Having read the actual study ( http://people.hbs.edu/bedelman/paper...ightstates.pdf ) it seems to me the author is pushing an agenda. It became pretty obvious when he began correlating porn buying with hypothetical religious statements.

Also, he concluded the highest rate occurs in urban areas.

Now, it's important to note that states are not homogeneous. You can have areas of Republicans in blue states and Democrats in red states. The author assumes that states are homogeneous. I also couldn't find the raw data, before he 'corrected' for several variables, that could show certain states with a much higher gross amount of porn buyers. As it happens, in the US, urban voters are more likely to be democrats.

So the porn-buyers the author is using to impugn red states are likely the urban democrats in that state.

CR