[concordDance]

That particular debate is often a semantics debate, so it isn't in the domain of science at all.

The main way I can think of off-hand to try and make it scientific is to ask about correlational clusters. And then you get way more than two genders, but you definitely get some clusters that contain both transwomen and men (e.g. if I hear a video game speed runner or open source software passion projecf maker using she/her pronouns they're trans more often than not).

[darksaints]

I have noticed certain groups where trans people are relatively over represented and group involvement more correlated with biological gender, but that’s not actually that interesting or meaningful in reality. Trans women having similar interests to men doesn’t make them men any more than me owning a gun makes me a Republican.

[concordDance]

It would by a "correlational clusters" gender definition put some transwomen in a mostly male gender (though, again, you'd have a lot more than two genders with with that definition).

And correlational clusters is one of the few ways it's not just semantics.