They did try to account for social conditioning: parents' education and jobs, local labor markets, school performance, the whole bit. The gap still didn't move much. If socialization were the main driver, you'd expect the most egalitarian countries to have the smallest gaps. They don't. In a lot of cases it's the opposite. Sweden, for example, shows bigger differences in occupational preferences than places like Pakistan.

So at that point you're not pointing to a specific confounder, you're basically saying "maybe there's something else." Sure, logically you can always say that. But if the evidence keeps stacking up in one direction and the only reply is "could be something," that's just refusing to update your view.