Mark, please read the study I linked above. It's open access and draws similar conclusions to the one that is behind the pay wall.
I am not talking about "tests aimed at the middle."
I suggest reading the actual research before dismissing something as "junk science."
The various links have included many conclusions drawn from tests aimed at the middle, and it is to those that I was referring. I did look at the AMS link, but it clearly didn't reach the same conclusions I was objecting to so I didn't feel the need to respond to it. I'll resist the urge to snark back here.
From the AMS link:
The data presented here neither prove nor
disprove whether the frequency of occurrence of
people with profound intrinsic aptitude for mathematics
differs between women and men. What
they do indicate, however, is that this scarcity is
due, in significant part, to changeable factors that
vary with time, country, and ethnic group.
The earlier articles were claiming to have data that contradicted variability by looking at data which would be least likely to show it, whereas this article looks at the right kind of data, and explicitly says it remains an open question. These are not the same conclusions at all.
If this is true, then it should be true universally, since it's purportedly a biological, hard-wired difference.
If there were data on the high end (IMO, USAMO, SET, etc) showing the split was 1:1 for at least some statistically significant areas, then this would be very convincing evidence against variability, but that data doesn't show 1:1, it shows 3:1 or more in many cases. This certainly doesn't prove variability true, but it also doesn't prove it false.
Arguing on the internet ... I'll leave the last word to you.
