you mean for how i know correlation=/=causation? i too would like a good source for understanding stuff like this, but i have yet to find one. i also haven't really looked...
anyways, a good way of explaining correlation=/=causation is:
when ice cream sales go up so do shark attacks. the logical fallacy correlation=causation would have you believe that the increase in ice cream sales caused the increase in shark attacks or vice versa (because they're correlated), but when understanding correlation=/=causation, we can see that the increase in ice cream sales and the increase in shark attacks went up at the same time and neither caused each other. they were both actually caused, in part, by the warmer weather and its provoking of human lust for beaches and sweet/cool food.
does that make sense?
i think it's by how T affects skin. higher levels and/or sensitivity makes skin oilier and stuff. i really dont know any biological mechanism, but there is tremendous empirical evidence that higher T levels and/or sensitivity increases acne. although, it really varies for different people, and of course genetics are very involved.
edit to add: i may have misread your second remark. i read that you said that there's no biological mechanism for T affecting acne, when after reviewing it again, it seems you actually meant that there's no biological mechanism for creatine affecting T. i honestly have no clue.