on that, except Patricia. Patricia’s only argument is that the paper does not subscribe to her point of view, and therefore must be propaganda. That is an empty claim that reeks of throwing with mud.

    Science is aimed to be objective, clear, and properly conducted. The Stoet/Geary paper attacks a popular theory using logic and reasoning (not subjective arguments) and rejects the stereotype treat explanation of the gender gap in mathematics. The Fields Medal is probably just one example of the enormity of the underrepresentation of women. Ultimately, we need to explain of why this is, or don’t we??? If you think it is because of discrimination of women, you need to come up with some arguments. I think nobody really buys that in today’s day and age.

  7. a part of the article claimed “girls earned consistently higher grade point averages than boys in all math and science subjects combined.”, this is because girls memorize…instead of visualizing the problem.

  8. Sam, are you arguing that “visualizing the problem” is a bad strategy or a superior one?

    One of the problems with the typical reporting of statistical differences between groups of people is the tendency to declarations like yours: that everyone within a group is defined by that difference.

    Regardless, you don’t present any evidence for your assertion, and you’re perilously close to troll territory.

  9. Given that many animals have number sense, and that humans are supposed to be masters of symbolic manipulations and language, the only rational conclusion I cam come to is that the difficulty of math literacy in our culture is a systemic problem in education.

    The differences in the math achievement and subsequent science literacy between males and females is so small when compared to the results; the scientific ignorance of the general public, displayed on the subjects of climate change and evolution.

    All this shows is that sometimes scientists, constrained by their beliefs in their ivory towers, can become totally disconnected from reality.

Submit a Comment