Who Marries Whom? Faith, Partners, and the Unequally Yoked

What new Pew data reveals about faith and who we marry

This post has been unlocked through a generous grant from the Lilly Endowment for the Association of Religion Data Archives (ARDA). The graphs you see here use data that is publicly available for download and analysis through link(s) provided in the text below.

---

There’s this phrase I heard over and over again growing up in a Southern Baptist Church - “don’t be unequally yoked.” It’s a reference to a verse from 2 Corinthians 6:14, “Do not be yoked together with unbelievers. For what do righteousness and wickedness have in common? Or what fellowship can light have with darkness?” In my experience it is always used in the context of finding a spouse. The admonition is to find someone to marry who has the same faith background as you. The worry for my evangelical group leaders was that if one of us married a non-believer, our spouse might lead us down a path toward secularism.

But the general admonition is the same in many faith communities: try to marry someone who shares your faith background. That’s certainly a well-established norm in Jewish communities. According to Kiddushin 68b, marriage between a Jew and a non-Jew is prohibited under Jewish law. In Islam, there’s not an outright prohibition on marrying a non-Muslim, but the Quran only permits people to marry other “People of the Book.”

There are theological reasons for this, as was commonly invoked in my evangelical upbringing. Each spouse can edify the other’s faith and grow closer together as they grow closer to God. But there’s also a really important sociological reason for it, too: kids who are raised by two parents who share the same faith background are more likely to be raised in a consistently religious home. This likely increases the chances a faith is passed down from one generation to another. This is a powerful reason religious groups might discourage interfaith marriages.

The ARDA is hosting the most recent wave of the Pew Religious Landscape Study, which was conducted in 2023-2024 and it has a whole host of insightful questions about who folks are marrying and how much they talk about faith in their home. It includes a variable indicating whether respondents’ current religion matches that of their partner’s. If a Catholic married a Protestant, that would not be defined as a match. If a mainline Protestant was married to an evangelical, that’s still a match because they are both Protestants. A Muslim marrying a Jew wouldn’t be a match and so on. You get the picture.

The groups most likely to marry someone of the same faith background are Hindus (86%), evangelicals (83%) and Latter-day Saints (81%). They are followed closely by Black Protestants at 79%. Then there’s a pretty noticeable gap and a whole bunch of traditions are clumped right around 70% including Catholics and the non-religious.

What really jumps out is that Jewish intermarriage rates are strikingly high. Only 58% of Jews marry a fellow Jew. Given Judaism’s strong norm against marrying outside the faith, it’s noteworthy that so many Jews find a partner from another religious background. I know what you all are wondering: what is the religious background of their spouse? I can’t answer that. The Pew survey did ask that question, but they didn’t include it in the publicly available dataset. So, it’s going to remain a mystery.

But let’s get back to the idea of being “unequally yoked.” I wanted to know if those who said that religion was very important to them also married a spouse who had the same view of religion. Or maybe people who just didn’t care about it that much managed to partner with someone who had the same general outlook.

The heatmap plots respondents’ own sense of religious importance against their perception of their spouse’s. The big thing that jumps out to me is the bottom left corner of the graph: over a quarter of all married respondents said that religion was not at all important to them and reported that their spouse felt the same way. In fact, over half the sample is located in the bottom left corner of this heatmap: they don’t think religion is very important at all and neither does their spouse. I think it’s fair to say based on this graph, that half of married couples aren’t placing religion at the center of their relationship.

In contrast, only 12% of married respondents said that religion was both very important to them and their spouse. The four squares in the top right account for 28% of the entire sample. So, just over a quarter of couples place a high value on religion.

The other thing that struck me was how hard it is to find a discernible pattern in this data. I expected most respondents to cluster along the diagonal. People who said religion was somewhat important to them would also marry someone who said the same. But that’s just not what we see here. Only three cells reach double digits — which tells you how diffuse the distribution actually is.

Let me show you that data in a different way, though.

The top row of this graph is people who said that religion was very important to them and 63% reported marrying someone who considers religion equally important. Another 21% said their spouse considers religion somewhat important.Taken together, this is strong evidence that highly religious people overwhelmingly partner with someone who shares that priority.

You can also see the reverse of that pattern among people who say that religion is not at all important to them. Two-thirds of them partner with someone who doesn’t place any importance on religion at all and 22% say that their spouse rates religion as “not too” important. So, 84% of the top row are in the two highest categories and 89% of the bottom row are in the lowest two categories. People tend to marry someone who agrees with them on matters of faith.

The share of people who are deeply religious who marry a non-religious spouse? 3%.
The share of people who are not religious at all who marry a deeply religious spouse? 3%.

People tend to want to marry people who have the same views of faith as they do.

Folks who took part in the Religious Landscape Study were also asked, “How often do you and your spouse or partner talk with each other about religion?”

The pattern is clear: theologically stricter groups dominate the top of this chart. About half of Jehovah’s Witnesses said they talked about religion every single day with their spouse. That was easily the highest of any religious tradition. Next were Latter-day Saints - 35% were speaking with their partner about religion every day and another 39% were talking about their faith on a weekly basis. Muslims and evangelical Christians were in this same general range.

Who doesn’t talk about faith that much? Well, I don’t think any of us would be surprised to see atheists and agnostics at the bottom of the list. Less than 8% of those respondents talk about religion on a weekly basis and nothing in particulars were just a bit higher than that. What really struck me, though, were the mainline Protestants. Just 28% said that they were talking with their partner on a weekly basis about religion. But Catholics weren’t that much higher than that, though, at 30%.

But any astute reader of this newsletter will know that the prior analysis is likely deeply related to the data I presented in the first graph of this post: people in interfaith marriages are less likely to discuss religion regularly than those who married within their tradition. That’s exactly what the data indicates.

Among evangelicals who married a non-Christian, just 21% of those couples talk about religion on a weekly basis. If an evangelical is married to another Christian, then faith is discussed on a weekly basis in 71% of those households. But the gap is not that large for other Christian groups, though. For mainline Protestants, it was discussed weekly in 10% of mixed-faith marriages compared to 36% when both partners were Protestants. For Catholics, the comparison was 17% vs 35%. So, it does appear that evangelicals (and Black Protestants) are in a different category on this metric compared to mainline Protestants and Catholics.

The Jewish graph is striking, too. When a Jew marries a non-Jew, religion is discussed rarely in about two-thirds of those households. Among Jews married to other Jews, that share drops to 38% — still notable, but meaningfully lower. This is a truism across every faith group: if you marry someone who shares your background, then religion is a regular topic of conversation. If it’s an interfaith union, it rarely comes up in discussion.

There’s a foundational concept in social science that helps explain these results: homophily. It’s pretty simple – people like to associate with people who are similar to them in as many dimensions as possible. That explains why neighborhoods are often stratified by race or income. But it’s also incredibly powerful when it comes to marriage (where it’s termed homogamy). It’s easier to spend your life with someone when you have a shared set of experiences and beliefs about the world.

If you wanted to read some of the literature about this concept I would recommend Kalmijn’s canonical work in the Annual Review of Sociology. Or Sherkat’s piece in Social Science Research which focuses more on religion.

One has to wonder how the rates of intermarriage will change based on a number of shifting societal patterns in the United States. With the rise of the nones religious individuals will have a smaller potential dating pool to draw from for a future mate. But because those who remain in faith communities tend to be “true believers,” their desire to find a partner from their own religious background will likely be heightened. These are two factors working in tension with each other. Whichever factor wins the day will say a lot about where American religion is headed in the coming decades.

Code for this post can be found here.

---

Ryan P. Burge is a professor of practice at the Danforth Center on Religion and Politics at Washington University.