> Next time flip it: Have the subjects interact over text as the proxy for intelligence It's possible to get good measurements of intelligence that way, but fo

> Next time flip it: Have the subjects interact over text as the proxy for intelligence It's possible to get good measurements of intelligence that way, but for purposes of evaluating the effect of intelligence on attraction it would be about as accurate as using "distance between the top of the person's head and sea level" as a proxy for height.