(click to enlarge)
The Public Religion Research Institute launched a new interactive feature last week called the "
American Values Atlas," which lets you access a treasure trove of polling data (conducted by PRRI themselves, though
Pew and
Gallup also do a lot of religion-themed polling) on policy issues, but also on the demographic questions of who belongs to what religion, and where they live (which, as I've pointed out often, is a hugely important part of political geography, but something that the Census Bureau doesn't cover, meaning we need to look elsewhere for data).
If you're looking for a quick summary, PRRI hits a few of the big findings. America has ceased to be a majority-Protestant nation, and in 19 states, white Christians (of all denominations together) have ceased to be a majority. Even as Christians become a smaller segment of the country, at the same time, Christians are becoming less white as well (especially among Catholics, who are increasingly Latino, but even among evangelical Protestants as well; there has been strong recent growth among Latino evangelical churches, for instance).
Also worth noting is the rise of the "unaffiliated," people with no religious membership at all. They now comprise 22 percent of the population, and that's poised to grow significantly: young people (34 percent) are three times as likely to be unaffiliated as senior citizens (11 percent). Other non-Christian affiliations are poised to grow as well (based on the age of members): Hindus and Muslims have an average age of 36. Compare that with white evangelicals, who have an average age of 54.
The deluge of PRRI data prompted a variety of interesting new maps and charts from other sources over the last few days, as well; a good starting point may be the collection of maps that the Washington Post put together, parsing out which states have a Catholic plurality, which have an evangelical plurality, and which have an unaffiliated plurality. It also contains dozens more maps looking at each particular religion, and what percentage of people in each state are adherents.
If you want to see all that information condensed to one map, though, community member Dreaminonempty put together a composite map (the one featured at the top of this post) that looks at whether states have a Catholic, Protestant (all Protestant, not just evangelicals), or Mormon plurality, and how dominant that plurality (or majority) is.
There's more over the fold.
Kyle Kondik and Geoffrey Skelley of the Univ. of Virginia Center for Politics delved further into the political implications of this new data, noting the high correlation between states that have a high evangelical percentage, and states that gave a larger share of the vote to Mitt Romney in 2012. They also scatterplot some other religious groups as well, finding a similar (but not as strong) relationship between percentage of "unaffiliated" in a state, and its Obama percentage in 2012. Catholicism doesn't generate as strong a correlation, though it's interesting to see that there's a stronger relationship based on percentage of white Catholics versus percentage of Hispanic Catholics (think about, say, the difference between how Massachusetts and Rhode Island vote, versus how Texas and Arizona vote—a lot of which has to do with different levels of turnout).
There's one other element to this question, though, that we should add to the discussion: there's a strong correlation between how white evangelicals in a state vote, and how white non-evangelicals in a state vote. In other words, Alabama, Louisiana, and Mississippi are both among the states with the most conservative white evangelicals and white non-evangelicals; Hawaii, Illinois, and Oregon are among the states with the most liberal white evangelicals and white non-evangelicals. (The graph below is based on 2008 exit poll data, not the new PRRI data.) In other words, there's a feedback loop; it may not be that making a state more evangelical will make it more conservative, but rather, that living in a state that's more conservative (and racially fraught), and taking your social cues from everybody else around you, means that white people are more likely to vote Republican regardless of their religious beliefs.