One that is not peer reviewed. Using Github data that doesn't even verify gender and anyone can enter anything into their profile. It guesses gender by linking

One that is not peer reviewed. Using Github data that doesn't even verify gender and anyone can enter anything into their profile. It guesses gender by linking the Github account to a Google+ profile using the email address. Something people accepting pull requests don't have access to. The sample size is different more men than women. Article doesn't state women are given discounts with tuition in college when they study programming and get special classes and mentors to learn to code better. That there are code schools for girls and women that don't accept men. That these programs and resources for girls and women are not available for men. Which might answer why their pull requests are accepted more often. Maybe the good female coders don't put their gender on their profiles? They used pen names or hacker names instead? Maybe the female coders that put their gender on their profile picture or name aren't as good as the ones who don't put gender on their profile. I found that women don't put gender on their profile to avoid getting hit on by men, so they use a male or gender neutral name instead. Some of the female profiles are transgender women, and that hasn't been taken into consideration either. The article seems to think there are only men and women in Github a binary gender and doesn't even cover transgender people and non-binary gendered people. There are all kinds of factors that could explain the results, which is why it needs a peer review.