| > What does it mean to "leave the field"? You can read it in the linked article: http://fortune.com/2014/10/02/women-leave-tech-culture/ * "716 women who left tech" * "I have collected stories from 716 other women who have left the tech industry" * "Of the 716 women surveyed, 465 are not working today." * "Two-hundred-fifty-one are employed in non-tech jobs, and 45 of those are running their own companies. A whopping 625 women say they have no plans to return to tech Which strongly implies "left the industry", not "moving to product management" or "managing". 2/3s of them aren't working at all, not "promoted to management" |
That seems like a shockingly high figure of people who have the means to avoid any sort of labor productivity. I can only conclude that "are not working today" is loosely defined or that these people were unfit for working at all, since someone that has the means to quit work and not have to look for other work elsewhere probably hasn't lead a very rigorous labor existence.
Also, this appears to be study of only the ones that left. I'm certain I could find 716 men that left the tech industry too, and come up with a set of reasons why they left. Looks like a classic case of selection bias. There doesn't appear to be one women who stayed in the industry in that study. Does that mean I should conclude from this study that 100% of women leave tech? Simply put, you need to survey more than just those that left.
That study comes across as far more biased than the biases it's trying to combat.
It's far more interesting to discuss base rates. Start with a sample of women in tech, follow them over n-years (you can get a good representative sample by choosing different cohorts like those that are recent grads to those with 5-10 years industry experience) and do the same with men. After 2-3 years check out how many from from each gender from that sample have left the industry. Interview them to find out why.
My ex was a documentary filmmaker interested in social causes and whatnot and I've seen how the sausage is made firsthand and know how data and statistics are twisted to support an agenda. Good statistics that strive to be impartial almost never produces numbers as "story-worthy" as the ones from that study, which means you need to question the numbers presented and also ask which figures were conveniently omitted.