Monday, 1 July 2013

Women's true maths skills unlocked by pretending to be someone else

There's an unfounded gender stereotype that says women aren't as good at maths as men. Reminding them of this prior to a maths task usually undermines their performance - just one example of a harmful phenomenon known as stereotype threat.

Research finds the threat comes in two flavours. Women can fear their poor performance will be used to bolster the "women are weak at maths" gender stereotype (known as "group-reputation threat"). Or they can fear that their poor performance will be taken as proof that they conform to the stereotype ("self-reputation threat"). Both can undermine women's ability to fulfil their true potential.

A new study shows a simple way to alleviate the self-reputational aspect of stereotype threat. Shen Zhang and her team tested 110 women and 72 men (all were undergrads) on 30 multiple-choice maths questions. To ramp up the stereotype threat, the participants were told that men usually outperform women on maths performance. Crucially, some of the participants completed the test after writing their own name at the top of the test paper, whereas the others completed the test under one of four aliases (Jacob Tyler, Scott Lyons, Jessica Peterson, or Kaitlyn Woods). For the latter group, the alias was pre-printed on the first test page, and the participants wrote it on the top of the remainder.

Overall, men outperformed women on the maths task. But women who took the test under someone else's name, be it male or female, performed better than women who performed under their own name, and they did just as well as the men. The effect was stronger for women who cared more about maths.

By separating their performance from their own identity, it seems the women performing under an alias no longer felt pressure to avoid being seen as an example of the harmful gender stereotype. Further analysis showed this had to do with feeling less distracted during the task and with experiencing less self-repuational threat. In contrast, male performance was unaffected by using another person's name.

Zhang and her team believe their findings have real-life applications for helping reduce the harm that comes from stereotype threat. "At the most practical level, they speak to the benefits of using non-name identification procedures in testing," the researchers said. "But more generally, they suggest that coping strategies that allow stigmatised individuals to disconnect their self from a threatening situation can be an effective tool to disarm negative stereotypes."


Shen Zhang, Toni Schmader, and William M. Hall (2013). L'eggo My Ego: Reducing the Gender Gap in Math by Unlinking the Self from Performance. Self and Identity DOI: 10.1080/15298868.2012.687012

--Further reading--
Simple psychological intervention boosts school performance of ethnic minority students
Female political role models have an empowering effect on women.

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.


ben said...

"Overall, men outperformed women on the maths task" -- which kind of confirms the "unfounded [sic] gender stereotype that says women aren't as good at maths as men".
Lolz, I'm sure some epicycle can be invoked to explain it all away...

Artemisynn said...

I'll help you, in this study it is expected that men would overall perform better because NONE of the male group had stereotyped threats (that is, men either performed under their own name (no threat) or another identification which they found produced no threat.

On the other hand, the female group who used their own identity did experienced stereotype threat and only some, who used other identities, did not.

TL;DR NONE of the men had stereotype threat, SOME of the women had stereotype threat. Stereotype threat is basically acting as handicap for some of the female group which brought the total female score down.

So the key part is really, when that stereotype threat is diminished, how do the women perform? Like 'men' who don't have a stereotype threat to begin with.

ben said...

Thanks, that response was precisely as convoluted as I had predicted. Again, this blog post claims that there is an unfounded stereotype that suggests that "women aren't as good at maths as men". Yet the study referred to in this blog post finds that, in fact, "men outperformed women on the maths task". Ergo, the stereotype that "women aren't as good at maths as men" is not "unfounded", but is instead correct. Whether some women may or may not gain a few points as a result of removing the stereotype threat is not relevant to the fact that, on the whole, "women aren't as good at maths as men".

Unknown said...

Ben - there's lots of research to suggest that women are just as good at men at maths. In this study, before they took the test, the participants were fed the myth that men are better at maths, which would have put the women at immediate disadvantage. It's therefore not justified for you to refer to the overall male superiority in this study as evidence that men are in fact better at maths.

Chris Wheeler said...

Interesting. At School my maths teacher believed that women out performed men until a certain age (I forget exactly, about 13-14), and it then switched. It certainly seemed to happen in our class. Perhaps that was the same stereotype threat in action.

Anonymous said...

So what would happen to the results if no aliases were used and they presented the myth at the beginning of the test that Women have edged out men as being superior in math in 2012?

Anonymous said...

"It's therefore not justified for you to refer to the overall male superiority in this study as evidence that men are in fact better at maths."

The study said it, did it not? Why is it not justified then? I am not denying that there are social effects that make women underperform in math. But it IS JUSTIFIED to say men perform better BECAUSE THEY DO. Not saying so is denying evidence.

Some women are better at math than some other men. It is pretty silly to lump all men and all women together like this in first place. It should really be a case-by-case study, but that's too expensive for researchers to do.

Look, we all want to believe that men and women are equal, but they're not. For example, men DO HAVE better depth perception. Women DO HAVE better color perception and sense of smell. Perhaps enhanced depth perception, higher-speed sports, and constantly integrating error during sports in our brains actually does make men better mathematicians. Who can really say for sure?

So where am I going with this? This is an age old problem - as old as humanity itself. The stance to take on this is: women are worse at math today, how can we fix this? Let's completely get rid of the social bias (oh yeah, that's really pragmatic, huh? It's only a fundamental human men vs women issue). No. So how can we fix it? Do studies like this add value? Should we have every girl and women take math tests under an alias? That seems like a bit of a clerical nightmare. So what then? Do we handicap the boys to make it equal? That seems counter-productive...

I'll tell you what we should do. We should be getting better books into younggirls hands: Jane Eyre, stuff like that. Stuff that makes girls accept responsibility for their own biases and their own beliefs. That way, when people say "men generally perform better", rather than "wasting time during the test worrying about perpetuating gender inequality", they can spend that time saying "Fuck that noise. It's bullshit anyways. i'm proud to have my name on this paper and I'm going to prove them wrong."

Happy to hear your thoughts.

Anonymous said...

Apparently they're also better than reading as ben must've missed this little nugget:

women who took the test under someone else's name, be it male or female, performed better than women who performed under their own name, and they did just as well as the men

Anonymous said...

The comments seem to indicate that women outperform m̶e̶n̶ ben in both reading comprehension and deductive reasoning skillz.

Jack said...

Thank goodness this article wasn't about elementary logic, Ben, because you would not be helping the male stats at all.

Anonymous said...

one of the points in this is:

women perform better under an alias than as their real name.

however, i believe there is a flaw in this assumption/point/theory with the related data. how can you compare 2 sets of women without a baseline of their mathematical skills? it seems that they are assuming that both sets of women (true name and alias name) have are at the same mathematical skillset. i do not believe that is a fair assumption, especially when the testing set were all undergrads even more so with a testing set of 110 women.

i would be curious as to see this same test being performed with some sort of standard or minimum requirement for mathematical knowledge in order to create that variable as constant as possible.

Therapist Los Angeles said...

It's like the book Black Like Me except it's Female Like Me. And for math.

Anonymous said...

It would be interesting to learn how women score on math tests vs. men in countries where there is less of the "geek stigma" that seems to prevent many American women from entering math and science careers. Judging by the numbers of females in American math and science careers who are from Asian countries, it seems that women in India and China don't have the same biases against math and science that many American women do. I wonder how they score on math tests compared to their male counterparts?

Anonymous said...

Ben: Ouch, you haven't heard of 2x2 factorial design and "main effect" and "interaction"? Oh dear. Maybe you shouldn't be displaying your ignorance by commenting

Poliwhirl said...

This is pretty depressing. My 8-year-old daughter is going through the same dilemmas that I went through as a child, namely: act dumb, look pretty and be accepted by most everyone, or act smart and have most everybody put you down for not acting dumb and looking pretty. And yes, she's good at maths and couldn't care less about clothes and hairstyles, but let's see how long that lasts.

Anyone, man or woman, who thinks that there isn't a gender bias stopping girls/women in our society from reaching their full potencial is either ignorant, an idiot or delusional.

Anonymous said...

Oh good Buddha. Men AND women, boys AND girls are helped AND handicapped by their respective sex/gender. We have been playing this "woe-are-the- girls" card for 40 years and it has worked wonders- for girls. It has been rather disastrous for boys. How about we try to do what is best for everyone? The goal is equality of opportunity, not equal results. It does not add to society to attempt to make girls and boys "equal" by improving opportunity (and results) for girls and women while decreasing opportunity (and results) for boys. You think it was a problem when women, who had historically been raised to be more accepting, figured out that they wanted more? Wait until the men, who are raised to fight and die, figure out what a load they have been handed. We either start doing the right things now or we will be forced to ride the tiger later.

Anonymous said...

When did they start calling it "maths"? It was called math when I was young.

Unknown said...

"maths" is British

Dean D'Souza said...

"Anyone, man or woman, who thinks that there isn't a gender bias stopping girls/women in our society from reaching their full potential is either ignorant, an idiot or delusional."

...or (on some level) in denial.

Dean D'Souza said...

"makes girls accept responsibility for their own biases and their own beliefs"

Ben: you don't seem to understand what biases are. In simple terms, they're "deep rooted". You can't hand a book of Jane Eyre to a child and believe that this will get rid of racism. Men have a role to play in reducing racism, not just women. We need to change our expectations of what men and women can and cannot do. It's hard. For instance, my girlfriend was given dolls when she was a child, even though she wanted lego! Dean

Post a Comment

Note: only a member of this blog may post a comment.