"unreported" is going off of, I assume, surveys on campuses. I can't imagine how else you'd get that information. Presumably good researchers are able to control for misinformation, but no survey-based data is ever going to be perfect.
I don't know how anyone would get "false allegation" data. I'm just saying that if it's based on the failure to obtain a criminal conviction then it's way too high - the failure to get a criminal conviction does not mean the allegation was false; it just means that the prosecutor couldn't get enough evidence to prove guilt BRD.