Related to the question of why there is a gap between the genders in math and the sciences is whether there are possible means of remedy. With respect to possible remedies it is often a good idea to look internationally at which countries don’t have this problem — to see what they are doing right.

Guiso et al. used data from the 2003 Programme for International Student Assessment (PISA) which surveyed 15-year-old students from 40 countries who took identical tests in mathematics and reading. They compared this data with measures of the gender equity in these countries.

One such measure is the GGI index from the 2007 Global Gender Gap report. The index includes measures comparing the economic participation, educational attainment, political representation and health between men and women.

The authors found a significant correlation between GGI and the disparity in math performance between boys and girls. This is depicted below. (Figure 1 in the paper)

The R^2 for the correlation between GGI and math performance is .32.

The authors correlated performance on math and reading tests with other indices for cultural attitudes towards to women using data from the World Values Survey and political empowerment of women using data from the World Economic Forum. Both these indices also showed significant positive correlations with reductions in the math disparity between men and women (R^2 = .23 and .21).

The authors also considered that the difference in math performance represent systematic differences in genetics between the countries surveyed — maybe women are better at math in Norway:

To verify that these results are not driven by biological differences across countries, we analyzed whether they persist in populations that have a similar or identical evolutionary history. To assess history, we used a genetic distance measure (14-17) based on the frequency of each allele across DNA polymorphisms.

According to this measure, there are 13 European countries with genetic distance equal to zero and 26 European countries with genetic distance less than 100 (table S5). When we restrict the regression of the table (above) to either one of these two groups, our findings are substantially unchanged (table S6).

(Someone who know more about this is going to have to say whether they did that properly. That part is straining the limits of my understanding.)

Interestingly, when the authors look at reading scores, there is also a positive correlation with gender equity — only this one in favor of the girls. They calculate that as equity increases this gap will likewise increase.

**The authors acknowledge that there are observable differences between boys and girls that do not vanish with increasing equity.** For example, the boys are on average better at math than they are at reading in all countries surveyed. Likewise, while the overall gap in math performance closes, differences in subfields of math (geometry vs. arithmetic) do not:

These results suggest that the gender gap in math, although it historically favors boys, disappears in more gender-equal societies. The same cannot be said for how boys score in mathematics compared with how boys score in readings. Boys’ scores are always higher in mathematics than in reading, and although the difference between boys’ math and boys’ reading scores varies across countries, it is not correlated with the GGI index or with any of the other three measures of gender equality (table S7A). Hence, in countries with a higher GGI index, girls close the gender gap by becoming better in both math and reading, not by closing the math gap alone. The gender gap in reading, which favors girls and is apparent in all countries, thus expands in more gender-equal societies. Similarly, although the gender gaps in all math subfields decrease in societies with more gender equality, the difference between the gender gap in geometry (where the boys’ advantage relative to the girls’ is the biggest) and arithmetic (where the boys’ advantage relative to the girls’ is the smallest) does not (table S7B).

This is a good piece of work that I think provides a good piece of evidence in explaining the disparity between men and women in math and science. Further, these results are very similar to a study that compared the performance of male and female chess players as a function of their representation in a particular zip code. The study found that in those zip codes where there was equal numbers of men and women their performance was equivalent. **Equal representation does appear to improve girls’ performance.**

I have argued before — so I won’t again — that innate differences between the genders are not sufficient to explain the disparity in representation in math and the sciences. Some people have been misinterpreting me when I say that. I am not saying that no innate differences exist on average between men and women. Any gender psychologist would take a great deal of issue with that statement. (And this study supports that notion because of the failure of the superior performance of girls in reading to disappear.) What I am saying is that these differences are not __sufficient__ to explain the disparity in representation. I will even grant that they may be part of the issue (a sum of variances approach), but from what I have read of the data I am unconvinced that the effect is very large or primary — particularly in comparison to the cultural effect that this study shows.

(Please actually read the above cited post before commenting about your prejudices to the contrary. Furthermore, since I know someone will say it: the upper tail effect is not sufficient either as men and women in the sciences do not necessarily come from the upper tail in mathematical ability.)

Now I will grant that this is a correlation study. It does not speak to the issue of causation. There could be some other cultural factor that is confounding. It could be income or access to child care, or a million other things. Also, the correlation factors — while significant — are not overwhelming. There are no doubt other issues at play.

What I am arguing, however, is that this is another piece of evidence suggesting that the disparity between men and women in math and science is __primarily__ cultural, not innate. How are differences in innate ability supposed to account for this data? **Basically for the innate differences hypothesis to be true, you have to say that not only are women and men different, they are differently different in each of the studied countries.** Is that statement accurate? Is the effect of prenatal hormones on brains different enough in France and Norway to explain this effect? Does the additional X-chromosome function sufficiently differently in various countries to explain changes in performance?

I for one don’t think so.

Guiso, L., Monte, F., Sapienza, P., Zingales, L. (2008). DIVERSITY: Culture, Gender, and Math. Science, 320(5880), 1164-1165. DOI: 10.1126/science.1154094