3 ms·
This seems to me like a rare "natural experiment" where correlation implies causality. Because whether a child is a son or daughter is essentially random, it's
by fogof 6y ago
This seems to me like a rare "natural experiment" where correlation implies causality. Because whether a child is a son or daughter is essentially random, it's hard to imagine a confounding factor causing both daughters and divorce.
- drjasonharrison 6y agoI think that the remaining factors will be the "difficulty" of parenting the behaviour of sons vs daughters in contrast to the difficulty a daughter or son experiences being parented. That is while you can use the sex of the child as an independent variable, you still need to explain why it has this effect.