Also, over the last two decades there has been a significant increase in cultural recognition and representation of male "nerds," as people who belonged to that category have gained positions of influence, thanks to the tech boom in the 90s, which can sometimes inspire students to pursue their interests despite the bullying. The lack of such representative role models for young females is disconcerting and could even cause some people to be slightly more affected by the bullying, because perhaps they don't know a female who works in STEM or know of anyone in recent years who made it big, and simply didn't know it was possible to go against the social tide (which happens all too frequently.)
All in all, I think there are a lot of factors at play here and reducing it to one cause seems like wilful ignorance at best.
The impact of all of these weird pro-STEM programs is a really complicated discussion, but let me suggest that it doesn't really make high school any easier for the vast majority. (The social structure of kids doesn't really look to DuPont for ideas about which way is up...) Your response almost reminds me of how some people are feeling left out of minority help programs ("what did I do to deserve a landscape devoid of scholarships," that kind of thing.) without realizing exactly how little help those programs really provide. In a sense, overestimating the effectiveness of a response to privilege so harshly that you think the privilege has been reversed.
I didn't mention my geographic location (Toronto) or the root causes of the cultural shift (I can't speak to that), so these are some strange assumptions to make. I'm also not really following the rest of your argument, tbh.
In any case, I stand by my argument that this:
>are you implying that male students who go for stem don't face bullying akin to being "geeky" or "nerds"? Please...
is a fairly tall order in 2018. Positioning yourself as someone who seeks a career in STEM, regardless of your gender, is not marking yourself out for bullying, and hasn't been for a long time.
Sure, but we can also present similar anecdotal evidence of visible forms of bullying from fellow male students for having geeky interests. So, where does that leave us, really? Similar to your criticism about jumping to causative conclusions, you've determined that this passive-aggressive bullying is the case of gender roles rather than the more obvious case of bullying nerds.
Compared to Algeria, Albania, Turkey, Tunisia, UAE, Romania, etc? Please get your head of western social "justice" bubble. The factors that you think the article doesn't take into consideration are not explicitly there because they are implicit to everybody who really knows these countries.
That's "natural inclinations" for you.
I also think that blanket questions such as "Would you like to have a job where you sit all day at the computer?" would be rather unhelpful because it really depends on the nature of the job. Both theoretical physicists and people working in building security have to sit at the computer all day, but that in no way implies that the nature of the job is similar.
The simple fact is that if your claim is that discrimination is the primary cause for the lack of representation of women in STEM fields, then these data show that you are wrong. Not because we can eliminate subtle pressures with 100% certainty, we don't need to do that.
The hypothesis is wrong because it makes the wrong prediction: if discrimination were the major cause, then more discrimination would lead to more unequal distribution. And the data show that it's the other way around: less discrimination leads to more unequal distribution.
It's not a matter of degrees, it's a matter of sign, of the direction of the change. And that's effectively boolean, the sign is either positive or negative. The "discrimination" hypothesis predicts the sign is positive, in reality the sign is negative. Can't really get any more clear-cut than that.
The question is: how do we measure the problem of discrimination? Many people assume that we can measure discrimination by looking at the percentage of women in engineering and comparing it to the percentage of women in the general population. Many people believe that any difference between these numbers proves discrimination. But an increasing amount of evidence suggests that this a bad assumption.