Heard of the study that finds that beautiful women have more girls and intelligent men have more boys? Andrew Gelman finds that at least the first finding is suspect on multiple levels.
You get a statistically significant effect if you compare the top quintile of beauty to the bottom four quintiles. The effect goes away if you simply estimate the sex ratio as a function of beauty.
The author apparently had three measures of beauty, but only used one. He should have averaged them instead of picking one. Why didn’t he?
The author miscalculated the implied probabilities of his logit model. His estimate implies an 8% effect, not a 26% effect. (Journalists then somehow inflated 26% into 36%!)
Gelman avers that he doesn’t want to set himself “up as some sort of statistical policeman.” Why not, I ask? The academy could use a good, honest cop on the beat.