I think that misunderstands the purpose of non-rigorous surveys like this one, which after all only included the "125 [out] of the 384 female founders who have

I think that misunderstands the purpose of non-rigorous surveys like this one, which after all only included the "125 [out] of the 384 female founders who have participated in YC", because that's who had signed up for the mailing list they used. It's a rough estimate, and probably a lowball one given that their survey mechanism excludes any respondents who chose not to maintain ties with YC for one reason or another. The purpose of this survey is to jump up and down while waving its arms in the air and yelling "WE HAVE A PROBLEM"! We have a problem. We don't know the exact size and scope of the problem, but at a glance it looks like it's probably bad. I would love to see reliable data about precisely how bad, including among many other things gender cross-tabs. But that data doesn't exist because nobody's gone to the substantial effort to properly collect and publish it. Maybe, if we're lucky, this report will help to create the political will to bring some of that data into existence.