Women and the Numbers: The Stereotype Threat

via Saturday Morning Breakfast Cereal

Web comics really do seem to say it best, don’t they? Following the article I wrote about the International Congress of Women Mathematicians last month, Psychological Bulletin published a new study on the so-called relationship between gender and mathematical performance. According to an in-depth look at previously published studies, the difference in scores between males and females was so close that it was deemed trivial. Though the chief author of this study, Janet Hyde, says that social scientists have long accepted this equality in mathematical ability between male and female, she does acknowledge the existence of something she calls the stereotype threat.

One reason I am still spending time on this is because parents and teachers continue to hold stereotypes that boys are better in math, and that can have a tremendous impact on individual girls who are told to stay away from engineering or the physical sciences because ‘Girls can’t do the math.’

There is lots of evidence that what we call ‘stereotype threat’ can hold women back in math. If, before a test, you imply that the women should expect to do a little worse than the men, that hurts performance. It’s a self-fulfilling prophecy. – Janet Hyde

For those unfamiliar with the term, here’s how Wikipedia defines it:

Stereotype threat is when a person, who belongs to a group that has a negative stereotype attached to it, subconsciously conforms to the negative stereotype by performing a task to a lesser degree than they would otherwise. For example: black people have the “less intelligent” stereotype attached to them, so a black person might perform poorly on an IQ test. If said person was either unaware of the stereotype or knew the stereotype to be wrong (stereotype threat is not present) then they would perform better.

Of course, there are different kinds of stereotype threats (e.g. where an individual is judged based on a negative stereotype) but today we will look at the threat which is subconsciously causing individuals to perform lesser and what is or can be done to combat this. But first of all, it is well worth showing that this stereotype threat is more than just a sociological idea and something that science has actually taken a good deal of interest in.

David Marx and Jasmin Roman wrote a paper in 2002 which involved several experiments. In one of the experiments, they found that women performed as well as men when they were tested by a female administrator. However, when a male administrator presented the test, the women did not fare as well as the men. Another study performed by Steven Spencer, Claude Steele and Diane Quinn examined whether changing the way the test was presented affected performance. In this experiment, two tests were given. In one, the participants were explicitly told that the test they were taking had shown gender differences in the past while on the other test, participants were told that the test had never shown gender differences before. As in Marx and Roman’s study, the women taking the latter test found higher scores than the women taking the former test.

Now, I know that there are probably sampling errors and contingencies that prevent this research from becoming generalized but I do think that these researchers have at least proved the stereotype threat to cause a substantial difference in performance. In Steele and Quinn’s conclusion, they highlighted the fact that just because boys and girls are taught in the same environment, doesn’t mean they are taught the same way. Though they have successfully managed to void the idea that any weaknesses women have at math comes from internal ability, they are unable to offer suggestions of how one can overcome the stereotype threat (understandably).

So what is the broader applicability of these experiments? And where can we go from here?

As I discussed last time, there aren’t a whole lot of female mathematicians for girls to look up to. One of the main focuses of the ICWM was to raise awareness on the achievements that women have obtained within the math field. While it may seem difficult for anyone to name either a famous male or female mathematician (or an achievement), I’m sure there are many who have walked into a math lecture or conference and discovered what the dominant sex was. Thus, the point remains. This perception of dominance batters away at self-confidence from an early age (Re: the Bill-Nye the Science Guy’s or Stephen Hawking’s or Carl Sagan’s that invade our young ideas of what a career in math or science might look like).

Besides having more role models for girls to look up to, Reducing Stereotype Threat offers some great immediate suggestions on how to reduce stereotype threat:

-       reframing the question/test to assure all participants that the test is sex-fair. I’m not sure if it’s possible to remove the male/female checkbox from tests, but having this sex-free disclaimer is similar to Steele and Quinn’s experiment

-       providing external factors that contribute to the test’s difficulty (e.g. students being told that the transition to middle school is difficult and any anxiety felt is normal)

-       effective feedback: communicating high standards but also assuring the student that they are capable of meeting the high standards.

RST has a few more examples of ways to reduce stereotype threat and beyond this site, I’m sure there are plenty more out there. I think what’s holding back this negation in stereotype threat is the fear that asking for solutions like having more women role models doesn’t sound very immediate. While that is certainly a long-term solution, I hope I have illustrated that direct ways to attack stereotype threat do exist, but as the SMBC comic implies, the younger you start, the better.

Are you new to Gender Across Borders? Please read this first. We may update the site, and you can stay in contact with us through our Twitter feed and our newsletter. Like Gender Across Borders on Facebook. Follow us on Twitter and Tumblr. Subscribe to our monthly newsletter.