How do you support the rather extraordinary claim that the technology professions have markedly larger gender imbalance compared to other cognitive non-routine professions? From the data I saw, 2014 numbers placed technology professions around the 70%-30% gender distribution which puts it directly into the majority of professions of any kind.
The two extremes: Fair distribution (40%-60% or closer to 50%/50%) has as group a rather poor 10% of all employed people. Unsurprising, it has about the same number of women and men that works in those professions.
The second extreme are those with 90%-10% or higher, and if I remember right about the top professions in that list, it was topping with midwife nurse (99.7%), nurse with dentist specialization (99.4%) and carpenters that construct floors (99%). Going down from that you got several other specializations of nurses, mechanics, midwifes, construction workers, and kindergarten teachers (94%). You have to travel really far into the list to reach professions with as high gender equality as 70%-30%. To name a few cognitive non-routine professions at the 80%-20%, we got professions like veterinarian, psychologist, librarian, and teacher, all with worse gender distribution than technology professions.
Is your premise is that female founders are a fundamentally separate group from founders in general?
FWIW, I think it's a great idea and a good way to encourage women to join a field that they perceive as not being welcoming to them but I don't have any hard data to support this (and don't think I need any)
It's so broad a classification I'd have trouble seeing how it created biased framing. It's just a more precise way of saying "the professions" (in the traditional sense of medicine, the law, accounting, and engineering).
Why use the classification at all, if it so broad? If you use "all professions", you get perhaps things like garbage men and miners who are worse than tech? So you can always select a filter that makes tech look worst.
I think with some goodwill, you could see the point I was trying to make.