That's our hunch, at least for now. I doubt that the code screen is somehow gender-biased, but I'm certainly not going to reach that conclusion with a sample size of 21.
I don't think it means much in the big picture but it does seem to be a reasonable example of a company where the hiring process is fair, gender-wise, from start to finish. The only problem is that a mere fraction of the applicants are women.
And it makes it easier to believe that, at companies unable to attract and screen that many candidates, they may get no qualified female applicants, period. It lends plausibility, even if it's not actually proving anything.
It's a question that others will probably ask themselves as they read the article. Some maybe don't understand significance and others, like you, will just miss the frequency counts. So I think your comment still had value, maybe with an EDIT added if you want to make it clear that you see the error now. It's your call of course, but I tend to think that comment deletion is something that should be avoided when possible.
