Gender bias in open source: Pull request acceptance by gender [Preprint]

I like how this paper is constructed. They form a bunch of hypotheses and test them and aren't afraid to be wrong. Rather refreshing since the typical papers I read are more of the "here's our successful test" variety. I feel like a supplementary qualitative analysis (code quality) would be helpful. My personal hypothesis is that women self-select and the pool of female contributes has a higher average skill than the pool of male contributors.

I have no evidence to back this up but by gut says that it's still socially harder for women to become developers and as a result the ones that "survive" tend to be better on average. While this is a bias I'd argue that the major reason (cause) for the acceptance/rejection of pull requests is the quality of said request which would argue against a bias in the acceptance. I'm not sure how to phrase it well but in summary I think there's a social bias against women in programming (as a career) but I suspect code quality is the main cause of accept/reject decisions and there is no bias there.

Was this flagged into obscurity while I was reading the paper? I can't find it on the first five pages anymore. Pretty sad.