Am I speaking Russian? I said this. We need a statistical trick, either arbitrarily saying you only need 1 out of 10 (or whatever) to have diagnoses or using a different way of deciding it altogether, and I like the 2 question questionnaire that identifies something like 80% of autistics and 70% of nonautistics, and has the advantage that you don't have to pry into someone's personal life in a way they may be uncomfortable with, especially if they're in denial. It would just look like you would any street questionnaire people; there's a normal, unthreatening way to approach that.
