If the issue is how common the people are going to be by demographic in the workforce, then shouldn't we start early in making them comfortable with the people they're going to be around, as the ice breaking generation, by mentoring them with the common rather than the rare?

I mean I genuinely do not understand why this is actually helpful. The lady engineers I talk to are either all pandering to me, or are as confused as I am.

I don't understand why we're presuming that a girl can't have a good, positive male role model, just like I, a boy, had a good, positive female role model.

.

"She might go on to be one of a tiny (and steadily decreasing) number of women in an overwhelmingly male university-level engineering program."

So, as we bucketted women out of the mainstream and told them they could only learn from people like themselves, their global participation began to drop.

I mean really, the data you're recommending seems to suggest that what we're doing is destructive.

It was not the case, in Elizabeth Cady Stanton's work, that the newly liberated woman was expected to learn only from other women. They're supposed to be equal.

The supporting claims seem to be assertions of better-ness with no supporting data or logic other than an intuitive feeling that the child's recognition of similarity is somehow more important than the child's early exposure to diversity (something most data backed child rearing studies disagree with.)

I do understand that your opinion is that a girl will have an emotional need to only learn from women.

Do you have any data to defend this? This seems bizarre and offensive to me, as it does to the female engineer who spoke up elsewhere in this thread.

I think that, just like men, women can learn from anyone, and should choose the best source, not the most similar source.

You can't create an inclusive society by educating explicitly exclusively, in my viewpoint.

I would love to see actual data, rather than beliefs or opinions, which suggest that learning from a demographically similar role model produces better results than learning from the best available role model, or that taking a role model from what you claim is an extreme minority, rather than building bonds and understanding with the majority, leads to more inclusion, despite that you also claim that inclusion is decreasing, despite that this sort of like-only mentorship is increasing.

It seems like your own data contradicts you. Maybe I'm misunderstanding.