A Deeper Dive into the Religious Gender Gap

Let me give you a “behind-the-scenes” look at how a story can wind its way through the media ecosystem. In June 2023, in the early days of this Substack, I wrote a brief post.

At that point in my Substack’s life, it did pretty well. It received a bit over 7,000 views—3.5 times the number of subscribers I had—and generated a nice bit of revenue. But it didn’t have much “tail.” In other words, within a day or two, no one was talking about it. That’s completely the norm in the newsletter game, by the way. If you can write something that endures for more than 24 hours, that’s atypical.

Every once in a while, I would tweet a graph about the gender gap in religion, and it would get a little traction, but it wouldn’t really reverberate around the discourse. However, in September, I sent this one out:

And it got quite a bit more social media traffic. Once you get above 100 retweets, things really start to move. I sent that out because Ruth Graham and I had a conversation a few weeks before about the gender gap in religion and I wanted to see if a story like that would have any loft. And, of course, getting 210K impressions was a pretty strong indicator that the story would land well.

It was published on September 23, 2024 with the title, “In a First among Christians, Young Men are More Religious Than Young Women.” I think it’s fair to say that (beyond the election), it’s the most well-read story about religion in the United States for 2024. In my world, that means that since I was quoted, I will get contacted by several other reporters who want to pull on the threads of that idea a bit more. When I scan my calendar the last two months, I think I have spoken to at least five media outlets as a result of the NYTimes story. That’s how the media ecosystem works - they set the agenda and other reporters live in their wake.

And, so do I, really. I want to write stories that people read. That’s why I continue to noodle around on that data. I did that for a piece a couple weeks ago that also got cited in the NYTimes by Ross Douthat.

In early November, I was speaking to a reporter from another major outlet, and they asked me to investigate whether a specific type of Christianity is seeing a revival among young men. So, I made an interesting graph, and now I can share it with you.

I broke the sample down into men and women, tracking the share of each who identified as evangelical Protestants, non-evangelical Protestants, and Catholics by age. This is what that data from 2022 and 2023 (total sample size of ~85,000) looks like:

Working left to right, there’s a clear gender gap in evangelical identification. Women are more likely to be evangelical Protestants, regardless of age. The gap is larger at some points and narrower at others, but it isn’t numerically huge. For thirty-year-olds, about 18% of women are evangelicals compared to 16% of men of similar age. It’s almost nonexistent for those around sixty years old, but otherwise it remains steady around 2-3%.

What about mainline Protestants? One clear conclusion is that women are slightly more likely to be non-evangelical Protestants compared to men. But by “slightly,” I mean a single percentage point. The lines never intersect or reverse; they run in near-perfect parallel.

Now, the Catholic line is where the real action is. The story is not as simple here. Among young people, there’s a clear gender gap—men are more likely to affiliate with the Roman Catholic Church compared to women. It’s not huge, but it’s there—about 2-3 percentage points. However, this gap narrows around age 45 and disappears by age 55, not reappearing afterward. This divergence is clearly a young-person phenomenon.

What does this mean in actual numbers? Here’s the share of 18-40-year-olds who are members of each religious group.

We can see a gap for evangelicals: women are about three percentage points more likely to say they are evangelical Protestants compared to men of the same age. For mainliners, the gender gap isn’t statistically significant. Notably, though, women aged 18-40 are twice as likely to identify as evangelical compared to non-evangelical Protestant.

But the Catholic gender gap is there and it’s reversed. Young men are two percentage points more likely to be Catholic than women of the same age. I can already see readers searching for more information on how JD Vance converted to Catholicism a few years ago. I’ll touch on politics a bit later, so stick around.

I also wanted to examine the other side of the gender gap conversation—this time looking at three types of non-religious Americans: atheists, agnostics, and those claiming no particular religion.

There’s definitely a gender gap among atheists, and it’s large compared to the gaps in the previous analysis. Among forty-year-olds, a man is about twice as likely to be an atheist compared to a woman of the same age (10% vs. 5%). This narrows slightly among older folks but remains around three percentage points. For agnostics, there’s essentially no gender gap; women are much more hesitant to identify as atheists in this data.

The “nothing in particular” trend lines are quite different. A larger percentage of people chose this identity—about 30% of the youngest adults. Among those aged 18-40, it’s apparent that women are more likely to be “nothing in particular” than men. The gap closes by their fifties, and older men may even be slightly more likely to identify as “nothing in particular” than older women.

What does this look like in actual numbers? Here’s the breakdown of nones among 18-40-year-olds.

This reinforces that atheism is more popular among younger men compared to younger women: one in ten men aged 18-40 are atheists, compared to just 6% of women. For agnosticism, there’s no gap at all—both remain at 8%.

For “nothing in particulars,” the gender gap reverses as discussed earlier. Among younger women, about 30% identify as such compared to 27% of men. In total, about 45% of men aged 18-40 are non-religious, compared to 44% of women. So, the overall number for nones isn’t a significant difference, but looking deeper, the composition of female nones slightly differs from that of male nones.

Now, as promised, let’s pivot to politics—another potential explanation for the gender gap in religion. Here’s the partisan composition of three types of evangelicals, broken down by gender, focusing on 18-40-year-olds.

Among all evangelicals, there’s no gender gap: 52% are Republicans, 32% are Democrats, and 15% are Independents. However, race plays a significant role here. Among white evangelicals, young women lean more toward the GOP than men of the same religious group; 59% of young white evangelical women are Republicans, five points higher than their male counterparts.

For non-white evangelicals aged 18-40, the partisan gap reverses: 36% of men are Republicans, compared to only 26% of women. It’s interesting that white evangelical women are the most right-leaning group, while non-white evangelical women are the most aligned with Democrats.

Let’s talk about Catholics, too. That’s where we saw the largest religious gender gap in the previous analysis - is politics driving some of that?

For Catholics, where we saw the largest religious gender gap in the previous analysis, politics seems to be a factor. Young male Catholics are significantly more right-leaning than female Catholics. In the whole sample, 38% of men are Republicans compared to 32% of women. Among white Catholics, the gap narrows: 42% of young female Catholics are Republicans compared to 45% of men. There’s still a gap, though.

For non-white Catholics, there’s a clear Democratic preference: 62% of young female Catholics of color are Democrats, and just 20% align with the GOP. For men, it’s 55% Democrats and 27% Republicans. There’s some tacit evidence here for the JD Vance pathway to Catholicism that intersects with politics.

Before I close up, I wanted to show you the partisanship of non-religious groups by gender, too.

There’s a pretty compelling conclusion here: non-religious women are significantly more aligned with Democrats than their male counterparts. Among atheists, 79% of women are Democrats compared to only 68% of men. For agnostics, it’s even wider: 75% vs. 56%. Among “nothing in particulars,” 51% of women are Democrats versus 41% of men. The plurality of the “nothing in particulars” are politically independent, suggesting that rejecting one label often correlates with rejecting others.

There’s a gender gap in religion, but it’s nuanced. Young women are more likely to be evangelicals, while young men are more likely to be Catholic. The most right-leaning group in this analysis is white evangelical women, with 69% Republicans. The most left-leaning group? Atheist women, with 79% Democrats.

Let’s just say I have many threads to pull on for future posts.

Code for this post can be found here.