Then just say that, instead of substituting the missing value with a sloppy and misleading calculation of your own. There were only 52 values to start with, and you can't non-randomly reject half of the sample size and pretend that the result has any meaning.
The study did cite their source, with the expectation that anyone wishing to check their work would use that, rather than Table 2.
I take this to mean that the barriers to including all the values are too great for you to bother with as well. Also, you absolutely can throw out values that may be less reliable for those that are believed to be more reliable. The result is meaningful. If you want a different calculation for comparison, go make it.
Then they go on to observe that under improved gender equality conditions, VR approaches 1.
What section are you referring to here?
This says when cultural conditions are equal, so are performance results. If there were a biological component involved, this should not be true... some inequality in results should remain despite equal opportunity. Therefore, the greater male variance hypothesis is false.
I do agree that their statement about VR being "similar in value" across nations does take things too far, and does argue for a "biology only" position. Some fluctuations due to social influences should be present. But if gender equality means that VR approaches unity, then that is a powerful "not biology at all" argument. And that's what the data shows.
Seriously, are you just making that up or is my copy of the study missing pages?
The countries the authors call out as having nearly equal VRs include the Czech Republic (GGI=0.6718), Indonesia (GGI=0.6550), Morocco (GGI=0.5676), Tunisia (GGI=0.6283), and the Netherlands (GGI=0.7383). Outside of the Netherlands, those are low GGIs (sometimes very low). I think that's why they specifically stated
we also conclude that VR is reproducibly essentially unity for some countries.
rather than stating as you have implied that countries with more gender equality are the ones with VRs closer to 1.
