In your mind, not your genes

Women do well in math if they’re told women have as many math genes as men. If they’re told there’s a genetic difference in math ability linked to gender, women’s math scores fall sharply.

In a Canadian experiment on “stereotype threat,” women were tested on math, asked to read an essay and given a second math test. Researchers wanted to see if the reasoning behind the stereotype would affect results. They divided college-age women into four groups.

Each took a three-part test, two math sections separated by a reading comprehension essay. The first essay argued that sex differences in math were due to genetic differences between men and women. The second cited a person’s experiences as the main factor. A third essay talked about gender differences but did not discuss mathematical ability. The final essay stated that there were no differences in mathematical ability between men and women.

Women given the essay focusing on genetic factors performed the worst of the four groups. Those focusing on experience did significantly better, and their performance was as good as the group that was told that gender differences do not exist, the team reports in tomorrow’s issue of Science.

Differences were significant. Low scorers got five to 10 questions right on average; high scores got 15 to 20 questions right.

About Joanne

Comments

  1. Whenever I see the words, “…if they’re told…” in a summary of the study, I wonder about one thing: “Why would they believe you?”

    Is there some special place where one can find a supply of trusting, gullible human lab subjects?

  2. I am really suspicious of the study. Looking at the way the study was designed, the meaningful result is the difference of test scores between the two tests. Yet we only heard about the average test score. I don’t know whether it is the fault of the paper or the reporter.

    Stereotyping can affect women’s interest, confidence and drive to learn math. That has to be a long term effect. I cannot imagine that women on reading an essay of genetic difference in mathematical ability, in half an hour’s time, got brainwashed and forget all about quadratic equation and other mathematical concepts, and do poorly in a math test. The more likely explanation is that she read the essay, became emotionally upset, could not concentrate and did poorly on the test.

  3. Richard Nieporent says:

    These same researchers tried a slightly different expeiment before and also got positive results. The summary of that experiment is presented below.

    Woman: I could while away the hours/conferrin’ with the flowers/consultin’ with the rain/And my head I’d be scratchin’/ While my thoughts were busy hatchin’/If I only had a brain.

    Researcher: Why, anybody can have a brain. That’s a very mediocre commodity. Every pusillanimous creature that crawls on the Earth or slinks through slimy seas has a brain. Back where I come from, we have universities, seats of great learning, where men go to become great thinkers. And when they come out, they think deep thoughts and with no more brains than you have. But they have one thing you haven’t got: a diploma. Therefore, by virtue of the authority vested in me by the Universitartus Committiartum E Pluribus Unum, I hereby confer upon you the honorary degree of ThD.

    Woman: ThD?

    Researcher: That’s… Doctor of Thinkology.

    Woman: The sum of the square roots of any two sides of an isosceles triangle is equal to the square root of the remaining side. Oh joy! Rapture! I got a brain!

  4. Cardinal Fang says:

    markm– Before making assumptions about the study, click the link.

    The women were given a three part test. The first part was a math test. The second part was a reading comprehension test with an essay asserting that women were worse at math than men, or that women and men were equally good at math, or that one’s math ability depends on one’s experiences, or nothing at all about math ability. The third part was a second math test.

    The ones who read the essay saying that women were genetically bad at math did poorly in in the second math test.

    This comports with similar research about blacks and intellectual ability: blacks who were reminded of their blackness did worse on tests than blacks that weren’t.

    It’s a strange result, but seemingly well replicated.

  5. At the risk of being really impolitic, what if there is a genetic difference? To me that represents one of the underlying problems here. I will admit a certain skepticism when I linked to the article and saw that they had linked Larry Summmer’s speech. That certainly didn’t inspire confidence that this was purely about science. But I have yet to see a study that does not reflect a pronounced overrepresentation of males at both extreme ends of the spectrum when looking at math scores. The problem with so many of these arguments is that the entry point requires a denial of much of the existing science rather than the formulation of a logical explanation for it.

  6. The study didn’t address the question of whether men are better at math than women. The issue was whether women’s scores would change if exposed to different ideas: women are innately inferior, women are equally able, etc.

  7. what if there is a genetic difference?

    If it is true *and* it becomes “accepted wisdom”, we can expect to see female’s representation in math fall far below what actual ability would predict as girls further internalize the “girl’s can’t do math” meme and establishment support for girls doing math falls as the institutions internalize “girl’s can’t do math” meme and fail to support those exceptional individuals by doubting their existence.

    If it is true *and* but the expression of the concept remains controversial, we can expect to see female representation roughly as ability would predict (or slightly below because of partial acceptance of the meme), and occasional (partially inaccurate) attributions of lower levels of representation to systemic discrimination.

    Between the two, I know which one I’d choose…

  8. I doubt that those are our only two options. I think that the consequences of lying to people for their own good are far worse than the effort required to make sure they actually understand something that isn’t that complex.

  9. The comparable study on blacks (actually on females as well) was done by Claude Steele. The conclusion is that when the subject worries about doing poorly in a test would lend support to stereotypes about their ability, the anxiety would interfere with his/her performance and resulted in poor performance. This is like a baseball player who has a reputation of choking in the post-season games, the added pressure would indeed make him fail to hit during the playoff. That is why I said what we are seeing in this study is the effect of emotion on the test performance.

    Every time we see a female successful in math or science, we should celebrate her individual achievement. Those who loudly proclaim that it is a blow for gender equality and a refutation of gender difference are not doing the female student a favor. They are putting unnecessary pressure on the individual, and that may end up in failure in future. This is the lesson we should learn from the study.

    I wish Michelle Wie would wait a few more years before trying to compete in men’s golf.

  10. Cardinal Fang says:

    It’s not a question of lying to students. It’s a matter of how the information is presented. Suppose that males are, on average, better at math than females.

    We can say to students, “Boys are better than girls at math.” Or we can say, “The best predictor of your math performance is how hard you work at math. Students who practice math are much better than students who don’t study.” Both of those things are true, but one of them offers an easy excuse and the other doesn’t.

    I don’t want to offer teachers or students excuses for girls’ poor performance. I don’t want a girl to think, “I’m a girl so I can’t do this.” I don’t want a teacher to think, “She’s a girl, no sense expecting her to be able to do this.”

  11. Cardinal: I think Richard’s hypothesis explains such experimental results equally well – and it’s far easier to believe.

  12. I doubt that those are our only two options.

    I’d really like that to be the case, but a lifetime of experience tells me that my two cases will sum up the vast majority of human experience. It’s like yelling “fire” in a theatre: It shouldn’t have to mean that people die, but 500,000 years of human evolution ensures that they will.

    And it’s not a matter of lying. It’s simply a matter of what information we choose to emphasize, publicize, fund, etc.

    I simply don’t see the point in funding and publicizing something that can have no positive impact on society.

    I’m also not encouraged by the fact that every indication I see is that most of the people who seem to be welcoming these studies intend to use them to dismiss any and all complaints about systemic discrimination.

    And I’ll say while it seems (in my opinion) to be unlikely that all the disparity of female representation in the sciences is due to systemic discrimination, it’s also a certainty that some of it is.

    I’d like to see what “positives” people will see out of these studies other than “finally we’ll have something to shut these stupid feminists up”. Anyone care to enumerate any?

  13. Joanne –
    I agree that the study did not address either the math ability of women on a normative scale, nor did it perform a comparative analysis between genters, but you are kidding yourself if you don’t think their point is that there are no genetic differences. Just read the first paragraph of the article – for all intents and purposes it says that raising gender difference issues should be off limits because that alone can negatively affect women’s performance. This is science? I thought the point was to ask the uncomfortable questions that force us outside our comfort zones and end up with us searching for and hopefully discovering the truth. It seems to me that in proposing that this stereotype adversely affects women’s performance in math they are reinforcing another, perhaps more harmful stereotype – that women are weak and need to be protected. It is worth noting that while they linked to Summer’s speech that they seriously mischaracterized the content and the context in the article as a means of butressing the relevancy of this study. He was not painting with a broad brush but was discussing those individuals more than 3 sigmas above the mean. He was talking about that group that is less than one half of one percent of the relevant population, and he was very specific about that. A better question to ask might be how a hypothesis about such a specific, very small group became a generalized argument regarding genetic differences across the entire population.

    I agree with bd that we need to celebrate individual achievments and not always try to use them to make some broader point about society. Any classroom teacher who says boys are better than girls in math is way out of line. But in a research environment, it should be acceptable to propose and test that hypothesis without being decried as a misogynist. My perception is that these authors would argue otherwise.

  14. Cardinal Fang says:

    I think some commenters are missing the point of the study. It had nothing to say about the mathematical ability of men. They didn’t compare women to men, because they didn’t test any men.

    Instead, the test results are about how women sometimes perform worse in math, and sometimes perform better– not better or worse than men, but better or worse than they otherwise did. In fact, I didn’t see anything in the linked article about how the women’s performance compared to male norms at all.

    Those of us interested in education want to know how to make students, including female students, perform better. Maybe this study can help us figure out a way to make girls and women perform better in math.

  15. But in a research environment, it should be acceptable to propose and test that hypothesis without being decried as a misogynist. My perception is that these authors would argue otherwise.

    Let me make a clearer analogy. If I’m a research medical scientist, and I spend considerable time and effort trying to discover if it is possible to modify crippling diseases to only affect white people, are you going to question my motives?

    It’s certainly an interesting endeavor from a pure biological science perspective, but I think most of us would question research into an area that can (at best) produce results that diminish general well-being. Well, the same (writ small), applies to the mental differences research.

    The reason people doing the research are being decried as misogynist is because about the only use these researches can be put is to further harm female achievement in math/sciences.

    More to the point, unless the researchers are impossibly dim, they’ll know that the results of research saying that women are underrepresented at high/low end will be translated into the general psyche as:
    girls/women can’t do math/science
    any girl who really is at the high end is probably not really qualified to be there.
    any and all under-representation in math/science faculties is automatically due to ability differences. None is due to discrimination/work-place issues.

    I’m *still waiting* for someone to come up with a positive use for the research.

    (And yes, I apologize to Cardinal Fang for hijacking the topic. My slant when reading the original post is that it was further proof that
    comments like Larry Summers *do* damage female achievement in math/sciences (and if he’s not smart enough to realize this, he’s no business being president of a University.) the societal benefits of mental sex/race difference research range from none (if no difference is found) to profoundly negative.)

  16. Richard Nieporent says:

    Interesting Tom. What you are saying is that for the “good” of society we are supposed to believe that everyone is equal in ability. Thus, when a difference appears in performance, such as more male scientists or mathematicians we are to take it as prima facie evidence that there is discrimination. Of course then we must rectify the situation by instituting affirmative action and preferentially hiring more women.

    If you are going to argue that the lack of female scientists is really due to discrimination, how are we going to know if you do not allow any research from being carried out on differences of ability? In other words it is perfectly okay to discriminate against men to prevent your hypothesized discrimination against women.

  17. Yes, given how human beings are generally built, the two choices (*if* we accept the science as truth), are as I enumerated above:

    If it is true *and* it becomes “accepted wisdom”, we can expect to see female’s representation in math fall far below what actual ability would predict as girls further internalize the “girl’s can’t do math” meme and establishment support for girls doing math falls as the institutions internalize “girl’s can’t do math” meme and fail to support those exceptional individuals by doubting their existence.

    If it is true *and* but the expression of the concept remains controversial, we can expect to see female representation roughly as ability would predict (or slightly below because of partial acceptance of the meme), and occasional (partially inaccurate) attributions of lower levels of representation to systemic discrimination.

    Both options have cost. But yes, I think the cost of the second option is a *lot* less than the cost of the second option.

    And remember we’re not talking legal action on those who want to espouse these ideas publicly. I simply believe that it doesn’t make sense to publicly support the research or presentation.

    And to be honest, I *do* believe that our society is based on the fundamental concept of equality. If that belief gets substantially eroded I fully expect the see “scientific” racism and sexism calling (for example) for policies that ban immigration from certain racial groups because they are inherently more criminal, less intellectually capable, etc. backed up by misinterpreted scientific results.

    Every society has its myths that bind it together. America’s myths I consider (for the most part) to have been beneficial for build up of its civil society. No, I don’t see eroding them to be beneficial.

  18. Richard Nieporent says:

    And to be honest, I *do* believe that our society is based on the fundamental concept of equality.

    It is equality of opportunity not equality of results. If we accept your premise then we must set quotas for every job so that all groups are equally represented. You do know that there are “too many” Asians admitted to college and for that matter too many women. Would you be willing to force universities to admit equal number of men and women and if not why not? The problem with forced equality is that it tears at the fabric of society. If discrimination is wrong when it is practiced against women and minorities they it is wrong when it is practiced against White men. Or do you believe that there is something inherently evil about White men that they deserve to be discriminated against?

  19. Tom is against research about “women are underrepresented at high/low end” and thinks “the societal benefits of mental sex/race difference research range from none (if no difference is found) to profoundly negative”.

    I am not interested in gender difference per se. Most of what I know about gender difference came from reading Baron-Cohen. I read Baron-Cohen because I am interested in autism. Baron-Cohen’s research interest is in autism. Autism is over represented by males, and it is a high/low end issue. That is why Baron-Cohen’s research in autism leads to research in gender difference. According to him, to understand autism, you have to understand gender difference. In an interview he said “It’s time to distinguish politics and science, and just look at the evidence.”

    So do I think that gender difference and over representation at high/low end research should be stopped for the sake of political correctness? As a father of a child with Asperger syndrome, I certainly don’t think so.

  20. Very interesting. That’s the first time someone has come up with what I consider some beneficial reasons for research in this domain. I’ll have to say that the tradeoffs for me are now no longer as clear cut. Sadly, by the time I’ve come to a formulated a new position to take this into account, this topic will likely have scrolled way off. Such is the transient nature of blogs. Luckily this topic keeps coming up. (Is that 🙂 or :-(…)

    As the father of two children with ASD (one Asperger’s, one PDD), I found the papers quite interesting. I’ve followed his work only peripherally until you mentioned him. Four papers later, I’m finding much of his stuff fascinating.

  21. Woman: The sum of the square roots of any two sides of an isosceles triangle is equal to the square root of the remaining side. Oh joy! Rapture! I got a brain!

    The more likely explanation is that she read the essay, became emotionally upset, could not concentrate and did poorly on the test.

    Ugh. Exactly the sort of misogynist stereotypes that women in the sciences have to confront.

  22. Reginleif thinks that “emotion cause underperformance” is misogynist. Obviously he/she does not understand what “stereotype threat” means. Let me quote from the authority on “stereotype threat”, Claude Steele: “Steele then talked about his findings in many studies that when a person’s social identity is attached to a negative stereotype, that person will tend to underperform in a manner consistent with the stereotype. He attributes the underperformance to a person’s anxiety that he or she will conform to the negative stereotype. The anxiety manifests itself in various ways, including distraction and increased body temperature, all of which diminish performance level.”

  23. John from OK says:

    I find it quiet plausible that some of these women became emotionally upset after reading the essay, especially if they had been taught for 4 years that they should become enraged at any form of “hate speech.” And I’m being serious here. A similar study performed on men, whereby a politically incorrect essay was presented between the tests, might yield similar results.

    It seems this test could be a good screener for potential employees: recruite those whose performance does not drop after reading a single essay.

    Finally, the idea that women are more sensitive than men is not an ignorant stereotype, except that only feminists are allowed to say it.

  24. Richard Nieporent says:

    Ugh. Exactly the sort of misogynist stereotypes that women in the sciences have to confront.

    Reginleif, get a life. I was making fun of the study. Clearly humor is lost on you.

  25. The experiment is missing a different obvious control to test bd’s – very likely – hypothesis. The control would be a different essay that said nothing about math but did insult women.

  26. Reginleif says:

    Reginleif, get a life. I was making fun of the study. Clearly humor is lost on you.

    Defensive much, Richard?

  27. I am not sure about this particular study, but I can vouch for myself. As a female in grade school I did very well in math. When I got to high school, I started having major problems in math. Everyone was telling me that I just didn’t have the genetic aptitude for higher math. So I concentrated on what I was good at and lost alot of interest in math. Alot of it was because I was female. However, I loved science which meant taking alot of math in college. I taught myself math and love math now. I totally turned myself around in math. What I did not realize when I was a kid was just because something is hard at first does not mean that one cannot learn it. So expectations are important. I talked to alot of my guy peers in science about this too and I discovered that when they struggled, they kept on going because they had the expectation that they could learn it. Too many females that I personally know were conditioned to give up on learning math. I have helped so many of my female friends to not fear math and to finally really learn it, even as adults. I hope times have changed. I am 36 and most of the similiar stories I hear about females being convinced that they lacked the aptitude for math are my age or older. I think there is something to the idea behind the study.

  28. “to unite, integrate, and collectivize”

    This is Commie-speak

  29. Kind of neat research.

    But results are supported by other work. Stereotypes DO matter when people believe them. Some of these women may have believed the stereotype, or become upset by them.

    I expect men would produce the same results.

    But for those who say such studies should be “off limits’ because “they can’t see how it will help” do not understand basic research. It may have some positive use.

    For these people, truth only matters when it supports THEIR STEREOTYPE – THAT WOMEN AND MEN HAVE EQUAL ABILITY IN MATH AND IT SHOULD NOT BE QUESTIONED.