Yeah, I noticed this too. (Edit: maybe this is because there’s no conviction. You don’t need a conviction to report the victim, but it might be problematic if recording perpetrators.)
That said, if intimate partner violence is marked as coming from a gay/lesbian relationship, I don’t see how this isn’t a close approximation for the perpetrator.
I’m not sure how you get to this conclusion. The screenshot I posted says (as I read it) that the rate for any kind of abuse from one partner on the other for gay/lesbian relationships where the victim is a woman is 9.8 somethings (I’m not sure of the units, the reddit commenter says it’s % but I’m not sure about that, and % of what?). Putting aside variance and assuming the units are comparable, how do you read 9.8 vs 7.0 or 4.5?
It seems like the data could be saying rates in lesbian relationships are higher than typical hetero relationships, but you seem to think it definitely doesn’t.