You start your post by bringing an anecdote which supports his point: STEM ability is, to a very large degree, biologically determined-as you point out, it runs in families. As does, for instance, athletic talent. Which does not mean that it is equally distributed between the genders-athletic talent certainly isn’t.
> let’s assume for the sake of this argument that those claims about the “average” woman are correct. Statistically speaking, you are describing about 68% of that demographic.
You are abusing statistics.
68% lie within 1 standard deviation of the median. Half of the curve is below the median. So the number you should be starting with is not 32%, but 16%.
Second, you’re purposely ignoring his point about overlapping normal distributions, and what rapidly happens as you move to one extreme or another.
Third, engineering is a field where productivity follows a log normal distribution. Top engineers are highly disproportionately productive. The sexy tech unicorn companies rely on their ability to attract not median engineers, but the top talent. See the second point-as you move further and further to the right on the talent curve, one of the normal distributions will be represented more and more over the other.
Hence, as you well know, competent lady engineers performing at the highest levels are hard to come by, and the tech companies are scrambling to attract them, to the point that they pretend MTF transformers (like Justine Tunney and this one,) are lady engineers and count towards the imaginary quota (obviously not true,) and even create special make-work teams for them.
