Pella, IA has a population of 10, 352 according to wikipedia. With 28 churches according to yahoo, and with a local claiming there is 42.
Go ahead, I know what it must be, but let’s hear your version.
Page, Arizona might be in the discussion. It’s in Mormon country so there are a fair amount of LDS churches. Page is on the edge of an Indian reservation and for whatever reason there are a bunch of small churches of multiple denominations that sit just on the outside of the reservation. I’ve always assumed there was some proscription against churches on the res so they set up just outside.
How about Zanesville OH ? churchangel.com lists 120 churches for a population of approx. 25,000
Are you sure? I thought that Colma, CA was No. 1-the whole town is a cemetery!
I went to college in Turlock (Turkey Tech) back in 1970. At that time there was a ton of churches in Turlock and exactly 2 liquor stores inside city limits.
I live there now. This page lists 56.
And there’s a lot more than 2 liquor stores. 
Suppose the numbers quoted are correct. I’m going to round to the nearest thousand on the population. Here are the numbers of churches per person:
Zanesville, Ohio has 120 churches and a population of 25,000, so it has a church for each 208.3 people.
Turlock, California has 56 churches and a population of 69,000, so it has a church for each 1,232.1 people.
Pella, Iowa has 42 churches and a population of 10,000, so it has a church for each 238.1 people.
Wheaton, Illinois has 46 churches and a population of 55,000, so it has a church for each 1,195.5 people.
I don’t think any of those cities hold the record. There are lots of small towns with 300 inhabitants and a couple of churches. This is because the church members mostly live in the farms nearby. This is an easy way to make a mistake when doing these calculations.
I remember some numbers quoted years ago. These numbers said that big city X had m men from there who were killed in the Vietnam War and small town Y had n men from there who were killed in that war. If you did the division, you could see that it must mean that, per capita, 10 times as many were killed from Y as from X. But the numbers were wrong. The number of men killed in the war who were supposedly from Y were in fact all those from the school district around Y, not necessarily those from Y itself. The school district had many more people living in it than just those living in the town itself. If you did the division using the number of people who lived in the school district around Y, not the town of Y, you got a number of men killed in the war per capita that was much closer.
The record hold for most churches per capita is almost certainly some small town. To do this right, you would have to split cities up into divisions according to their population. Then you could say that, for a city with more than a given population, they have the most churches per capita. Comparing cities of wildly different populations is worthless since the churchgoers in small towns largely come from the nearby farms.
I’ve been to weddings in a couple of tiny Texas towns like that- Walburg and Frelsburg, and neither is probably over 300 people, and both have 2-3 churches.
I thought the number of churches in Zanesville, Ohio (120, supposedly) was suspiciously large for a city with only a little over 25,000 people. I looked at the website that george klavins claims to have used to find the number of churches there. The number is 113 according to one place in the website but 108 according to another place on the same website. But it’s worse than that. I wondered if the churches that are supposedly in Zanesville are really inside the city limits or if many of them are outside it. The answer is that some are certainly outside it. Look at the map for one such church listed as being in Zanesville:
http://www.churchangel.com/map/meadow-farm-united-methodist-church-zanesville-oh-43701.html
That’s quite far from the city. So there’s no way to use that website to tell how many churches are in a given city unless you click on the individual churches and find their actual locations. The area with a Zanesville mailing address may have 108 or 113 or 120 or whatever churches, but that includes a lot of the surrounding rural area. This is an important thing for people who’ve never lived outside a big city to remember. The number of people with an address of X, where X is a small town, could be five or even ten times as many people as the number who live within the city limits.
Yes, the population of Zanesville’s 43701 ZIP code was more than 55,000 in 2010.
By the same measurement, the village of South Zanesville has no churches at all, as there are no churches with a South Zanesville mailing address. (It actually has at least three.)
There is another reason why it’s likely to be a small town: small towns have cheap real estate and that allows for smaller churches. Churches often have doctrinal or social disputes. In a small town, 50 people can split off and open their own church in what used to be the Chicken Shack or whatever. Small town churches can be primitive: pretty much a roof and four walls is the only requirement. And the preacher is often someone with a regular job who preaches on Sunday. The logistics are just too complicated for the same kind of splitting in a city: you need a couple hundred people, at least, to support an institution. So dissenters find a different, existing, church, or they just keep grumbling.
Edna Bay AK, population 42, 1 church