A coherent article trying to establish evidence of a particular argument or claim would include all available studies and weigh them against their various differences in order to find a viable conclusion. It would, at the very least, attempt to search for alternate explanations and present them, and then make an attempt to debunk them.<p>The article's main problem is it's trying to attack a very specific problem, which is actually part of broader trends in human behavior and culture. It doesn't address that at all.<p>The author claims "gap != discrimination", which is true in basic definition and theory. But if you look at all the other studies that show things like hiring discrimination, discrimination getting into a good school, discrimination based on where/how one grows up, cultural discrimination, etc, gaps immediately form and cause a ripple effect.<p>That there are about as many black men as white women (or less) in a given tech field is one example of this. Pick any other discriminated-against minority and you see the same, except when hiring from overseas (economic benefit) or hiring someone with a preferred social stereotype.<p>If the author's argument is that women simply prefer not to go into this field due to some weird connection between math and speech, then you have to ask why the connection? Is it due to "biology", or social conditioning, or the wariness of sexual harassment in a male-dominated field, or the low pay, or a combination of these things? This article doesn't attempt to look at those, and its limited view of the subject leaves it running into the same correlation-is-causation conclusions it laments.