January 09, 2010

Decimals Are a Girl's Best Friend

It's a lovely headline, that. It's also some fun new research.

A new study that looked at math scores for boys and girls in 69 countries corroborates her experience.

"There are no differences in girls' abilities in math," says Nicole Else-Quest, who led the study that analyzed the tests of nearly a half million students aged 14 to 16. "When they have the same resources boys have, nothing holds them back."

Her research linked individual achievement in test scores to the local status of women. In countries where men and women are perceived as equal, girls perform better in math. In Iceland, girls outperform boys; in Korea, boys outperform girls. In Canada, the difference in girls' and boys' scores is too small to be meaningful, Else-Quest said Wednesday.

Of course this means that people will step up to say, "Well, guys just have a greater range of math ability and the top mathematicians are just going to be guys." I strongly recommend they read this first.


Becca said...

The weird thing about the extreme tails argument is that people think it might have anything to do with genes. Whereas everyone who knows anything about genetics in different populations knows that people with African ancestry have higher genetic variation; ergo, logically they should also have the most genetically-predispositioned to both dullards and geniuses. Yet, strangely enough, I never hear white men arguing that their greatest mathematicians ought to be inferior to the greatest mathematicians Africa should produce.

Joshua said...


Someone could consistently argue that their is a variation effect but that we don't see this in Africa and with people of African descent because of the other, essentially environmental problems. Or they could argue that the the variation that occurs there is just not the relevant one. Both of these are pretty clearly ad hoc hypotheses but both are testable. But yes, the problem of African ancestry does pose at minimum a pretty serious problem for any simple claim that the differences are due to changes in the level of variation.

Incidentally, while on this topic, there's also a recent study by Sian Beilock's group at U Chicago looking at how children pick up cultural norms about gender and math that is worth looking at.

The original study can be found at http://hpl.uchicago.edu/Publications/PNAS_2010.pdf

and I discussed it at:

It appears that children may pick up on cultural beliefs about how their gender should do mathematically in some very subtle ways.