If you calculated your inter-country variance based on the information from Table 2, your data set is incomplete.
Table 2 is restricted to information from countries in which 3 separate measurements had been made. By averaging these samples, the result should be more reliable than those from countries in which fewer measurements were taken. The selection process had nothing to do with their statistical consistency, so any bias is unintentional. If you would like to compute a mean and variance for all the inter-country VR measurements available, please be my guest.
I'd agree with the statement that 0.91 to 1.31 is a wide variance, and 0.91 to 1.52 even more so.
My point is that it's sloppy (or misleading) when you are presenting a statistical analysis to quote the range of variation as being large without calculating and stating the variance. The range is largely meaningless, while the variance is telling.
Since the study is out to determine how mathematical performance differs among the sexes amid different cultures, that was the right choice.
The authors take their performance data and use it to comment on biology. Their tests are not well suited for this purpose.
So, what biological revolution occurred in the US between the sexes from the 70s to today that can explain these results? Answer: none. What social revolution can explain them? Answer: feminism.
When you realize that we agree on this you'll be one step closer to understanding my position.
Similarly, this
Next, we tested the greater male variance hypothesis.
If true, the variance ratios (VRs) for all
countries should be greater than unity and similar
in value.
is a complete misrepresentation of the natural greater male variance explanation, based on the assumption that this explanation excludes the possibility of any non-biological effect on performance, as well as the assumption that if the explanation were true for any group it must be true for all groups around the globe. I don't agree with these assumptions.
In this way, the authors essentially argue that the evidence they have purportedly demonstrated in support of a nurturing effect absolutely contradicts any natural discrepancy. However, it only contradicts a nature-ONLY discrepancy. Just because nurture is effective, doesn't mean biology plays no part.
In order to determine what the natural VR ratio should be, we have to try and strip away or control for the other contributing factors to that measurement. This study did not do that. The authors demonstrated that VR varied across populations, and jumped to the conclusion that the natural VR simply must be 1.0. Maybe it is, but that has not been demonstrated by the study. To do that is a much more ambitious undertaking, with difficulties I alluded to in earlier posts.
