There are so many blind spots in survey research that it’s impossible for the average person to understand how much we just don’t know and can’t know with the current repositories of data about religion. Most surveys that do ask questions about religious attendance, belief, or affiliation are focused on capturing a snapshot of where things are right now. They don’t ask about the religious affiliation of someone when they were growing up, or how often they went to church.
The end result is that we are very well equipped to understand the share of Americans who identify as Catholic, atheist, or Hindu. However, we don’t really have a very good handle on how they got there. Did that person who told us that they are currently an agnostic always espouse that affiliation? Or did they grow up in a highly religious, evangelical household and just leave it behind at some point? We know who they are now - but the rest of their religious history is a mystery.
How Many People Leave Their Childhood Religion?
Being an academic is an exercise in the absurd. That’s what I’ve realized after doing this for the better part of two decades. For instance, I have devoted thousands upon thousands of hours to trying to untangle a very difficult puzzle: when people are asked about their current religion on surveys, what is the mental process they go through to arrive at…
This, by the way, is one of the central problems that Michael Graham and I are trying to solve in a two-book series that is going to be released through Zondervan in 2027 and 2029. We currently have a survey in the field that asked an incredibly detailed set of questions about the religious history of every respondent. We want to know how much they went to church and what kind of house of worship at every single life stage. The number of avenues for research from that data is nearly limitless.
But there are some other datasets that can help shed some light on this type of question right now. A recently released one is the Pew Religious Landscape Survey. It was fielded in 2023 and 2024 with a total sample size of about 37,000. It asks a nice little battery of questions about the childhood religion of each respondent. Let me start by showing you how two questions interact - the individual’s religious importance when they were growing up and the importance their family placed on religion.
The most frequently chosen option in this sample is the top right - these are people who say that religion was very important to them as a child and it was also very important to their family. About a quarter of all Americans are sorted into this category. But there are other squares that contain a significant share of the sample, too. For instance, 16% of people said that religion was somewhat important to them and their family in their childhood and nearly the same share said that their family was very religious when they were a kid but they thought that faith was just somewhat important. That’s over half the sample in just those three squares.
You know where you don’t find a lot of folks? In the bottom left - these are people who grew up largely irreligious. Just 6% of all people taking the survey said that religion was not at all important to them as a child and was not important to their family either. In fact, just 8% of the sample said that religion was not at all important to their family when they were growing up and just another 16% said it was “not too important” to their family. That means that about three-quarters of folks grew up in an environment where their family said religion was somewhat or very important to them.
Of course this has to have shifted over time, right? To see how these percentages look different for older respondents compared to younger ones, I broke the sample down into decade of birth and did the same bit of analysis.