| >strong correlation exists between a country's measures of gender inequity and the size of the math gender gap both at the mean and the right tail of the distribution 1. How does that correlation disprove it? It's entirely consistent with one common cause for all inequalities. 2. "Mean gender gap" is for people who did school-related tests, mainly PISA, not the general population. If the greater variability is true there will be much more severely mentally retarded males unable to even read. Depending on the size of the effect, the different mean could in fact confirm the hypothesis (that the mean is the same but std is greater). 3. No analysis whether the distribution of countries' result is consistent with greater variability for the whole human population. Especially jarring when they mention Iceland, a country so small that there are many individual schools elsewhere with more students than they have. >These findings challenge the Greater Male Variability Hypothesis, which, if valid, should hold for all representative populations, regardless of ethnicity or nationality. 4. That's an entirely unwarranted assumption (in addition to 3). Are Asian-Americans in particular really representative of a general Asian population (with an American culture)? What about brain drain? The more recent the immigration, the harder it's to get in; simplifying for the sake of an example, if you only allow those with a X234 mutation in a B5 DNA location, don't be surprised when there's zero variability in a B5 location. Even without a filter, why should the standard deviation in math skills be identical between genetically different populations? The identical variability should be true for the X-chromosome in general, not math-skills in particular. |