al and tipper must really have been in love

An article in Slate reports on the finding that parents are more likely to divorce if they have a girl than if they have a boy. The author credits this basic insight to a recent study by economists although a sociological study reported that more than 10 years ago. In any case, the new findings appear to expand the finding to more cultures. Also, I had not been aware that the findings had been extended to show that remarriage probabilities are higher for women with boys than they are for women with girls, and that they had been extended to stepchildren from a second marriage, with divorce in the second marriage again being more likely if the family has sons than daughters.

1. The introduction of this article--"If you want to stay married, three of the most ominous words you'll ever hear are "It's a girl.""--is ridiculously overblown given that the finding is a 5% difference in the probability of divorce (from what I can tell, that's five percent, not five percentage points, so the equivalent of a difference between 20% and 21%).

2. The general interpretation of the finding is annoying because it presumes that parents have a preference for sons. Instead, all that is necessary for this kind of finding is that fathers have a preference for being in a familial relationship with sons, or even just that they feel more comfortable entering or staying in a father-hood relationship if the child is a son than a daughter.

3. I'm not sure if some people are going to try to put an evolutionary spin on this by providing some interpretation of why Darwinian theory would predict a preference for sons. This would possibly involve the Trivers-Willard hypothesis, which I've already written two articles criticizing (its application to contemporary societies). If so, the application would be wildly misguided given the findings about remarriages, where parents are showing the same preference for children who are not their own that they show for children who are their own.