Faith and Family in America — Commentary

October 19, 2005

Read comments of Anna Greenberg of Greenberg Quinlan Rosner Research, Inc.; University of Akron political science professor John Green; and University of Virginia sociology professor Brad Wilcox at the October 19, 2005 press conference in Washington, DC releasing results of RELIGION & ETHICS NEWSWEEKLY’s national survey on Faith and Family in America:

ANNA GREENBERG: Changes in the American family are probably the most significant changes we’ve seen in the last, I’d say, 30 years, and it’s happened in some ways almost unnoticed, though certainly not by some folks, especially some political actors in our society. But it’s gone somewhat unnoticed. In the 1950s, 80 percent of all Americans lived in a household where the head of household was married. Now 50 percent of Americans live in a household where the head of household is unmarried. Almost half of all people in America are not married.

There are lots of other statistics we could talk about: the increase in the number of kids living in single-parent homes; obviously, the level of divorce. There are lots and lots of ways we could slice the changes in the American family, but nevertheless it is quite significant, and we can have an entire conversation about why we’ve seen these changes, ranging from the women’s movement to the birth control pill to no-fault divorce to changes in the economy which require two-income families. This is a huge conversation.

But we are going to talk about the relationship between these changes in the family and religion in America, and we’re going to do it in three ways: first, looking at what we call “traditional” versus “nontraditional” families and how they interact with religious life — and I will give you a definition of that in a moment; second, how religious institutions are confronting these changes; and third, what are the implications for the future of religious life in America?

The survey itself was a random-digit dial telephone survey of about 1,100 adults over the age of 18 conducted this past summer, and we had oversamples of traditional families and nontraditional families, so that we could really dig into the differences between these two groups. Overall, the survey has a margin of error of plus or minus three, but obviously, whenever we look at subgroups, the margin of error is higher. We did not report, when we looked at the differences between traditional and nontraditional families, on any differences that weren’t beyond the margin of error.

Both John Green and Brad Wilcox played a key role in helping construct this survey, making sure that it’s rooted in the literature, in what we know about family life and know about religion, and [they] were invaluable to making this study as good as it is.

In this survey we are defining “traditional” families as any married couple with children under the age of 18, and that’s 24 percent of the American population. We’re defining “nontraditional” as unmarried parents with children under the age of 18, and that is 16 percent of Americans. That can include single parents; it can include cohabitating couples who have children but are not married; it can include same-sex relationships with children. We have a broad definition of “nontraditional” and, as you can see, there are still more “traditional” families, but obviously that has changed dramatically over the years.

I’m going to focus on three of the most interesting and major findings. The first is the gap between beliefs and reality. The second is looking at what is a family and how do people see family and the importance of family. And the third is to look at the differences between traditional and nontraditional families in their relationship to religious life. Then John Green and Brad Wilcox are going to have a lot more to add to what I say.

There’s a significant gap between what we call beliefs and reality. Nearly everybody in this country supports an idealized vision of family. Seventy-one percent of Americans agree that it’s God’s plan that marriage should be between one man and one woman, and four in five agree that it’s better for kids if parents are married. There [are] very strong beliefs and — this is across traditional and nontraditional families — you might even say a romanticized kind of vision of what family should look like. And yet there is a very basic acknowledgment of the reality that family life does not look like that, both in attitudes and actual behavior.

Attitudinally, 52 percent of Americans say divorce is usually the best solution if parents can’t work out their problems. So more than half — not much more than half, but more than half — say in fact divorce is a good solution if parents can’t work things out. Only 22 percent of Americans think that divorce is a sin. There is just a dramatic, dramatic change in attitudes about divorce. And about half, 49 percent, accept the notion that cohabitation is just fine. Now there’s a difference between cohabitation and trial marriage. In fact, a majority do not think a trial marriage is a good idea. The notion that you would live together for a few years before you get married to see how it works out — only 40 percent agree that it’s a good idea to try out marriage. But half say it is okay for couples to live together if they don’t intend to get married.

There is, I think, a fairly dramatic disjunction: 71 percent think marriage should be one man-one woman for life, and then at least half of this country accepts divorce and cohabitation. Behaviorally, lots of people are getting divorced. [Yet] lots of people who are divorced or lots of people who never get married and have children will hold these idealized views.

It’s not surprising, however, that there are big differences between traditional and nontraditional families. Traditional families are less likely to support cohabitation and trial marriage, less likely to think divorce is okay. What’s interesting is that 69 percent of people in nontraditional marriages believe that is God’s plan for marriage, [to be] one man-one woman for life. And I should say that the majority of people who are in nontraditional families are divorced. About 60 percent of them are divorced, and the rest are people who have had children without getting married. You have this interesting contradiction or tension.

The other not surprising finding is there are very big differences by religious tradition. One of the frameworks we used to analyze the data was looking at evangelical Protestants, mainline Protestants, traditional Catholics versus liberal Catholics, and people who had no religious preference. Obviously there’s a group of others, but that sample size wasn’t big enough to do real analysis of them. We don’t think they’re unimportant; we just couldn’t get enough of them in the survey to talk about them.

What’s quite notable is, not surprisingly, very big differences between religious conservatives and religious moderates and liberals and people who have no religious preference. Evangelical Christians and traditional Catholics are much less supportive of the notion of divorce, less supportive of cohabitation, and less supportive of trial marriage. Only about a third of evangelical Christians say that divorce is a sin, and 30 percent of traditional Catholics say divorce is a sin. This notion of divorce as a sin is not supported among religious conservatives. There are a lot of ways we can speculate about that, but evangelical communities have been very welcoming of people who are divorced, and so [disapproval of divorce] is not an attitude that is expressed.

Divorce is quite common among evangelical Christians and among traditional Catholics. There’s one difference that’s important to note: if you are very religious — in other words, if you attend church or another institution more than once a week — you are less likely to get divorced than if you were less religious. But if you are more religiously conservative in terms of your faith tradition, you are not less likely to get divorced and in some cases more likely to get divorced. The issue is not so much tradition as the intensity of your religious commitment that’s related to divorce, and there are lots of things we can talk about in terms of evangelical Christians — class and income and how that all relates to divorce. We know, for instance, that people without a college education are much more likely to get divorced than people with a college education.

What is family? This is a huge issue. Anybody paying attention to politics over the last couple of years knows that gay marriage is a huge debate in our politics. But we also have other issues. We have marriage initiatives coming out of the current administration encouraging people to get married, especially people in poverty. We certainly have a lot of activism among conservative religious groups around family breakup and the state of the family. Anyone who goes on the Metro in Washington sees the signs about “Marriage Works.” This question of what is a family is obviously very important and really is in flux.

We found that people’s notion of family is quite flexible. There’s no doubt that it’s about kinship for most people, in other words, being related by blood. In an open-ended context, we just said, “Tell me what family is” before we even asked any of the other questions in the survey, so that we wouldn’t bias their answers, and there’s no question that most people — three quarters — think it’s someone you’re related to genetically or by blood. But that does not necessarily mean a nuclear family. In fact, only a third said a mother, a father, and children, or married parents and children. There is a sense that you should be related to each other to be family, but it can be an extended network. It can be grandparents, it can be people who are divorced in blended families with stepchildren and half-brothers and sisters. And of course this, I think, is in part a reaction to the reality of the decline of the nuclear family — that we are seeing fewer and fewer traditional families where a man and woman get married and have kids and stay married for life.

In fact, over half — 55 percent — agree that “Love is what makes a family, and it doesn’t matter if parents are gay, straight, married, or single.” We have had a change over time in what marriage is supposed to be about. It’s very much in our postmodern times about partnership and love, and much less about economic relationship. This notion that being in a family is just about love, I think, is a quite modern notion.

A little over half of Americans agree with that. There are, as you can imagine, differences between traditional and nontraditional families about what family is. People in nontraditional families quite strongly believe in a flexible definition of family because they aren’t in what we think of as a traditional family. And there are big differences by religious faith tradition. Traditional Catholics and evangelical Christians [are] significantly less likely [to agree] that love means a family whether or not [it is] gay or straight than either liberal Catholics, mainline Protestants, or people without a preference. There is still evidence of polarization, even as we have the majority of Americans agreeing about a flexible notion of family. When we get to the issue of gay marriage, obviously this issue gets much more divisive.

A majority of Americans in our poll — 59 percent — believe that marriage is about one man and one woman. This is quite consistent with other national polls that have opposition to gay marriage, if it’s a binary yes or no [question], around 55, 59 percent. When it comes to the issue of gay marriage and gay rights in general, this is one of the areas where we’ve seen some of the biggest shifts in public opinion over the last 15 years. This is not a static number. This is a dynamic number, and there are a lot of reasons to believe that it may change even more. Certainly if you take marriage issues out, on most issues around gay people you have a majority of Americans against discrimination and a whole range of other issues, a majority favoring civil unions. But the gay marriage issue remains the one, and I don’t think it is unrelated to a larger issue of what’s happening to family in America, where we’ve seen a little less change. And again, not surprisingly, [there are] big differences by denomination, though not by whether you’re in a traditional or nontraditional family.

Nontraditional families are religious, and a lot of them are religiously conservative. We have a big chunk of African American women especially in the nontraditional family category. This is a group that while very progressive, say, on economic issues and other [issues], are not necessarily progressive on issues of, say, gay rights or gay marriage. In fact, there aren’t very big differences between traditional and nontraditional families when it comes to the question of gay marriage, but there are huge denominational differences.

The only group for whom a majority favors gay marriage is people without any religious preference at all. This is a group that is growing, by the way. About 15 percent of the population has no religious preference. For people under 30, it’s about 25 percent with no religious preference. A plurality of liberal Catholics support gay marriage, but among evangelicals and traditional Catholics there is opposition. They’re split on gay adoption, and again, this goes to the question of flexibility about what family is. While a majority — 59 percent — oppose gay marriage, 49 percent favor gay adoption. Forty-seven percent are against gay adoption.

I think what you’re going to see over time [are] changes in family around same-sex relationships, where they become more and more like every other family with kids, and that probably will precede changes in what their legal status is. [It is] important for some of those changes to happen so that kids have the legal protections of marriage and all those sorts of issues.

The final section I’m going to talk about is religious observance in general. I’m going to neutrally give you data, but Brad has some pretty provocative, I think, conclusions about the future of religious life, given some of the differences between traditional and nontraditional families around religious observance. There is no question that, when it comes to traditional worship and participation in institutions, nontraditional families are less religious than traditional families. As an example, 36 percent of people in nontraditional families attend church or religious services at least once a week, compared to 50 percent of people in traditional families. That’s a pretty significant difference.

Because there are more African Americans in nontraditional families, there are fewer denominational differences than you might expect around evangelical/nonevangelical, because most African Americans are evangelical. But 17 percent of nontraditional families say they have no religious preference. They’re not religious at all, compared to 9 percent of people in traditional families. There’s no doubt that there are religious conservatives in the nontraditional family group, but there are also more people who are not religious, who have no preference.

What is interesting, however, is that if you look at informal measures of religious observance, the difference is actually close. The gap closes between traditional and nontraditional families. Fifty-five percent of nontraditional families say religion is very important, [as do] 59 percent of traditional families, even when we control for race. In other words, because we know that African Americans are more religious, we looked among whites and still see that traditional and nontraditional [families] are closer to each other when it comes to this perceived importance of religion, and if you look at other informal measures — reading the Bible at home — 50 percent of both traditional and nontraditional [families] say they read the Bible. We know that’s over-reported, but what I’m really interested in is just what the differences are between the groups. Daily devotions at home: again, about half of traditional and nontraditional families say that they have daily devotions. Similarly, they’re equally likely to say that they talk to their friends about religion.

I think this has a couple implications. The first is that there are huge stresses in the lives of nontraditional families. People in nontraditional families are lower income; there are more minorities. They are more economically marginal than people in traditional families. We know that one of the things marriage does is increase economic stability. But if you look at the range of questions about what people in families worry about, nontraditional families are 10 and 20 points more worried on every single measure about their kids, whether that’s economic issues or values issues. I infer from some of these data that the kind of stresses around nontraditional families may make it harder for them to participate in traditional religious institutions. They’ve got incredible time challenges. Mostly they are raising their kids alone. A majority of nontraditional families are single parents. They’re not people with partners, so it may just be hard for them to get that into their life. I think a lot of people would argue that if they did, there would be a lot of benefits to their family by being integrated into a religious community. Maybe some of them find religious institutions unwelcoming because they are in nontraditional families. I don’t think there’s a lot of evidence of that. Brad can speak to this more, but a lot of evangelical churches, for instance, are quite welcoming of people who are divorced. They have ministries to people who are single parents and divorced. Certainly in the more liberal traditions there’s no reason to believe that they’re unwelcome. I think that the issue is probably not religious institutions themselves, but the kind of stresses around time and other aspects of nontraditional families’ lives, where traditional participation is just much harder for them.

JOHN GREEN: Let me say that it was a real privilege to work on this survey with the sponsors and also with Anna Greenberg and Brad Wilcox. I think it’s a very rich set of findings that will reward a great deal of attention, but let me just pick up on one of the things that Anna talked about at the very beginning of her remarks, which was the tension between the ideal of marriage and family life and the reality that people experience. If you look at this from that point of view, then churches and other religious institutions have been fairly successful in maintaining a certain kind of ideal that’s accepted even among people who can’t live it out. On the other hand, when we look at the kinds of households and families that people live in and their attitudes towards the stresses of everyday family life, we see a different reality — that relatively few Americans actually live in those types of idealized situations. And that’s something to think about.

On the one hand, it may be that the ideal has persisted precisely because the reality has changed, and there are an awful lot of people that would very much like to have the kind of family life that’s idealized in religious traditions, but also in the sort of popular images of the family.

On the other hand, Americans have become much, much more tolerant of deviations from that ideal, I suspect because they experience those deviations — that they themselves may have been in divorce or may have been involved in other kinds of nontraditional families, and they certainly know an awful lot of people, may even be related to a lot of people, who have those different kinds of family structures. And this shows up in the politics of family values, if you will, in an interesting way.

As Anna pointed out, Americans do not tend to see divorce as a sin any longer, even among very conservative religious people. That figure has become much lower than it used to be. There is strong support for traditional marriage in the law and not very much support for same-sex marriage and, depending on how you ask the question, certainly not overwhelming support for civil unions, although there is somewhat more support than for same-sex marriage.

In some sense, I think the support for maintaining the traditional legal status of marriage reflects that ideal — that sense that this is the way it really ought to be, even though not everyone can live up to it. On the other hand, there is a sense that perhaps the government, the state should not be too involved in family life. One of the statistics that Anna didn’t bring up, but I’ll bring it up right now, is we also asked people about the various programs sponsored by the government to encourage marriage. And 82 percent of the people in our survey were against having government involvement in encouraging marriage.

I think that’s another part of this ambivalence, this disjunction between idealized families and the reality. Many people would certainly like to see the ideal enshrined in the law about marriage, but they’re not necessarily sure that they want the government telling people how to live their lives and what kinds of families they should be in. This disjunction, I think, is really very important, and it helps explain the politics that surrounds family values.

BRAD WILCOX: I just want to add a few points here, and I’ll speak both on the survey itself and on the General Social Survey (GSS), which is a large national survey that’s been conducted since 1972. I did some new analyses of that survey for this press conference and for the series.

One of the interesting things, as we’ve talked about just now, is this gap between belief and behavior, and a lot of this centers around the issue of marriage, obviously. I want to just once again highlight the issues that don’t always come to the fore in this discussion. There have been a number of media accounts around questions of evangelicals and born-again Americans being more likely to divorce, and that is, indeed, true. But I think we have to remember that there are also important issues of race and class here, and part of this phenomenon is the phenomenon of race and class and not really religion per se, because African Americans and working-class Americans are more likely to face stresses in their lives that make them more vulnerable to divorce. That’s an important, I think, qualifying point.

Another thing that relates to that, as Anna mentioned, is that there is a strong association between religious practice and marital stability, so folks who attend services on a weekly basis are between 30 and 40 percent less likely to divorce. We have to be careful, I think, here in talking about evangelicals or born-again Christians in the abstract and then also talking about folks who actually attend services on a regular basis, and that could be evangelical Protestants, it could be Catholics, it could be Jews. But folks who go to church or to synagogue on a regular basis are much less likely to divorce.

Another interesting point that comes out in the survey is this gap in attendance between traditional and nontraditional families. About a third of adults having nontraditional families attend services weekly, compared to about half of adults in married households. This gap is actually even larger in the population as a whole. In the U.S. population, 32 percent of men and 39 percent of women who are married with kids attend weekly, and this compares to just 15 percent of men and 23 percent of women who are single without kids. There’s a big gap in the U.S. population between folks who are married with kids and folks who are either not married or don’t have kids, and this gap is particularly big among men.

Men are much less likely to attend church if they’re not married with kids. We have to just think about how certain types of families are associated with religious practice.

I also did some analysis of changes over time in religious attendance. In 1972, 41 percent of Americans reported attending [church] on a weekly basis, and that falls to 31 percent in 2002 over a 30-year period. When I looked at the link between these declines in attendance over the last 30 years and family structure or family trends, what I found is that about a third of the decline in attendance is associated with these changes we’ve been talking about in the American family. If we had the same number of adults who are married with kids as we did in 1972, we’d see a lot more folks in the pews on Sunday.

This is important in part because it has important consequences for the congregational life of many churches and synagogues in the U.S. And so while it may be true that folks are still doing things on their own, whether it’s reading the Bible or praying or whatever, if they’re not integrated into a community, it has consequences for those communities, of course, but also for themselves. We know that being integrated into a religious congregation puts you in touch with social networks and norms that can then impact your family life in a variety of ways.

There seems to be a strong association between religion and marriage for men, a stronger one than for women. So, for instance, just looking at issues of martial happiness, we find that folks who attend on a weekly basis are happier in their marriages. We also find in the survey that they’re happier with their family life and happier with life in general. But these effects seem to be particularly strong for men. Seventy-two percent of married men who attend weekly report that they are very happy in their marriages, compared to 60 percent of married men who don’t attend weekly, whereas for women it’s 64 percent of married women who attend weekly are very happy, compared to 58 percent of married women who don’t attend weekly. In a sense, the religious boost for marital quality is stronger for men than it is for women. And, once again, the link between marriage and attendance is also stronger for men than it is for women. There is something about this religion-gender link, which I think is interesting to highlight.

I want to just conclude by once again stressing this idea that the RELIGION & ETHICS NEWSWEEKLY survey shows some important differences in attitudes toward things like cohabitation, divorce, and same-sex marriage between, on the one hand, traditional Catholics and evangelical Protestants, and on the other hand, mainline Protestants, liberal Protestants, and liberal Catholics. Most surveys until recently haven’t actually broken out the Catholics. This is one of the first surveys that actually break out Catholics along traditional and liberal lines, asking the respondents to say, “Are you a liberal Catholic or a traditional Catholic or just a Catholic basically?” So we are able to see for the first time that the divisions we’ve seen at the elite level on a lot of hot-button social issues, like abortion or same-sex marriage, are also being mirrored to some extent in the pews. One of the things I really think is good about the survey is that it does break out Catholics in ways that allow us to get at that phenomenon in the pews.

JOHN GREEN: We thought one of the most interesting findings in our survey was the whole question of moral values. I’m sure you all remember that caused quite a stir after the 2004 election, because one of the questions on the exit polls showed that the largest group of priorities that voters had were “moral values.” About a fifth of the public held those views, and that set off a storm of argument. We ask a very similar question on our survey, and what we found was about the same number of people chose “moral values” as opposed to jobs, the war in Iraq, other sorts of issue priorities — a little bit less than on the exit polls, but roughly about one fifth of the population. We do have a group of people who identify or respond to the term “moral values” when asked about their political priorities.

But then we did something really, I think, quite innovative. We went and asked people what they meant by “moral values,” and then we found something really quite interesting. If you look at the entire sample, only about 10 percent of people define “moral values” as issues such as abortion or same-sex marriage. Another 25 percent, roughly another quarter, mentioned family issues such as protecting children from sexual abuse or from violence and sexual content on television. But the single largest definition for “moral values” was personal values, personal honesty, personal responsibility. So we find a really interesting thing. A lot of people identify with this term that may, in fact, influence their voting behavior, but it means different things to different people. And the most common meaning isn’t those hot-button social issues that are oftentimes connected with that term “moral values.” They’re, rather, questions of personal behavior — honesty and responsibility.

If you take out the 18 percent that picked moral values as their top priority, they are somewhat more likely to name abortion and marriage as their definition of “moral values,” but it’s only a little bit higher than the group as a whole, and still these questions of personal values come in first. If you look at evangelical Protestants, regular worship attenders, people in traditional families, the numbers on these hot-button social issues as a definition for moral values [are] also a little bit higher. But even for those groups, it’s these questions of personal values that are really the most common definition for “moral values.”

This is really an interesting finding, because it tells us for the first time how people think about the meaning of that term “moral values.” A lot of people care about morality in the United States, but the hot-button social issues are by no means the most important thing that they care about when they use that language. This is not to suggest in any way, shape, or form that people who care about abortion and marriage are not an important part of the electorate, an important part of the political process. They certainly are. But the whole question of moral values is really a much more complicated issue and extends to a wide variety of values and not just to those issues.

ANNA GREENBERG: This survey is a snapshot in time. This is not a longitudinal study. It would be interesting to know, are these attitudes [about cohabitation and divorce] moving? We know if we look at census projections that the increase in, say, single-headed households is continuing. In other words, this is a dynamic process. We’re not at point A and it’s going to stay like this. We’re actually going to be a majority unmarried country in the not too distant future. What I want to see is, are these more progressive attitudes about cohabitation and same-sex relationships moving in that direction, and do they continue to move that way as family structure changes?

If you look at the two groups that are growing in the religious landscape, it is conservative Christians, whatever denomination, and it is people who are nothing. If you look at younger people, you see that even more dramatically. I don’t know that there are limits to changes in peoples’ attitudes. I don’t think there are, actually. I think we’re in the middle of a process that hasn’t ended. But I do think there are going to be more pitched battles and more polarization, because those are the groups that are growing most quickly, and it’s the groups in the middle that are in some ways — for instance, if you look at mainline Protestants that are on the decline, one study talks about the fact that one of the reasons why we see the growth in people who say they are nothing is that you have a movement of people who are more politically liberal out of religious communities because of the polarization on cultural issues in religious communities. Again, I don’t see limits. I see polarization.

BRAD WILCOX: And just the fact that we actually see these family dynamics affect church attendance are most consequential for mainline Protestant and moderate or liberal Catholic churches. One of the fascinating ironies of our day and age is that as the family itself has become less centered around the traditional model, we’ve actually seen a growth in evangelical churches. I think one of the reasons that’s happening is that people are reacting to these trends. But I think another reason that it’s happening is that folks who don’t fall into this kind of conventional family style are less likely to go to a mainline Protestant church or to a Catholic church for the religious and moral formation of their children. This is one of those interesting ironies where the center in a sense is getting smaller as the polar extremes are getting larger. And the changes in the American family are related to these developments.

ANNA GREENBERG: What is interesting about some of these churches in the center is they are less well equipped to handle families. What evangelical churches do very well is provide a whole set of ministries — small groups, for example. But, if you look at mainline churches, for instance, they are much less likely to have activities every night of the week or over the weekend to help. Similarly with African American churches — they are also more likely. It’s not just that [people] are less likely to go to these mainline churches to inculcate values in their children. Those kinds of institutions are less likely to have things for them to do with their children. It’s a real institutional issue as well.

BRAD WILCOX: Singles often don’t feel there is a place for them in their local church or synagogue; that is part of the equation. Part of the story is a lack of pastoral offerings for folks who are not married with kids, and part of it is the fact that folks who aren’t married with kids tend to be less interested in the kinds of things that churches and synagogues and mosques can offer to them.

ANNA GREENBERG: We also know that people are delaying marriage. Twenty-six, 27 is the average age for women, and it was much different in the ’50s and ’60s. You have this much longer period in people’s 20s where they’re not getting married. Some are having kids, but other aren’t, and you wonder. I mean, some people come back [to church] when they have kids. But when you spend such a long time out of religious life, what’s the likelihood of going back? I think that is not insignificant. If you look at the National Election Study or the GSS and the change over time and how many people are married in that 18-to-30 group, it just [drops dramatically]. And that has a big impact on their participation in religious life.

BRAD WILCOX: In my analysis of the General Social Survey, I looked at adults from 1992 to 2002 by the religious denomination that folks indicated they are affiliated with. Forty-two percent of evangelical Protestants — these are folks who have to have been married at some point — are divorced or separated. That compares to 39 percent of mainline Protestants, 53 percent of black Protestants, 35 percent of Catholics, 36 percent of Jews, and 53 percent of unaffiliated folks. The bottom line is that basically, evangelical black Protestants and secular adults are more likely to report having been divorced or separated.

ANNA GREENBERG: Evangelical and mainline Protestants are more likely to get married than others, but they are no more likely to stay married. Twenty percent of evangelicals have been married more than once, compared to, say, 15 percent of people with no religious preference. But I want to reiterate Brad’s point. A lot of this is related to class and race. There is a new study that just came out that showed that women who were married between 1990 and 1994 — if you were college educated, you were half as likely to get divorced than if you didn’t have a college education. A lot correlates with the demographic reality of these folks’ lives, and it isn’t a religious explanation per se for this.

BRAD WILCOX: My colleague Penny Edgell at the University of Minnesota found in her study — this was in upstate New York, but I think her study is probably suggestive of trends elsewhere — that it was actually the evangelical Protestant churches that were most likely to have ministries that combine on the one hand, this notion that we have an ideal about what marriage should be, and on the other hand, have ministries that were targeting folks who weren’t able to live up to that ideal for one reason or another. Having more singles ministries, having more ministries for stepfamilies, having more ministries for folks who had been divorced — I mean, the irony, of course, here is that the tradition that in some ways is most opposed to these developments symbolically is also the tradition that practically has more ministries on the ground. I’ve done a study on this with the National Survey of Congregations that finds some similar patterns. It’s one of these ironies in American religious life, you know, that we find mainline Protestant churches, as Anna said, less likely to have ministries serving nontraditional families, even though symbolically they’re more likely to affirm their commitment to family diversity.

JOHN GREEN: As Anna indicated, when we asked people about the stresses and worries in their lives, we found that people in nontraditional families expressed more concern about [values and their kids]. Many of the concerns that people have about raising their children and so forth are exactly the sorts of things where congregations have historically provided a great deal of help. And yet those people who need that help are not as involved in congregations. Although, as Anna also indicated, by other measures of religiosity, the subjective importance of religion, prayer, and so forth, [nontraditional families] are just as religious in that sense. One implication of this finding is that there is a real opportunity for congregations to find ways to help nontraditional families cope with the stresses in their lives.

ANNA GREENBERG: There is important research by Christian Smith and others that talks about how kids do when they grow up in families that are more religious. I’m not making any kind of normative statement here, but kids and teens who are from religious families — and I saw this in a study I had done on youth and religion as well — tend to have higher self-esteem, do better in school, are less likely to use drugs, less likely to drink. There are a whole lot of things that parents are worried about, which you see dramatically in the survey, especially among nontraditional families. When you look at kids that grew up in religious families, and by “religious” I mean [regular church] attending, they actually do better on a lot of those measures.

JOHN GREEN: Married couples that have never been divorced and that don’t have children at home make up about 25 percent of the population but wouldn’t be counted in our definition of “traditional” family because they don’t have kids at home. But they’re an important part of the public, and in terms of attitudes they look a lot like married couples who have kids at home. In other words, they tend to have very traditional values. While only about a quarter of the population fits that definition [of “traditional”], there are other people who have participated in that definition at some other point in their life and who probably think of themselves as being in traditional families. Another group is widows with no children at home, who are people unmarried probably for involuntary reasons. And their attitudes look a lot like [those of] traditional families of one kind or another. Because they tend to be older, though, they’re a little bit more traditional on some of these things.

The moral values question, whether it’s in the exit polls or in our survey, is a priorities question. We’re asking people, what are their priorities? We don’t actually ask them what their attitudes are. We have other questions where we ask people what their attitudes are, and what we find is just that a lot of people who have conservative attitudes on marriage or on abortion simply don’t rate those issues as their top priority. Perhaps it is because jobs are more important, or foreign policy concerns are more important, or whatever. So there is a potential disconnect, if you will, between people’s attitudes and their priorities. But one of the things you have to understand about a priority is that only so many things can have top priority. By nature you rank them, and a lot of people who have traditional values in one sense or another don’t make them their top priority. And then, on the other hand, people when they think about moral values as a priority have a broader set of definitions. Other questions fall in there besides just abortion and marriage.

ANNA GREENBERG: I would add that if you look at our evangelical study from last year, we asked about a whole range of concerns, and you see that evangelicals, who are going to be more likely to say “moral values” is an important issue, have a whole set of economic concerns about health care costs that are similar to everybody else. In a lot of ways evangelicals look like every other American, and so it’s not surprising that when you ask these moral values questions that even for people who are conservative, gay marriage and abortion are lower on their list than other things like, “How am I going to pay for my health care?”

BRAD WILCOX: What’s also interesting is the survey suggests that about 10 percent of traditional Catholic and evangelical Protestants think that all the kinds of priorities articulated in the survey are important to them — personal values, family values, social issues, and social justice also. I think for a lot of these folks there’s a sense in which they are to some extent in our current political context homeless, because neither party really in their view does a good job on both economic issues and social issues.