Do Educated People Believe in God More or Less?

Here’s a lesson that I’ve learned over the last couple of years: if the first book you write gets any traction at all, you will be remembered for that work for years to come. Not that it’s a bad thing. I am still giving presentations to different groups that use graphs that I first put together over five years ago. People want to read The Nones and talk about it. That’s truly a blessing. I think I speak for most academics when I say that I’m just glad that anyone wants to read what we spend years writing.

The second chapter of that book is the one that I’m the proudest of when looking back. I try to answer a seemingly simple question: what caused the nones to rise? It’s obviously not a simple question to answer. There are dozens of reasons. But the first one I point to is probably the most foundational: secularization theory. In short, it’s this simple idea:

As a country becomes more educationally advanced and economically prosperous, it will become less religious.

I am most certainly not going to take readers down a very dry and boring academic explication of the architects and nuances of secularization theory. All I’m going to say is this: Western Europe is a pretty strong case study of secularization theory.

I’ve written a lot about the impact that education has on religiosity in a bunch of different pieces. When it comes to religious attendance in the United States, there’s basically no empirical evidence to point to the fact that education makes one less likely to attend. If anything, there’s a decent case to be made for the opposite: educated people are the most likely to report weekly religious attendance.

But that is, of course, only one facet of religiosity. One I haven’t really touched on a whole lot yet is religious belief. It’s probably the hardest one to poll on in surveys because it’s such an amorphous concept. I don’t want to get too bogged down in the minutiae of that methodology discussion, so let’s just cut to the chase. I am going to use the General Social Survey’s question about belief in God. They first asked about it in 1988, but only included in every survey since 2006. Here’s how it’s setup.

Please look at this card and tell me which statement comes closest to expressing what you believe about God.

1. I don’t believe in God

2. I don’t know if God exist and I don’t believe there is any way to find out

3. I don’t believe in a personal God but I do believe in a Higher Power of some kind

4. I find myself believing in God some of the time, but not at others

5. While I have doubts, I feel that I do believe in God

6. I know God really exists and I have no doubts about it.

Here’s how responses to those questions have changed since 1988.

This graph isn’t great, to be honest. This is a big problem in data visualization: one response option is clearly much more popular than the other ones. It throws off the scale for all the others. But you can get the gist here: the certain belief in God number was incredibly high in 1988—over 60%. It’s clearly declined a bunch since 2000. Now it’s about 50%. The other lines don’t look like they are moving up or down too dramatically. But let me visualize this in a (hopefully) more helpful way.

The share of Americans who said that they didn’t believe in God was 1.5% in 1988. Today, it’s not quite 7%. That’s a pretty substantial jump, but it’s worth pointing out that the share of Americans who are atheist by belief is not that large. The agnostic story is fairly similar: about 3.6% said that we couldn’t know if God exists in 1988. Today, it’s about double that rate—7.4%. In 1988, about 5% of Americans were atheist or agnostic by belief. Today, it’s about 14%.

The share of the public who believes in some general Higher Power has also doubled during this time period from 7.4% in 1988 to 13.3% in 2022. The next two categories are pretty flat. Combined, they were 24% of the sample in 1988. In the 2022 data, it’s 23%. Not much change there.

The share of Americans who expressed a certain belief in 1988 was 63%. It declined to just about half of the sample in 2022. Despite this numerical decline, it’s still very much the most popular response option, though. Today, there are over three times as many people with a certain belief in God than those with an atheist or agnostic belief.

Let’s add the education piece to the mix now. I calculated the share who expressed an atheistic belief in God by four levels of education. I did that for every wave of the GSS since 2006.

There’s an inherent problem when doing this kind of analysis: there just aren’t that many people who pick this option on the belief question. What that means practically is that it’s hard to get to statistical significance when the baseline is so low to begin with. For instance, in the 2006 data, this response option was chosen by 2-3% of the sample, regardless of education. However, I do think it’s fair to say that the trend line does move upward: those with graduate degrees are more likely to have an atheist belief, but it’s not a huge increase in most years.

Looking at this data in most years doesn’t really tell a compelling story. In 2016, there’s absolutely no relationship between education and religious belief. That’s also the case in 2018. In the last couple of years, I think it’s fair to say that there’s a statistically significant difference in atheist belief among those with graduate degrees compared to folks with a high school diploma or less. But the size of the effect is small—a three percentage point difference in the two estimates.

Let’s expand our dependent variable to those who selected the atheist or agnostic belief option.

Okay, I think this does bring things into sharper focus. In some years, there’s a clear stair-step in the trend line. Higher levels of education lead to a greater likelihood of having an atheist or agnostic view of God. But it’s certainly not consistent across every survey, especially in the last couple of years. In 2021, there was no difference in belief among those with some college credits compared to those with a graduate degree. That was also the case in 2022, too. I think it’s generally (but not universally) true that education is positively related to an atheist or agnostic view of God.

But let’s put this to a more rigorous test. I combined the data from 2006 through 2022 so that I had a really nice sample size. Then I analyzed the relationship between belief and education.

The top row is just those with an atheistic view of God. As can be clearly seen, if there’s a relationship between those two variables, it’s an incredibly weak one. About 6% of those with graduate degrees say that God doesn’t exist compared to 4% of folks with no more than a high school diploma. Certainly not a substantively large difference.

However, the difference does become much larger when looking at the second graph which is folks with an atheist or agnostic view of God. It was 7% of folks with no more than a high school education. But as education increases, so does the percentage of folks with no belief in God. Among those with a graduate degree, it’s 16%. That’s a doubling in the share from the bottom of the education spectrum to the top. That’s certainly notable.

What about the share who express a certain belief in God? This is clear: those with lower levels of education are more likely to say that they believe in God without any doubt—it’s 64% of those with a high school education or less. Among those with a graduate degree, it drops to just 45%. That’s nearly a twenty-point decline from left to right. There’s no doubt about the takeaway here: educated folks tend to be less religious on this belief metric.

But let me point something out. Even among people who have graduate degrees, they are still three times more likely to say that they have no doubt of God’s existence than to express an atheist or agnostic view of God. Certain religious belief is still incredibly robust in the United States, even if it has declined somewhat over the last 35 years.

Code for this post can be found here.