> that is the result of sociological rather than economic factors Economic and sociological factors aren't completely separable. Specifically, you can't assum

> that is the result of sociological rather than economic factors Economic and sociological factors aren't completely separable. Specifically, you can't assume that economic factors don't drive marriage rates. For example, people with college degrees are more likely to get married. Is that because they can afford to? Because the same type of people who commit to schooling also commit to marriage? Would decreasing the cost of tuition increase the marriage rate? It's a correlation, and trying to infer causation from it is always going to be dicey. That is still an interesting graph though. For example, looking at the top line (dual income) and light green (married, only husband has income), it looks like basically since the 70's wages for men have been flat. The growth appears to be almost entirely accounted for by growth in wife's wages.