Female coders are rated more highly than men, when nobody knows they're women | Hacker News Reader

Female coders are rated more highly than men, when nobody knows they're women

qz.com

11 pointsgpresot10 comments

Rehash of this previous submission: https://news.ycombinator.com/item?id=11074587

The drop in pull request acceptance from gender-neutral to gendered is worse for men than for women. (Figure 5) Anyone can cherry-pick parts of that preprint to support their argument.

Apart from that it isn't. As Figure 5 shows. But ok..

This is why I do not believe people when they say "I hire people solely on the basis of ability" as an argument against making any efforts towards equality or equal representation.

I don't think (many) people are willfully biased, but every study bears out that bias exists. Make selection blind to race, sex, etc., and the results are different. Orchestras switched to blind interviews and hiring profiles changed dramatically. Suddenly females and minorities were rated higher and got more job offers.

Its sad, but true, that we have this unconscious bias and we need to fix it. I don't discount the conscious biases that clearly exist, I'm just saying a claim of "I'm not biased, my hiring is fair" needs to be backed up with data, not assertions.

I guess that's why, in youth orchestra contests, the judges sit behind a barrier when auditioning and making their notes.

The results seem reasonable to me: with 90-95% of coders being male, that means that this group has probably a wide distribution curve in terms of skills. The group of female coders probably has a narrower curve and likely skewed towards "good" (i.e. gifted and truly passionate about coding).

That is exactly what their study concludes:

"One explanation is survivorship bias: [...] Another explanation is self-selection bias: [...] Yet another explanation is that women are held to higher performance standards than men [...]"

Or in other words: Women who stay long enough to end up submitting PRs submit higher quality code, while the much bigger population of men, under much less social pressure, has a more well-distributed quality curve; and even women who get this far are biased against if their gender is known.

Obviously my comment only refers to the part of the article talking about the better grading of female coders, not the part about the bias.

I've had one female coder tell me she's picked gender-ambiguous names and doesn't show her pic online for this reason. Only once she's proven her ability does she reveal her gender.

From Kon_air in the other discussion thread.

And then you discover in Questions Section; "Our analysis (not in this paper -- we've cut a lot out to keep it crisp) shows that women are harder on other women than they are on men. Men are harder on other men than they are on women." https://peerj.com/questions/2002-do-you-have-data-on-the-gen....

The research paper is here if anyone is interested: https://peerj.com/preprints/1733.pdf

Reply on news.ycombinator.com