It's kinda like how one has to be careful measuring people interviewing for jobs, since someone who interviews poorly will spend many interviews before being hired but someone who interviews well will only do a few (on average). So measuring the average ability of the candidate in interviews will be worse than the average ability of people who have been interviewed.
Another thing I've noticed is how many urban women get their exercise from "classes": barre, yoga, pilates, "orange theory", etc. Go into one of these and you probably won't see a single male there.
Yoga and pilates attracts mostly women, but they won't kick out males either (as in, I seen guys there, but like 1-2 for full class of women).
https://www.npr.org/2018/03/19/594719471/guys-we-have-a-prob....
When I thought about it afterwards, maybe the mistake is to treat your socialization strategy as something inflexible "I do it that way" instead it is set of skills, habits and decisions that can be learned and tested.