Slow but steady acceptance

I largely stopped reading Nate Silver after the election, and I really shouldn't have, because he's still pumping out interesting stuff. Recently, he produced a little regression model predicting state votes on gay marriage bans. It turns out that you can predict a huge amount of the variance in those votes just by knowing state religiosity and the year of the election. With regards to the year variable, state support for gay marriage bans appears to be declining at about two points per year. So a gay marriage ban would probably pass in Iowa today. But by 2012, the odds will be even.

Yes, Nate's going beyond the data a bit in making these projections, but they're still consistent with what we know about public attitudes toward gay rights.