Correlation != causation. Might it be possible that women who play the woman card are not as capable as those who let their code speak for their ability?

>women may in fact be better coders

How can someone make such a blanket statement like that with so many confounding variables. Do you think it possible that funding sources and expected (and socially acceptable) results might have something to do with the conclusion these sorts of studies come to? Do you think there are grants available for researchers who would interpret this data differently?