It’s something that I’ve said before during presentations that felt right, but I wasn’t 100% sure, “Agnostics are a light version of atheists.” Agnostics seem to get overlooked when it comes to talking about the nones. I know that when I’m writing about the extremes of American religion, I tend to focus on atheists the most. And, in evangelical media circles, there’s never an agnostic philosophy professor – it’s always an atheist. So, are agnostics just a slightly more religious, slightly less liberal version of atheists? I dug through some data and I think I can say that the answer is pretty clear – “yes.”
A quick aside about the theological differences between the two groups. Atheists, by definition, believe that there is no Higher Power. They contend that everything in the world has scientific explanations and not Divine ones. Agnostics are a bit more ambivalent about that. While atheists state, “There is no God,” agnostics would say that they don’t know if God exists and there’s no way to prove that either way. The term agnostic was coined by Thomas Henry Huxley in 1869, when he stated “(agnostic) simply means that a man shall not say he knows or believes that which he has no scientific grounds for professing to know or believe.”
Let’s compare those two groups on the religious questions that exist on the Cooperative Election Study to get a sense of their theological differences.
When asked how important religion is to their lives, 92% of atheists say “not at all” while another five percent say “not too.” Agnostics are a bit more ambivalent with 74% saying “not at all” and 20% saying “not too important.” When it comes to church attendance, the same general pattern emerges – neither group goes to services that much but atheists are even less apt to admit to any church attendance (88% say that they never go vs. 72% of agnostics). Finally, when it comes to prayer, the gap grows larger. Over nine in ten atheists say that they never pray, but it’s only six in ten agnostics. Another 26% of agnostics say that they seldom pray.
Taken together, I think it’s accurate to say that neither group is particularly religious – which makes sense given that they are atheists and agnostics. However, it’s also fair to say that agnostics are more religious, especially when it comes to prayer. An atheist is fifty percent more likely to report never praying than an agnostic. Thus, saying that agnostics are a slightly more religious group than atheists is borne about by this data.
How about politics? A good place to start is tracing the vote choice of these two groups across the last four presidential elections.
The atheist vote is one of the bluest in American religion (behind only Black Protestants). Eight-two percent of atheists backed Obama in both 2008 and 2012. However, Clinton didn’t enjoy that level of support – she only garnered 78% of the atheist vote. That’s not because atheists like Trump, by the way. The cause of this drop was a lot more third-party voters. But, then Biden did incredibly well with atheists, gathering 85% of the vote.
Agnostics are a very Democratic leaning group, as well. But, they aren’t as blue as atheists. Obama got 80% and 77% of the agnostic vote, but that big dip for Clinton appears among agnostics, too. Clinton did nine points worse than Obama’s reelection effort. But, there’s some evidence that the cause was not just third party votes, but that some agnostics supported Trump. By 2020, agnostics looked almost identical to their 2012 vote shares.
The fact that atheists are a bit further to the left side of the political spectrum also appears when the mean partisanship and ideology of atheists and agnostics is plotted from 2008 through 2020.
There’s no doubt that atheists are both more liberal and more Democrat than their agnostic counterparts. There’s also ample evidence here to indicate that both groups have shifted further to the left in terms of both political partisanship and ideology since 2008. The average agnostic in 2020 is where the average atheist was back in 2008. It’s also worth pointing out that these groups are moving in both dimensions. It’s not like one is becoming more liberal, but not more Democratic. So, politically, it’s also accurate to say that agnostics are just a less extreme version of their atheist cousins.
But, do these differences persist in a basic regression model? I put one together that tried to predict an atheist affiliation and an agnostic affiliation and included the usual suspects of control variables. Factors like education, income, age, gender, church attendance, region, and political ideology were modeled. Interpretation of this coefficient plot is straightforward – if the circle is to the right of the dashed line, it means that the variable predicts a greater likelihood of that identity. If it’s to the left, that’s a lower likelihood. But, if the circle or the horizontal lines intersect with zero, there’s no statistical relationship.
It’s noteworthy that none of these variables work in different directions for atheists or agnostics. Generally speaking, the foundational aspects of an individual predict atheism or agnosticism in the same way. However, there are instances where the magnitude is different.
For instance, being male is more predictive of identifying as an atheist than it is being an agnostic. The same is true for income. Higher incomes make one slightly more likely to be an atheist. But there are two variables where the gaps are much bigger. Lower church attendance is more predictive of an atheist identity than an agnostic one. And, the same is true for being politically liberal. A political liberal is more likely to be an atheist or an agnostic, but the impact of that variable is larger in predicting an atheist identity.
That regression model is the icing on the cake, in my mind. The factors that make an atheist, namely political liberalism and the lack of church attendance, also work to predict an agnostic but at a much smaller magnitude. Taken together, I think it’s statistically accurate to say that agnostics are in many ways just a toned down version of atheist. And, as I show in my recent book, many agnostics do become atheists over time. That’s likely as their political identity begins to shift to the left.