Estimating Religion from Census Data

[RI]

As an extension of my world religion map and in light of the fundamental flaws in ARDA data, I'm attempting to estimate religious affiliation based on only data from the Census/ACS with an eye on finding an adequate model to project to sub-county areas. The single most useful figure available is ancestry data, to which I've attempted to assign each of ~190 ancestries an approximate religious distribution. In light of the recent Pew results, I've tried to create a model that matches their state-by-state results. I also used their results to identify a possible proxy for the variance in the nonreligious distribution: white, never-married males.

The model I've come up with isn't perfect, but I think it's pretty good so far. Here's a county map for the contiguous US:



Compare to this map based on ARDA data:


I'm aware of a few issues, such as not yet having a great proxy for the Jewish population, but if there are any other glaring issues, let me know. When it's adequate, I'll try making some sub-county maps.

[All Along The Watchtower]

Very curious to see where this goes! Though, be careful with ancestry; to take just one example,  German-Americans in the Milwaukee region are different from German-Americans in Missouri, who are different from German-Americans in Texas or Pennsylvania or Oregon or...

Also, watch out for "Irish" ancestry in the South and (much of) Appalachia. Not gonna be like Irish ancestry in New England or NYC. Let's leave that at that. 

[publicunofficial]

Does teal/green represent unaffiliated? If so, I'm surprised at Kittitas County in Washington.

[RI]

Does teal/green represent unaffiliated? If so, I'm surprised at Kittitas County in Washington.

Yeah. I think my method may over estimate the non-religious population in college areas. Here's a census tract map for Washington using this method; as you can see, Kittitas being nonreligious is entirely due to Ellensburg and CWU's influence.





Generally, it seems fairly accurate to me. It might overestimate the nonreligious population near military bases (see Fort Lewis and Naval Station Whidbey).

[publicunofficial]

That would also explain Whitman County, as well as Albany County in Wyoming

[RI]

I think the Pullman area may very well be nonreligious, but I don't think it's as nonreligious as my map shows. However, to get Washington to be 32% nonreligious, you need a lot of those types of areas to show up as green.

[publicunofficial]

I think the Pullman area may very well be nonreligious, but I don't think it's as nonreligious as my map shows. However, to get Washington to be 32% nonreligious, you need a lot of those types of areas to show up as green.

I can believe the alumni of WSU swinging Whitman, it's a pretty empty county outside of Pullman. But does CWU really swing Kittitas? CWU isn't that large of a school.

Whatcom County is also interesting. Bellingham is extremely non-religious, even outside of the WWU campus. But it gets counter-acted by the strong Dutch Reformist presence in Lynden, where all the businesses still close on Sundays.

[Asherzeleg]

Fascinating! Are non-Christian religions being included?

[RI]

Fascinating! Are non-Christian religions being included?

Yes, but they don't have pluralities outside of a few major cities.

[Sol]

Yeah, there's noway that non-religious is a plurality in Watauga County.

[bgwah]

Very cool, and it seems more accurate than the ARDA map. How Catholic are you saying Hispanics are? I would have expected places like East Pasco to be darker shades of red.

Also, do you have a color key?

[RI]

Very cool, and it seems more accurate than the ARDA map. How Catholic are you saying Hispanics are? I would have expected places like East Pasco to be darker shades of red.

Also, do you have a color key?

Hispanics in eastern Washington tend to be more LDS than in other places.

The key is standard Atlas shades (>20%/>30%/etc.) with Red being Catholic, Blue being Protestant, Green being Non-religious, and Yellow being Mormon. The grey area in Benton didn't have any people reporting an ancestry.

Yeah, there's noway that non-religious is a plurality in Watauga County.

Yeah, probably not. As I said, college towns are a bit of a problem, but I'd imagine Watauga isn't hyper-Protestant either. It should probably be similar to the shading of Orange County, NC.

[Sol]

Very cool, and it seems more accurate than the ARDA map. How Catholic are you saying Hispanics are? I would have expected places like East Pasco to be darker shades of red.

Also, do you have a color key?

Hispanics in eastern Washington tend to be more LDS than in other places.

The key is standard Atlas shades (>20%/>30%/etc.) with Red being Catholic, Blue being Protestant, Green being Non-religious, and Yellow being Mormon. The grey area in Benton didn't have any people reporting an ancestry.

Yeah, there's noway that non-religious is a plurality in Watauga County.

Yeah, probably not. As I said, college towns are a bit of a problem, but I'd imagine Watauga isn't hyper-Protestant either. It should probably be similar to the shading of Orange County, NC.

Hahaha no.

[True Federalist (진정한 연방 주의자)]

At first I was surprised Maryland had no Catholic counties, but then I realized African Americans don't tend to be Catholic.

It's more the case that despite being established for Catholics, Maryland was mainly Protestant even in colonial times.

[bgwah]

Very cool, and it seems more accurate than the ARDA map. How Catholic are you saying Hispanics are? I would have expected places like East Pasco to be darker shades of red.

Also, do you have a color key?

Hispanics in eastern Washington tend to be more LDS than in other places.

The key is standard Atlas shades (>20%/>30%/etc.) with Red being Catholic, Blue being Protestant, Green being Non-religious, and Yellow being Mormon. The grey area in Benton didn't have any people reporting an ancestry.

I'm just using East Pasco as an example, but it is around 85% Latino if I recall correctly, but you have it only >30% Catholic. That doesn't seem like enough. I'm not saying you're wrong - just curious what inputs you used and where you are getting this data (such as the LDS fact).

Are you using a universal formula for the entire country, or by state? County?

I am only nitpicking because I really like this, btw.