Gender inequality in mathematics

[fclbrokle]

A paper I'm reading right now featured the following quote with respect to boys' and girls' mathematics performance:

Moreover, the variability of female identification and participation across countries changes over time and is significantly correlated with national indicators of gender inequality (Hyde & Mertz, 2009).

So I looked up Hyde & Mertz

Comments

[arrowedumbrella]

Hyde and Mertz do good work. I saw Mertz give a presentation at an MSRI conference (Critical Issues 2009), and it was one of the first times that I felt utterly convinced of the argument that girls and boys should perform at parity, instead of just hoping that it was the case.

[fclbrokle]

That very accurately describes my response, but being male, I felt uncertain if I could say it that way. :)

[meep]

Should?

What if they don't want to?

[meep]

Let me be more clear by what I mean. I have no problem with the averages being the same, but if we're talking right tails, then a lot more work goes into it, no matter the original genetic proclivities

[campbellsoop]

i recently read an interesting article about the physics GRE.... some study with some test i forget.. when people told girls that it was a special test designed so that there wouldn't be any gender discrimination in it, the girls did way better. and keeping the same questions, when people told girls it was just a regular test, they did worse.

[fclbrokle]

Yeah, that doesn't surprise me at all --- there are all kinds of similar results for minority groups as well.

(To be honest, though, it doesn't sound like they tried the case where they tell the girls that but also tell the boys something similar, but aimed at them.)

[ukelele]

Stereotype threat -- people have been doing lots of studies on it lately, with various tests and populations, and results like that keep cropping up.

thanks for nice reference!

[ajawa_took]

It is short/concise, full of data, reasonably all-encompassing, and at least on the face of it doesn't seem to be lying too much

Re: thanks for nice reference!

[fclbrokle]

In terms of variability, given that there are countries that consistently show equal variability between males and females (the Netherlands, Denmark), I think there isn't evidence for inherent differences in variability. (Their arguments about variability are based on high school level tests which are given to a broad sample of the population, so there's no affirmative action in there except maybe getting into higher-level courses that better prepare women for the tests, which seems a stretch.) Anyway, there are caveats --- how does one measure what "inherent" means when the surrounding society makes such a difference, and just because we haven't found evidence doesn't mean it's not true --- but lacking evidence to believe in higher variability, I don't see a particular reason to believe it's there

Re: thanks for nice reference!

[ajawa_took]

I think I was mostly aiming toward the point you and Meep make: that when environment is known to have a huge effect, it is unclear how to measure "inherent", and that something un-inherent keeping girls out of math is not necessarily evil, nor even external, to the girls.

I missed the weight of the Denmark/Netherlands evidence in reading too quickly. That's an extremely interesting question in itself, how could culture influence, not means, but variabilities?

Re: thanks for nice reference!

[fclbrokle]

Yes, it is quite interesting!

Let me give you an easily-stereotyping example of how culture can influence variabilities. In the US, for girls to succeed socially, it is important for them to avoid sticking out from the crowd as much as possible. For boys, it is no great social harm to do poorly in school, nor is it a big social harm to do well. (You might be called a nerd, but that's about it.) For girls, sticking out in either way is very damaging socially: you become ostracized from the other girls very quickly in the back-biting middle school culture. Thus, higher variability among the boys.

Now, this is vastly simplified and stereotypes in a whole bunch of ways, but it's a simple model for how a culture can do this. Now just scale it up to the real-world complexity we face. :)

[oxeador]

Out of curiosity (and this is an entirely honest and candid question): have you ever encountered other studies which concluded something different, and you have discarded them, instead of stating that they proved or refuted a certain point?

[oxeador]

To explain the reason for my question. I have recently realized that with topics like this, where I want one certain thesis to be true, I can catch myself doing exactly that.

[fclbrokle]

Well, yes and no. I have certainly encountered studies (or, rather, raw data) that showed both a higher performance level for boys and a higher variability for boys. These were in the US, and thus with easy to identify flaws. However, I didn't discard them; for example, I felt that Larry Summers was being treated quite unfairly when forced to resign Harvard. I felt that the issue needed more research, and there was contradictory evidence, and he was right to call for that additional study and remain open to all possibilities. (I *think* I have an LJ entry way back when to back that up, but I don't know for sure.)

[ukelele]

http://en.wikipedia.org/wiki/Confirmation_bias

Humans, alas, do this.