送交者: mangolasi 于 2006-1-16, 09:12:46:
A paper from Lancet (I hadn't read it yet but will this afternoon), providing evidence to (indirectly) refute Emily Oster's Hep B hypothesis. The method to use is simple: sex ratio conditional on the sex of first birth (Andrew Gelman has similiar idea on another unrelated topic on the effect of sex of children to the parents' ideological leaning). With the first birth of girl, the girl-boy ratio is 759:1000, while with the first brith of boy, it's around one.
The most scary fact: "The other piece of suggestive evidence is the fact that the effect is stronger from educated women, who presumably have more access to ultrasound and abortions."