The map of faith

Looking for God? Well this is where best to find Him in America.

This map above shows “religious adherents as a percentage of all residents”. Click on the map to enlarge the image. Essentially, the redder it is, the stronger the faith.

It seems to me that the Bible Belt, which was traditionally seen as stretching from Texas, across states like Tennessee and Alabama, to Virginia – has been flipped up, through “tornado alley” and into Northern states like the Dakotas.

As though the importance of the Almighty in American politics wasn’t clear to you, below is the electoral map following the 2004 Bush-Kerry presidential election for you to draw your own conclusions.

Comments

[Ismael Klata]

Interestingly, various maps for depopulation (e.g. the first map on this pdf: http://www.census.gov/population/cen2000/atlas/censr01-103.pdf) look much the same.

[Candadai Tirumalai]

It is interesting to consider Massachusetts, which has earned its reputation as one of the most politically liberal states in the Union. That is where the Pilgrim Fathers landed early in the 17th century, running the state for some time virtually as a theocracy. In the 19th century very influential Boston intellectuals (Ralph Waldo Emerson was their spokesman) moved into Unitarianism, which could be seen as a very liberal version of Christianity, though some might deny that it is Christian. Later in the 19th century, Italian and Irish immigrants (John Kennedy's ancestors were among the latter) strengthened the Catholic population of the state, which is now 40% and more.
More than a century ago, Massachusetts politics was dominated by patrician Republicans of the relatively liberal variety. But the new Catholic immigrants voted largely Democratic and the political composition of the state shifted. In the 1950s John Kennedy was elected to the United States Senate in preference to a member of a very old Republican family. Today the state's two Senators (Edward Kennedy and John Kerry) are both Democrats (and Catholics). Ronald Reagan wooed a fair proportion of American Catholics to the Republican party in the 1980s but perhaps not in Massachusetts. Still the state elected Mitt Romney, a Republican and a Mormon, as its Governor a few years ago.

[M.C.]

I think you've found a map of people living in small towns. People in rural areas belong to churches for a lot of reasons besides religious observance. Churches are social, and they may be the only places in town to hear (or participate in) concerts. Groups like Boy Scouts may use the church basement. These factors are less likely to bring people to church in urban areas, simply because urban residents have more options.

As for the Bible Belt, you need to map by denomination to see that. You'll find a lot of Baptists in the Southeast, and that's important. People in the north/central parts of the country are more likely to be Lutherans or some other low-key denomination, and Catholics dominate in a lot of other places. What it means to belong to a church varies a lot depending on what kind of church it is.

[Dr. Keith Caserta]

It's interesting to note that Ohio, which barely went to Bush in the 2000 and 2004 elections, isn't particularly religious. In fact, northern Ohio, which is clearly the more religious section of the state according to the map, voted heavily for Kerry in 2004 while unreligious southern and central Ohio were solidly for Bush. I suspect something else is at play here. In addition, the "soccer moms" in unreligious, exurban counties like Warren (Cincinnati) and Delaware (Columbus) went for Bush.

[Henry Racebrook]

The statistics in the map are misleading, since they are based on reports by 149 cooperating religious institutions. Many Americans attend unaffiliated megachurches and the like, which would account for the lower numbers of reported church membership in places like Alabama, where almost everyone goes to church. And it would also ignore groups like Hindus, which do not have memberships to report.

[Mary Downes]

The result of the 2004 US general election had far less to do with religious adherence than it did "faith-based" electronic voting (put your vote in, pray it's counted correctly).

There's also the secret ritual called 'caging' by which people have their legitimate right to vote raptured away without warning, and the Election Day Passover, a plague that whisks away voting machines only from heavily minority districts which typically vote Democratic.

Charlatans and snake oil salesmen have always been around. The Republican Party has merely perfected the art.

[Tony Francis]

Dan, the map of religion is a little misleading, if one is trying to correlate it with voting patterns. Most blacks (Afro-Americans), especially in the south probably have a large church attendance, but will consistently vote the Democratic ticket. If one looks at the county break-down on the last two presidential elections, most of California, Washington State, Oregon, and Illinois voted "red". It's the cities such as Los Angeles, San Francisco, Seattle and Chicago with large populations which vote heavily Democrat that turns the whole state "blue". Pennsylvania has been described as having Pittsburgh on one end and Philadelphia on the other, with Alabama in-between. This means the big cities vote "blue" and the rest is "red". But the state went "blue" the last two elections. Candadai, you are correct: Massachussetts (Tax-A-Chussetts) is an odd case. Iowa is another funny state: went for Gore in 2000, then Bush in 2004, and only by a few hundreds of votes. Minnesota will probably be "blue" for a long time, but those counties with large church attendance are proabaly red. It's Minneapolis/St. Paul that carry the day.

[Robert]

In response to Mary Downes, its Democrats who tried such tricks as having dead people's votes count, voting twice or more (including overlapping districts in GA...thrown out by the courts) and even to 'estimate' instead of 'count' the US population in the 2000 census. Believe me, as a former census worker, I find it highly suspect that the 2000 population of New York City turned out to be exactly what the politicians wanted it to be...they demanded 8 million and got 8,008,000...a 9.4% increase? No way is that real.

[Ian Goodson]

What interests me is that the palest areas on the map are 'less than 35%'! If we had as much as that in the UK we would consider it a revival.

[CW]

Red & blue are the colors of politics. Using red to denote religion pushes hot buttons. Around 2000, the Ann Arbor {Michigan} News printed some maps of voting paterns over many years. The higher the population density, the more the voting trended Democratic Party. It seems that when locations grow, they need the offerings promised by the Dems. Places with populations spread out can afford to rely on traditional Republican mantras like 'personal responsibliity' etc. In general, the blue leaning areas were metropolitan; red more rural which fits with the necessities of rural life from the past. The MI 7th is blue closer to Ann Arbor and real red in Hillsdale county to the west. I went to a Party for the Planet in Hillsdale (county seat) to see who my fellow subversives;~) were. There were 5 of us. As far as church goes, you can't spit in most places in the US without hitting a church.

[Jared]

The election was so close (differing by 2-3% between Bush and Kerry -popular vote) that the correlation between religion and 2004 voting patterns is rather weak.

Statistics requires larger differences to prove a solid correlation.

[jenny perry]

In the US, we have a diverse population, who belong to equally diverse religions, as well as those who do not.

Unlike, the UK, where there is a willingness to undermine the right of it's citizenry to have religious belief (well, only a certain part of the population) while giving a wink and a nod to another part of the population who believe they have a right to impose their authoritarian religious beliefs on everyone. The US protects the right of all it's citizens to have the right to worship free of government interference, and visa versa.

Then again, historically, the UK was willing to turn a blind eye and to act as a facilitator to the rise of naziism in Europe more than half a century back. No one should be surprised.

[Shivas]

Interesting, the more you move away from the coasts, the more dense the religious people become.

[Scary]

Jenny Perry writes: "the UK was willing to turn a blind eye and to act as a facilitator to the rise of naziism". Is that the same UK that fought alone against the Nazis for 3 and a half years while the rest of the world outside the British Empire capitulated, fought with the Nazis or if they were lucky enought sat on the sidelines? Jenny, you wouldn't be from the U.S. would you? That country was in the last category, waited till Britain was exhausted, then came in when the war was almost won and scooped up all the prizes, ensuring a US hedgemony that is only now fading principally because of the retreat of British-Christian values around the world.

[Knemon]

Snake-oil salesmanship: it's a bipartisan thing.

"If we do the work that we can do in this country, the work that we will do when John Kerry is president, people like Christopher Reeve are going to walk, get up out of that wheelchair and walk again."

John Edwards - October 2004

[Malcolm Parker]

Hmmm, Statistics on religious adherents compiled by people reporting for a society whose justification for existance is the promotion of religion. Could they not possibly be just a little bit biased for this type of thing?

[Nick]

I'd like to think that Shivas doesn't mean intellectually dense, just more of the population proportionately. (However, those on the coasts do tend to think those in the middle are dense, and vice versa.)

Despite the cultural religiosity of the US, honest-to-goodness atheism is staging a revival and perhaps the map is reflective of this, too. Certainly some years ago this map would have been a bit "redder."

[G Austin]

America has been overly blessed by God. we are a stupid people. We do not know where Canada is or that Poland is a country not a chain saw. Our abundance is a gift and we have returned the love in our faith that God guides our destiny. I lived in Europe for many years and sadly see it dying and being replaced by mosques and horrible screaming sounds. Pray.

[Rob]

Henry Racebrook's comment is precisely right. This map's statistics are highly suspect. It ends up telling us more about which institutions cooperated in the survey than where the religious people live. For instance, it looks like the evangelical megachurches didn't participate but the Lutherans did.

[Alex]

Sorry, but these maps don't show a correlation between religiosity and Republicanism, unless that's what you expect to see.

If you want to do this accurately, you should distort the map so that the size of each county is proportionate to its population density.

Then you should break down the states by voting area. Finally, you should abandon the binary red/blue colouring, and adopt shades of purple for divided votes, with at least the same number of distinctions as the other map.

Do that and you'll see a very different picture emerge. For a start, most of your voting map will be within a narrow spectrum of purple. The US is one of the most homogenous countries in the world, and certainly much more than the UK. Say it quietly, though. The NY media types don't like to admit it.

[Clifford Grammich]

I was one of the compilers of the data mapped at the top of the page, and have an abiding interest in religion and politics. Some comments on the comments, in the order in which they were made.

Ismael Klata--It is possible the statistics reported by religious bodies don't change as fast as populations move, i.e., church records may not change as quickly as their members move. Hence church members may sometimes show in places they no longer live. But I'd also admit some of the demographics in places with "depopulation" may be more amenable to church membership.

Candadai Tirumalai--There has actually been very little correlation between % Catholic and % for either party at the county level in recent years. I suspect this is in large part because the Catholic population is so large and assimilated that it simply reflects the electroate at large.

Henry Racebrook--Largely correct, and I'm always somewhat surprised to see maps like this, given that we always note the data are for participating bodies only. Nevertheless, there are some interesting (if predictable) county-level correlations, e.g., Evangelical Protestants for Republicans, Mainline Protestants less so, etc. Overall, we counted about 50 percent of the population, while in surveys 80%+ claim religious affiliation. I suspect the truth is somewhere in between those numbers, but, again, you are correct that our data are an undercount. I'm not sure, however, that "megachurches"--many of which are in our data--comprise as much of the undercount as you seem to imply.

Tony Francis--One of the traditional weaknesses of these data has been in our undercount of predominantly African American bodies. One of the reasons for that is the limited resources those churches have to provide data for enumerations such as this, although the problem alsomay be diminishing as more churches become more integrated. Anyway, regarding the South, I strongly suspect racially polarized patterns of voting overwhelm any religious effects. Regarding California, Washington state, Oregon, and Illinois, there are actually many Republican counties there that have trended more Democratic in recent decades (while remaining Republican overall). The opposite has occurred in the South. I think that has to do with demographics, maybe some shifts in church affiliation, and other reasons many others have cited in these comments.

Malcolm Parker--I'm always interested in improving the accuracy of these statistics, so if you have any evidence of overcounts then I'd be glad to hear about it. I note for the record that I had responsibility for the Catholic county-level enumeration in 2000, with results that showed the population is 22 percent Catholic--in contrast to the 24 to 25 percent Catholic one will see reported in surveys.

Nick--The "revival" of atheism is probably better addressed in survey works (see, for example, some of the work by Michael Hout on persons leaving religion altogether and those who are still religious but not affiliated with a religious body).

Rob--Again, I largely agree with Henry Racebrook, but am not really convinced evangelical megachurches were grossly undercounted.

Alex--I more or less agree, although there might be some more interesting variation in considering the effects of different types of churches on politics.

[Jeff Green]

God clearly lives in Utah!

[Clifford Grammich]

Whoops, one more comment, regarding M.C.--You raise two good points. One, on the type of church influencing one's politics, is about spot on. You're also quite correct about the rural church having to serve many functions that others do in urban areas. I'm not convinced the rural churches can always serve these functions, but that's another story, and, again, there's no question there are fewer alternatives to it in rural areas (meaning a church may fall into roles it wouldn't consider elsewhere).

[Tony Francis]

Clifford, thanks for your comments and clarification. One thing that occurred to me: it is frequently lamented that only 50% (or less) of declared Catholics attend mass every Sunday. Yet people who haven't been to a church in years will still say they are Catholic. In many Protestant churches, some may not attend for months at a time, or only go to one Sunday service a month. Yet they are considered "active" members. Anyway, I think religious belief is deep in the US, whether these people go to church or not. There aren't a lot of atheists "preaching", because their message isn't generally well received. Traditionally, Catholics voted Democratic, but the abortion issue has split many of them away. Right now, Iraq and inflation will probably skew the next election, irrespective of religious affiliation. Comments?

[DAVID TIMMER]

Can anyone (Alex?) explain the relative "secularity" of southern Ohio, eastern Kentucky and western West Virginia according to this map? Based on my own perceptions of these areas, that doesn't fit. (Of course, my perceptions could be the problem.) Another possible source of undercounting religious affiliation besides "megachurches" might be small, independent Baptist and Pentecostal churches that aren't on the list of "149 religious bodies" used in the survey.

[Patrick]

It reminds me of a cute line from the Economist about 15 years ago: statistics should be used as a drunk uses a lamp-post--for support rather than illumination. The trouble with this map is that the more I look at it, the less it tells me. Most of it has to do with intra-religious diversity. Church attendance is produced by a wide variety of factors, including social instrumentalism and communitarianism, as well as religious devotion and evangelicalism. These are the underlying values that rub off on the political culture and social environment of places. Oregon and Southern Ohio look a lot alike on this map, but gosh, those are very very different places for values such as tolerance, multiculturalism, and the like. Southwestern Minnesota looks like a hot-bed of religiosity, but Lake Wobegon-style Lutheranism is a far cry different from the seemingly tepid religious support of the populations in north-central Georgia, coastal South Carolina, and central North Carolina. The posts concerning methodology are interesting and perhaps contain some of the problem, but I think, rather, the fundamental operationalization of the concept we're trying to get at through this data is flawed--the interesting and important regional variation in social and political environment in the U.S. cannot be adequately represented by a measure of aggregate church membership or attendance.

[Clifford Grammich]

Tony Francis--That's a fair point about attendance. Let me note what I think are the two strengths of the data, and then address yours. Strengths of the data: (1) providing insight on specific religious bodies (e.g., Wisconsin Evangelical Lutheran Synod) that would "fly under the radar" of national surveys, (2) providing insight on all religious bodies in local areas that won't have reliable surveys on religious affiliation. As I mentioned yesterday, I'm much more leery of using these data to document "total" religiosity in the United States. Anyway, regarding attendance, in 2000 we did start reporting attendance for religious bodies that had such data. Unfortunately, many don't have this data, or (as in the case of the Catholic Church) don't have reliable data for the entire nation. What a church may consider "active" or even belonging will vary widely, as a cursory comparison of our membership and attendance data will show. The number of Catholics I ultimately produced is close to the proportion of the population who tell survey researchers they are both Catholic and attend church at least occasionally. (About two percent of the population, incidentally, will both claim to be Catholic but say they attend church "never"--accounting for much of the discrepancy between our data and survey data that I noted yesterday.) I'd be quite happy to just get that base number for all religious bodies in all counties across the country before worrying about the varying levels of commitment reflected in it. But you are right to raise the issue. I'm less convinced Catholics "split" from the Democrats over abortion. In fact, looking over the past half-century of survey data on religion and voting, I'm not sure there is much of an overall "Catholic" effect in any election except 1960. Yes, the "Catholic vote" will sway two or three percentage points from one election to another, and some make legitimate points about Catholics' broader attitudes and their politics (see, for example, http://www.uakron.edu/bliss/docs/TheAmericanReligiousLand.pdf). But, overall, I just don't know if the Catholic distinction is all that interesting anymore. I understand why James Carville or Karl Rove might get exercised (or do I mean exorcised, har har), but, again, I just think there are other, more interesting things in American politics.

David Timmer--you are precisely right that our failure to count many "small, independent Baptist and Pentecostal churches" is the problem in southern Ohio, eastern Kentucky, and western West Virginia." With apologies for the self-promotion, I'll note I made a concerted effort to include many of these churches in the 1990 count (and even ended up writing a book about their politics). In one county in eastern Kentucky that I know fairly well, for example, we only counted 7 churches (and their members) in 1980. In 1990, my efforts resulted in 42 churches being counted there. In 2000, I unfortunately didn't have time for these particular churches, and we only ended up counting 21 churches there. My conservative estimate is that there are more than 80 churches in that county--meaning even in our best year we left half the churches there uncounted. An interesting thing about their churches not evident in the maps above is how many of these "small, independent" churches adapt to their communities. Many, for example, have members who are or were in the miners' union, leading to quite different politics than what one sees in some other, larger Baptist churches of the South.

Patrick--Your general point about aggregate data being inadequate for analysis of regional variation is well-taken. I think a far more interesting map with these data is of "denominational" families, showing, for example, whether "Baptists," "Lutherans," "Catholics," etc. are most prevalent by county. (You can see a state-level map using the 1990 data at http://www.adherents.com/maps/US_denom_prot_adh.jpg). I completely agree that "Lake Wobegon-style Lutheranism is a far cry different" from the religiosity of places like "north-central Georgia, coastal South Carolina, and central North Carolina," but, with apologies for going back to the questions of methods, I suspect part of the reason religiosity in those latter places looks so "tepid" is that church records haven't caught up with population migration there. It's also the case that persons moving to such areas from, say, southwestern Minnesota, may change their religious affiliation from Mainline Lutheranism to something more "Evangelical," but that may just be personal speculation.

[Clifford Grammich]

Tony Francis and Patrick--something else that just occurs to me that may interest you is the work of Mark Shibley on "evangelicals" on the West Coast. Shibley makes some very good points about how "evangelicalism" has adapted to the West Coast in ways that one wouldn't think possible in the South. I suppose one could also see this point by comparing the websites, for example, of the late Jerry Falwell's church (http://home.trbc.org/) with that of Rick Warren's church (http://saddleback.com).

[Clifford Grammich]

Whoops again--I just realized the adherents.com map link that I posted earlier is for *Protestant* denominational families. The only such image I can find for all "denominational" families, including Catholics (who dominate in the Northeast, North Central, and Soutwestern regions) is a thumbnail at http://www.glenmary.org/grc/grc_shopping.htm (page down). Nevertheless, I suppose the Protestant denominational family map still makes the point about different regions having different religious influences even if equally "religious."

[Clifford Grammich]

Good grief, in calling the adherents.com map a "Protestant" map I called Mormons "Protestants." I need to quit while I'm behind . . .

[Dominic Shields]

Gosh a gullibility map, how useful for a budding confidence trickster. Anyone want to invest in my perpeptual motion machine?

[Maurits van Eijck]

we all aint got much use for them churches out there in the west country....
I wonder what this map looked like 20/30/40 years ago. Lots more little red blocks I'd guess. Sanity evolves. Now to squeeze the little red centre like a big pimple.

[Nick]

Knemon - as you well know, Edwards was talking about stem-cell research being possible once the bigoted scientific illiteracy of Bush was removed from the White House. This is not snake-oil salesmanship, just optimism that the Dark Ages were coming to an end. Pull the snake out of your own oil!

[Clifford Grammich]

Maurits--or, conversely, the (Mormon) red block currently centered in Utah becomes bigger. I also suspect the light shades of (immigrant) red in the southern half of California will become darker.

[Tony Francis]

Clifford, if I am reading you correctly, you are indicating that the Catholic vote in the US mirrors the general electorate, within a few percentage points. I know that priests and bishops in these churches are quite careful about discussing topics in elections. They will discuss them in a general way, and tell the congregation to vote its conscience. Candidate's names are never mentioned. This is to avoid accusations of "mixing politics of Church and State". It makes me wonder if the Catholic voters are more of an unmovable block than we might have imagined. I am also thinking of predominantly black churches where political rallies are held, and candidates are endorsed.

[Clifford Grammich]

Tony, "unmovable block" might not be my choice of words, although I suppose it gets the point across. I think the more salient one is that some Catholics may interpret their faith politically one way, some may do so another, and some (most?) may just follow other cues.

[Tony Francis]

Clifford, I had been under the impression that many Catholics had switched party affiliation over the abortion issue in the 1970s. You have indicated this is not the case. Is that true? In a similar manner, LBJ predicted that the south would be lost to the Democrats after the civil rights movement of the 1960s. Prior to that, African Americans in the south had been predominantly Republican, in the mode of Lincoln. Thanks for your comments.

[Clifford Grammich]

Tony, I suspect the 1970s were a period of "dealignment" caused by disaffection with the political parties more than anything else.

For example, according to online data for the National Opinion Research Center's General Social Survey, the proportion of the total population identifying as Democrats decreased from 47 percent in 1972 to 38 percent in 1980. The proportion identifying as Republicans increased from 22 to 23 percent, while the proportion identifying as "independents" or with another party increased from 30 to 39 percent.

Catholics weren't much different. The proportion identifying as Democrats decreased from 54 to 41 percent, the proportion identifying as Republicans increased from 15 to 18 percent, and the proportion identifying as independents or with another party from 31 to 41 percent.

Given those numbers (and assuming their accuracy), what would you say the story of the 1970s was for Catholics (who remained more Democratic than the general population in their party identification)? Their switching parties because of abortion? Or their "dealigning" and becoming more independent like the rest of the nation?

I won't deny the influence of abortion on some Catholic voters, but I think it's overemphasized. I don't see how it can be compared to the effect of, say, the civil rights movement on Southern politics.

[Tony Francis]

Clifford, these are interesting points you bring up, and clarify the situation. If I recall the numbers correctly, George Wallace got a large percentage of the 1968 vote, and a good number of electoral votes. Perhaps this was the beginning of the realignment you are writing about. This was before the white south went from Democrat to Republican. I think in those days, there were lower numbers of "northerners" in cities such as Atlanta, Birmingham, Miami, etc. Miami may be a special case because of the Cuban population.

[Kurt]

I really find it hard to believe that church membership data collected from church organizations is at all reliable; it seems near certain that overstatements of membership would result. I just think it's dangerous to rely exclusively on data from sources that stand to gain from certain results.

Also, maybe I'm just hanging out with the wrong crowd, but, as a long time resident of Chicago and a past student of both Northwestern U. and the University of Chicago, I have a very hard time grasping the '50-75%' figure for the Chicago area. It has been several years since I have met an actual believer...

[Clifford Grammich]

Kurt, as noted above, I'm always interested in improving the accuracy of these statistics, so please let me know of any evidence you have of overcounts. Regarding the Chicago area, of which I'm a long-time resident (including of Hyde Park, though not since my days at the U of C), the Catholic population alone, according to several sources, accounts for about 40 percent of the population. (Maybe not in Hyde Park or Evanston, but probably more than that in, say, Bridgeport or Beverly.)

[Clifford Grammich]

Tony, in the 1964 election, shortly after LBJ signed a civil rights act, Barry Goldwater carried but six states: five in the Deep South and (barely) his home state of Arizona. That might have been the beginning of the Southern realignment. Or maybe it stems to the Dixiecrat ticket of '48. Regardless, as you note, migration has surely played a large part as well.

[Stan(expat in USA)]

"The Republican Party has merely perfected the art.
Posted by: Mary Downes"

Learnt from the Democrats in 1960 ?

Electronic voting, mail voting etc there will always be fraud on all sides, the trick is to get turnout high enough to render the fraud insignificant.

In 2004 Bush's share of the popular vote was the highest for any party since his father's election as President.

The Dems loss in 200/2004 is entirely due to their poor choice of candidates.

[Mr. King]

I love how the South is always labelled and stereotyped as overly religious and yet this graft shows the Northeast (who everyone portrays as extremely liberal - e.g., New York) has a higher percentage of religious adherents than the south.

[Eric LaRue]

According to the two maps, religion and politics isn't one-to-one. Massachusetts is shown as being more religious than Arizona, yet Arizona is a red state while Massachusetts is solid blue.

[Doug Routledge]

I would love to see a contrasting or comparative map of economic health county by county. I am in one of those pale communities and I believe, subsequently one of the poorest.