Church Attendance Rate Maps

[RI]

I've been working for a few years now on a project using smartphone data to map church attendance in the US. Here's a preliminary map showing the percentage of people who I have identified as attending a church by county during the period of August 2016 to January 2017:

Image Link

I can also post census tract level maps for individual counties by request. Previous data that exists has some severe methodological issues, so I think this is a one-of-a-kind map.

There are a ton of interesting things about this map, some of which confirm some long-held suspicions. A few observations:

1) Church attendance is not actually that high in the South outside of suburban counties. In general, this map correlates well with social capital, which also peaks in the upper Midwest and the Mormon Belt, and the middle of the income spectrum. Dutch Reformed areas also stick out.

2) Suburbs in general have more churchgoers than elsewhere, but especially in some of the anomalously Republican suburbs around Milwaukee and Indianapolis.

3) There seems to be a bit of relationship with Ted Cruz's primary share in 2016 (see Illinois, for example).

I'm sure you can find some other interesting things on the map.

[RFayette]

Thank you so much for this!

[xelas81]

I've been working for a few years now on a project using smartphone data to map church attendance in the US. Here's a preliminary map showing the percentage of people who I have identified as attending a church by county during the period of August 2016 to January 2017:

Image Link

I can also post census tract level maps for individual counties by request. Previous data that exists has some severe methodological issues, so I think this is a one-of-a-kind map.

There are a ton of interesting things about this map, some of which confirm some long-held suspicions. A few observations:

1) Church attendance is not actually that high in the South outside of suburban counties. In general, this map correlates well with social capital, which also peaks in the upper Midwest and the Mormon Belt, and the middle of the income spectrum. Dutch Reformed areas also stick out.

2) Suburbs in general have more churchgoers than elsewhere, but especially in some of the anomalously Republican suburbs around Milwaukee and Indianapolis.

3) There seems to be a bit of relationship with Ted Cruz's primary share in 2016 (see Illinois, for example).

I'm sure you can find some other interesting things on the map.

What about people without smartphones who goes to a church?

[Storr]

The low attendance rates in stereotypically religious Appalachia stuck out during my first glance.

[afleitch]

That's actually a really valuable piece of work that you should try and share to bigger companies (perhaps Pew?) who might be able to add to it.

[Former President tack50]

So basically, Church Attendance is directly correlated with income? (ie rich people go to church, poor people do not)

[RI]

So basically, Church Attendance is directly correlated with income? (ie rich people go to church, poor people do not)

Yes, very much so (but with some important exceptions), which is also pretty commonly seen in survey data. I recall a post by ExtremeRepublican which talked about the income-church attendance relationship pretty well. I believe the idea is that religiosity generally declines with income, but conditional on being religious, church attendance rises with income. Someone who is poor and religious likely doesn't attend church that often, if ever, whereas higher income religious individuals generally attend church quite often. This somewhat explains the Appalachia data, which can also be partially explained by the link between church attendance and social capital, which is also very low in Appalachia.

[Stranger in a strange land]

The low attendance rates in stereotypically religious Appalachia stuck out during my first glance.

Contrary to what a lot of media would have you believe, the most religious folk in the South are Middle Class & above, not the "trailer trash" or "poor white trash" or "hillbillies" or what have you.

[TJ in Oregon]

So basically, Church Attendance is directly correlated with income? (ie rich people go to church, poor people do not)

Yes, very much so (but with some important exceptions), which is also pretty commonly seen in survey data. I recall a post by ExtremeRepublican which talked about the income-church attendance relationship pretty well. I believe the idea is that religiosity generally declines with income, but conditional on being religious, church attendance rises with income. Someone who is poor and religious likely doesn't attend church that often, if ever, whereas higher income religious individuals generally attend church quite often. This somewhat explains the Appalachia data, which can also be partially explained by the link between church attendance and social capital, which is also very low in Appalachia.

I'm not necessarily disagreeing, but isn't it also possible that bringing a smartphone with you correlates to income instead (or at least the number of smartphones a family brings)?

For example, I could see wealthy suburban families going to church with each member also carrying a smartphone at unusually high rates.

[Sprouts Farmers Market ✘]

The New York metro is pretty impressive! I know it's immigrant-driven, but that's a solid percentage - lot of non-Hispanic immigration there too plus the lapsed incumbents of the areas circling Manhattan.

Chicago is significantly lagging, which is not something I would have guessed.

[they don't love you like i love you]

Worth noting the numbers could be skewed by people who attend church in a different county than they live in, which I imagine is somewhat common in Appalachia. Might also boost suburban counties a bit too.

[RI]

Worth noting the numbers could be skewed by people who attend church in a different county than they live in, which I imagine is somewhat common in Appalachia. Might also boost suburban counties a bit too.

The data is based on where the person resides.

So basically, Church Attendance is directly correlated with income? (ie rich people go to church, poor people do not)

Yes, very much so (but with some important exceptions), which is also pretty commonly seen in survey data. I recall a post by ExtremeRepublican which talked about the income-church attendance relationship pretty well. I believe the idea is that religiosity generally declines with income, but conditional on being religious, church attendance rises with income. Someone who is poor and religious likely doesn't attend church that often, if ever, whereas higher income religious individuals generally attend church quite often. This somewhat explains the Appalachia data, which can also be partially explained by the link between church attendance and social capital, which is also very low in Appalachia.

I'm not necessarily disagreeing, but isn't it also possible that bringing a smartphone with you correlates to income instead (or at least the number of smartphones a family brings)?

For example, I could see wealthy suburban families going to church with each member also carrying a smartphone at unusually high rates.

Something like 80% or more of American adults own a smartphone, so I'm not hugely concerned about that side of things. It's possible, but my experience suggests that poorer families don't really have fewer smartphones than richer ones, and, if anything, poorer families might be more likely to actually bring their smartphone into church than a wealthier one (e.g. to use as a distraction for kids, etc.).

[Associate Justice PiT]

Something like 80% or more of American adults own a smartphone, so I'm not hugely concerned about that side of things. It's possible, but my experience suggests that poorer families don't really have fewer smartphones than richer ones, and, if anything, poorer families might be more likely to actually bring their smartphone into church than a wealthier one (e.g. to use as a distraction for kids, etc.).

     Is it possible that there is geographic variation in the distribution of smartphone ownership?

[RI]

Something like 80% or more of American adults own a smartphone, so I'm not hugely concerned about that side of things. It's possible, but my experience suggests that poorer families don't really have fewer smartphones than richer ones, and, if anything, poorer families might be more likely to actually bring their smartphone into church than a wealthier one (e.g. to use as a distraction for kids, etc.).

     Is it possible that there is geographic variation in the distribution of smartphone ownership?

All of our validation measures suggest that smartphones are not spatially distributed differently from the general population relative to any important demographic or political dimension.

[True Federalist (진정한 연방 주의자)]

Worth noting the numbers could be skewed by people who attend church in a different county than they live in, which I imagine is somewhat common in Appalachia. Might also boost suburban counties a bit too.

The data is based on where the person resides.

So basically, Church Attendance is directly correlated with income? (ie rich people go to church, poor people do not)

Yes, very much so (but with some important exceptions), which is also pretty commonly seen in survey data. I recall a post by ExtremeRepublican which talked about the income-church attendance relationship pretty well. I believe the idea is that religiosity generally declines with income, but conditional on being religious, church attendance rises with income. Someone who is poor and religious likely doesn't attend church that often, if ever, whereas higher income religious individuals generally attend church quite often. This somewhat explains the Appalachia data, which can also be partially explained by the link between church attendance and social capital, which is also very low in Appalachia.

I'm not necessarily disagreeing, but isn't it also possible that bringing a smartphone with you correlates to income instead (or at least the number of smartphones a family brings)?

For example, I could see wealthy suburban families going to church with each member also carrying a smartphone at unusually high rates.

Something like 80% or more of American adults own a smartphone, so I'm not hugely concerned about that side of things. It's possible, but my experience suggests that poorer families don't really have fewer smartphones than richer ones, and, if anything, poorer families might be more likely to actually bring their smartphone into church than a wealthier one (e.g. to use as a distraction for kids, etc.).

I'm glad to be part of the 20% or less. That said, dismissing income and other effects that might keep people from going to church with a smartphone turned on and tracking location without some surveying to prove it isn't skewing the data doesn't seem very wise.

[Basil]

Could you do Lorain and Cuyahoga in Ohio please? I want to see if it confirms my suspicions.

[Amenhotep Bakari-Sellers]

Covid 19 was an adjustment to the system. Patrons have the freedom to worship online and not subject to Churches getting them into the Church for membership purposes. I worship at home now

[Tintrlvr]

The low attendance rates in stereotypically religious Appalachia stuck out during my first glance.

Contrary to what a lot of media would have you believe, the most religious folk in the South are Middle Class & above, not the "trailer trash" or "poor white trash" or "hillbillies" or what have you.

It depends what you mean by "most religious". Certainly the ones most bound up in church learning and religious hierarchy to the point of attending church regularly are middle class and above. I think saying that constitutes being "more religious" is somewhat dismissive of what you might call "modern folk religion" that is quite common with poor people in places like Appalachia who are untethered from institutional churches but nonetheless have strong religious beliefs (predominantly Christian or at least institutional-Christian-adjacent, of course).

[Starry Eyed Jagaloon]

Very interesting that the most religious part of the South is the Piedmont--not the coast, Black Belt, or Appalachia. Anybody have an explanation for this?

[True Federalist (진정한 연방 주의자)]

Very interesting that the most religious part of the South is the Piedmont--not the coast, Black Belt, or Appalachia. Anybody have an explanation for this?

At least part of this is due to the uncritical use of phone geolocation data as the means of measurement.

There's also the potential issues of whether rural church locations have been adequately mapped, and whether urban churches are getting a higher rate of false positives from people driving by on their way to somewhere other than church, or simply to another church.

[Del Tachi]

Very interesting that the most religious part of the South is the Piedmont--not the coast, Black Belt, or Appalachia. Anybody have an explanation for this?

Church attendance is correlated with social capital, especially in the South.

[Gass3268]

New York City and the surrounding area is much higher than I would have expected.

[RINO Tom]

New York City and the surrounding area is much higher than I would have expected.

I mean, this is ENTIRELY anecdotal, but I always stereotype your average Midwestern Mainline Protestant like myself and a lot of my friends growing up as less avid church goers than the largely Italian Catholics I have met from New York over the years.

[RINO Tom]

RI, how difficult would it be for you to estimate church attendance by (broad) denomination using this map?  I'd be very interested to see how Catholics, Evangelicals and various Mainline denominations compare.  Anecdotally, growing up ELCA Lutheran, I feel a lot of Mainliners tend to be part of the "C&E Club" and not attend church often, though I will say a lot would still identify as being Mainline Protestant.

[FEMA Camp Administrator]

RI, how difficult would it be for you to estimate church attendance by (broad) denomination using this map?  I'd be very interested to see how Catholics, Evangelicals and various Mainline denominations compare.  Anecdotally, growing up ELCA Lutheran, I feel a lot of Mainliners tend to be part of the "C&E Club" and not attend church often, though I will say a lot would still identify as being Mainline Protestant.

Depends on how he stored/recorded data on individual churches. DK how this would work in interaction with the cell phone data but I imagine it's matched with a list of locations/coordinates. If that sheet already includes denomination, it shouldn't be that hard to filter, though rendering that # as a percentage of a hypothetical larger body of believers might be difficult. Maybe census info could provide the raw numbers of "adherents" in each denomination to compare attendance numbers with.