November 16, 2015

Event: Is the American Public Becoming Less Religious?

A Discussion of the 2014 U.S. Religious Landscape Study

Is the American public becoming less religious? Yes, at least by some key measures of what it means to be a religious person. An extensive new survey of more than 35,000 U.S. adults finds that the percentages who say they believe in God, pray daily and regularly go to church or other religious services all have declined modestly in recent years.

But the Pew Research Center study also finds a great deal of stability in the U.S. religious landscape. The recent decrease in religious beliefs and behaviors is largely attributable to the “nones” – the growing minority of Americans, particularly in the Millennial generation, who say they do not belong to any organized faith. Among the roughly three-quarters of U.S. adults who do claim a religion, there has been no discernible drop in most measures of religious commitment. Indeed, by some conventional measures, religiously affiliated Americans are, on average, even more devout than they were a few years ago.

On Nov. 18, 2015, Pew Research Center brought together religious leaders and scholars for a discussion about the latest data on religious beliefs and practices in the United States and what the data means for congregations, clergy and religious groups.

Gregory A. Smith, Associate Director of Religion Research at Pew Research Center
John Green, Distinguish Professor of Political Science, University of Akron

Alan Cooperman, Director of Religion Research, Pew Research Center

Alan Cooperman: Thank you all for coming. I’m Alan Cooperman. I’m the director of religion research here at the Pew Research Center. The Pew Research Center is a nonpartisan fact tank. We do public opinion research, we do demographic research, we do content analysis and other data-driven social science research, but we do not take positions on public policy issues.

This year, we’ve published two major reports on the results of the U.S. Religious Landscape Study, which is largely a repeat of a massive survey that we first did in 2007, and we largely again repeated it seven years later, in 2014.

There are a few questions that are new in the 2014 survey. One of those questions is about gratitude. So let me hasten to add, I am very grateful for the serious coverage that we have received of our work in major media. However, many of the findings of this very massive survey have gotten little attention in the media, and we think that this rich trove of data may be of particular interest and use to religious leaders, congregational clergy – religious groups of all varieties in the United States.

And so this session today is an attempt to get data of interest to religious leaders, directly to religious leaders. In addition to the people here in this room, we are joined by religious leaders on live stream around the country.

Here in the room, we are delighted to have a diverse group of representatives of religious organizations in the Washington area, including the Seventh-day Adventist North American division, the District of Columbia Baptist Convention, the Washington Buddhist Vihara, St. Paul’s College, the Religious Action Center of Reform Judaism, the Episcopal Diocese of Washington, and the Episcopal Church’s Virginia Theological Seminary, the United Methodist Church’s General Board of Church and Society, the Military Chaplains Association, the Church of Jesus Christ of Latter-day Saints, the Council for Christian Colleges and Universities and the Progressive National Baptist Convention. And, again, we have many others joining us live from around the country.

It is not our place to give recommendations, and we are not going to pretend to tell religious leaders what you should or should not do on the basis of the data that we have. But we do want to help you to understand the data as best we can.

Today’s briefing will begin with my colleague Greg Smith, the associate director for religion research at the Pew Research Center. You will find a copy of his bio in your handouts, so I won’t take much time introducing him. But Greg holds a Ph.D. in political science from the University of Virginia, and he was the principal researcher both on the 2007 Religious Landscape Study and on the 2014 Religious Landscape Study. So he’s the ideal person to brief you on the findings.

And after he presents an overview of the results, Greg will be joined by John Green, whose full – and fully appropriate – title is Distinguished Professor of Political Science at the University of Akron. John’s bio is also included in your folders.

And before I turn it over to Greg, I would like to thank The Pew Charitable Trusts and Lilly Endowment Inc. for the generous funding that made the 2014 Religious Landscape Study possible. And I would like to note for all of you that this briefing is on the record. It’s not aimed at media, but it is being live-streamed and taped, and we may post a transcript on our website.

So with that, Greg, would you walk us through some of the main findings of the Religious Landscape Study?

Gregory A. Smith: Sure, I’d be glad to.

Well, Alan mentioned that we included a question in this survey on gratitude, and so I should start by saying that I’m grateful for the chance to be with you today and to provide this briefing on our research, and to try to answer any questions that you might have. I’m also anxious to hear your comments and get your feedback on our work.

Now, as Alan mentioned, the data that we’ll focus on today come from our 2014 Religious Landscape Study. This was a massive survey of more than 35,000 people nationwide, done in all 50 states and the District of Columbia. We conducted interviews last summer in English and in Spanish on landlines and on cellphones.

The survey’s very large sample size, with more than 35,000 respondents, means that it has a very small margin of error – plus or minus less than 1 percentage point for results that are based on the full sample.

The study examines religious identity, people’s religious background and religious intermarriage. It takes an in-depth look at religious beliefs and practices, social and political values and other topics.

Also, as Alan mentioned, this is the second time we have conducted a study like this. We first conducted a Religious Landscape Study in 2007, so now we have the opportunity to see how things have changed in recent years. And change they have, in some very important ways, starting with religious composition…

To measure the religious composition of the country, we ask people a question about their religious identity: “What is your present religion, if any? Are you Protestant, Roman Catholic, Mormon, Orthodox – such as Greek or Russian Orthodox – Jewish, Muslim, Buddhist, Hindu, atheist, agnostic, something else, or nothing in particular?”

In response to this question, we find that 23% of the public tell us that they are religiously unaffiliated, that they are atheist or agnostic, or that their religion is just nothing in particular. We often call this group the religious “nones” – that’s N-O-N-E-S – because they tell us that they have no religious affiliation.

The religious “nones” have been growing very rapidly in recent years. Indeed, you can see here that their share of the population has grown by about 7 percentage points just in the last seven years. Over the same period of time, we’ve seen corresponding declines in the share of the population that identify with several Christian groups, most notably Catholicism, whose share of the population now stands at about one-in-five, and mainline Protestantism, whose overall share of the U.S. adult population now stands at about 15%.

Now, some have wondered what to make of the growth of the “nones.” How should we interpret and understand this development? Is the growth of the “nones” an indication that the United States is becoming less religious? Not necessarily. After all, not all “nones” are nonbelievers.

The “nones,” the religiously unaffiliated, are not uniformly secular. In fact, most “nones” tell us they believe in God. One-third of “nones” say that religion is at least somewhat important in their lives. One-fifth of the religiously unaffiliated even say that they pray every day.

However, while some religious “nones” do exhibit a proverbial religious pulse, the religiously unaffiliated as a whole are far less religious compared with those who do identify with a religion. And you can see here, for instance, that religious “nones” are 36 points less likely to believe in God as compared with those who say they identify with a religious group. Religious “nones” are also far less likely than those who identify with a religion to say that religion is very important in their lives, and far less likely to pray. They are far less likely to attend religious services with any regularity, and so on.

Furthermore, as the “nones” have grown, they have become even less religious than they were to begin with. Fully 65% of the religiously unaffiliated now say that religion is “not too” or “not at all” important in their lives, up from 57% when we first did this study in 2007. Sixty-two percent of “nones” now say they seldom or never pray, up from 56%. Very few “nones” have ever reported attending religious services with any regularity, and while most “nones” still say they believe in God, the share who don’t believe in God is growing very rapidly. Thirty-three percent of religious “nones” now say they don’t believe in God, up from 22% in 2007.

As a result of these twin trends – the growth of the religiously unaffiliated, coupled with their increasing secularism – the rates of religious observance that we see among the public overall are being tugged downward by a modest, but nevertheless noticeable, degree.

These figures you see here are based on all U.S. adults, including the religiously unaffiliated and those who identify with a religion. You can see here that the United States is still very much a nation of believers, where nearly nine-in-ten adults say that they believe in God. But the share of Americans who believe in God has ticked downward slightly in recent years, from 92% to 89%.

Similarly, we’ve seen a 3-point drop in the share of Americans who say they pray every day and a 3-point decline in the share of Americans who say religion is very important in their lives. And we see a 4-point drop in the overall share of the public who say they attend religious services at least once or twice a month.

Now, we should not overstate the magnitude of these changes. These are modest changes, and the United States remains a very religious country, especially when compared with much of the rest of the industrialized world.

Still, though they are modest, these trends are significant, and the fact that they are pointing in the same direction and changing by about the same amount across so many important indicators of religious observance is very striking. Moreover, these changes – the growth of the “nones” and their increasing secularism – really are quite consequential for religion in American society, serving to pull down the nation’s overall rates of religious observance.

So what’s behind these trends? What’s behind the growth of the “nones”? Here there are a number of things we can point to, but perhaps the first thing to understand about the growth of the religiously unaffiliated is that there is a very, very strong generational component to this change.

You have older generational cohorts of Americans – Baby Boomers and those from the Silent generation, for example – who are overwhelmingly Christian and who, by and large, are pretty devout. As these older cohorts begin to pass away, and as their numbers begin to dwindle, they are being replaced by a new generation of young people – by Millennials, who are simply far less religious than their parents and their grandparents. You can see that here.

Fully 85% of adults from the Silent generation identify themselves as Christians, and just 11% say they have no religious identity. And among Baby Boomers, fully 78% identify with Christianity, and just 17% are “nones.” But look at Millennials. Among Millennials, fewer than six-in-ten identify with Christianity, and more than one-third say they have no religious affiliation.

We see the same kind of generational pattern when we look at religious beliefs and practices. Whereas two-thirds of those in the Silent generation and six-in-ten Baby Boomers say that religion is very important in their lives and that they pray every day, only about four-in-ten Millennials say the same. And whereas half of those in the Silent generation and four-in-ten Baby Boomers say they attend religious services on a weekly basis, fewer than three-in-ten Millennials say this. In fact, the majority of Millennials say they attend religious services a few times a year at most.

So there are obviously a number of important changes underway in the American religious landscape. The religious composition of the country is changing quite rapidly, driven by generational replacement, with dramatic growth in the religiously unaffiliated share of the population. And the religious beliefs and practices of the country are changing more slowly, but nonetheless trending downward. These are very important parts of the story, very important conclusions from this study. But they are only part of the story.

It is also true, as I mentioned, that the United States remains a nation of believers, and despite the growth of the “nones,” the big majority of U.S. adults do identify with a religion. Three-quarters of U.S. adults identify with a religious group, primarily Christianity. And this group – those who are religiously affiliated – is about as religious today as when we first conducted the Religious Landscape Study in 2007.

You can see that here. Among those who are religiously affiliated, virtually everyone, 97%, say they believe in God, which has changed not a bit since 2007. Two-thirds of the religiously affiliated say that they pray every day and that religion is a very important part of their lives, also very steady compared with 2007.

And 62% of religiously affiliated Americans say they attend religious services at least once or twice a month, little changed since 2007. In some ways, religiously affiliated adults are perhaps even more observant today than when we first conducted this study. Forty-three percent of religiously affiliated people now say they read scripture at least once a week, up 3 points.

We also see small but significant increases in the share who say that they regularly participate in prayer groups or scripture study groups or other small religious groups, and who say they share their faith with others on a weekly basis.

One way to illustrate the stability in American religion is to shift from looking at percentages to looking at raw numbers. When we do this, what the data show is really very interesting, I think. And that is that there are about as many highly observant religious people in the United States today as was the case in 2007.

Demonstrating this can be a little tricky, given the declines in the share who are religiously observant, so I hope you will indulge me and let me walk very quickly through what I mean. I’m going to try to use the mouse here to help me.

In 2007, there were 227 million adults in the United States, and 83% of them were religiously affiliated, identifying themselves as Protestants or Catholics or Buddhists or as members of any number of other faiths. This means that there were about 189 million religiously affiliated people in 2007.

Our data show that 64.5% of them said that religion was very important in their lives. If we do the math, multiplying that 64.5% times the 189 million who were religiously affiliated, we get an estimate that in 2007, there were 122 million adults in the United States who identified with a religion and said that religion is very important in their lives.

Now, as of 2014, there are far more adults in the United States, 245 million in total. A smaller share of them are religiously affiliated, but we’ve still got roughly the same number of those who identify with a religion, and a similar share of them who say religion is very important in their lives.

Our 2014 data suggest that there are now about 123 million religiously affiliated adults who say religion is very important in their lives, almost exactly the same number we got in 2007, right? A picture of perfect stability. So there’s not much change here in the number of devout American believers.

What has changed in American religion is that there has been very rapid growth in the number of people at the other end of the spectrum, in the number of adults who are religiously unaffiliated and who say, furthermore, that religion is not important to them.

I’m not going to walk through all the math again, but let me just cut right to the chase. In 2007, there were 21 million U.S. adults who were religious “nones” who said religion was not important in their lives. By 2014, driven largely by generational replacement, the number of religious “nones” in the population who say religion is not important in their lives has swelled to fully 36 million adults. That is an increase of 15 million people in this category, religious “nones” who really are not particularly religious.

It’s this growth that is shifting the relative composition of the public. Religiously observant Americans constitute a declining share of the population, not because there are fewer of them, but because there are now so many more adults who don’t identify with any particular faith and who, furthermore, say that religion really doesn’t matter in their lives.

Those are some of the big, overarching trends that emerge from these data. The United States does, in fact, appear to be growing modestly, but noticeably, less religious. But there’s also a lot of stability in the American religious landscape. That’s the big picture.

But there’s a lot more of interest here. The data show, for example, that although those who are religiously observant in traditional ways account for a shrinking share of the population, certain forms of spirituality appear to be on the rise. We’ve seen a 7-point increase, for example, in the share of American adults who say they regularly feel a deep sense of spiritual peace and well-being, and a similar 7-point increase in the share of Americans who say they regularly feel a deep sense of wonder about the universe.

The data suggest, furthermore, that this increase in spirituality is not a replacement for more traditional forms of religious observance. In fact, feelings of deep – common feelings, regular feelings of deep spiritual peace and well-being are most common among those who are most religiously observant in traditional ways. And the increases on both of these measures are seen across the board, including among those who identify with a religion and among those who do not.

The data also show that religious institutions are widely viewed as a force for good in American society. Nine-in-ten Americans say that religious organizations help bring people together and strengthen community bonds. Nine-in-ten say that religious organizations play an important role in helping the poor and the needy, and three-quarters of Americans say religious institutions are vital to helping protect and strengthen morality in society. These views are even common among those who are religiously unaffiliated.

On the other hand, many Americans also express certain reservations about religious organizations. About half of the public says that religious organizations are too concerned with money and power, that they are too focused on rules and that they are too involved in politics. These views are especially common among the religiously unaffiliated.

Now, if we put these questions together, combining our questions about the positive roles and the negative attributes of religious organizations, the data show that 42% of the public, four-in-ten people, express mostly positive views of religious institutions. They agree with two or three of those positive statements and they disagree with one or fewer of the negative statements.

Half of adults expressed mixed views of religious institutions and relatively few people, fewer than one-in-ten, expressed mostly negative views of religious institutions.

I know we’ve covered an awful lot of ground here, and I want to get to our discussion, but I do want to reiterate and emphasize that we have really, truly just scratched the surface of the data that are available and of the questions that can be explored using these twin religious landscape studies.

In wrapping up, let me just point out that one key strength of the Religious Landscape Study is that the very large sample size allows us to analyze groups that account for a very small share of the U.S. population. We’ve looked at broad trends here in this presentation, and that’s what we’ve looked at thus far in our reports, but our reports and that data also have available information on denominational families like Baptists and Methodists and Lutherans, as well as on specific dominations like the Southern Baptist Convention, the United Methodist Church and the Episcopal Church.

We can analyze the data on groups that are as small as three-tenths of 1% of the population, including members of the African Methodist Episcopal Church; the Church of God in Cleveland, Tennessee; the Church of the Nazarene; the Presbyterian Church in America; Seventh-day Adventists and many, many more.

And one of the things that I am most excited about with this project is that we have made much of this data available through an online, interactive database that we hope you will find useful and informative. And I’d like to show you very quickly how to get to that database, what it looks like and how you might be able to use it.

I start here at our home page,, P-E-W From there, I can click right here at the top on the religion tab, which takes you – takes us to a page that includes all of our religion-related research. The Religious Landscape Study is right here at the top. I click on it and from here I can either read the report, which is here below the mouse or I can click over here where it says “Interactive Database: Explore Religion in America.”

The database has a ton of information in it, starting at the top with the religious composition of the country. Now scrolling down, I can click on any state or on any of about two dozen cities to learn about religion in those places. And scrolling further, I can click to learn more about any number of demographic characteristics or about religious beliefs and practices or about the social and political views of religious groups.

But scrolling back to the top, to the information about religious identity. Let’s say I’m interested in finding information about the United Methodist Church. This is how I can find information about specific denominations. Now, the United Methodist Church is a mainline denomination, so I can click on this yellow arrow here next to mainline Protestant to expand the category. I see all the constituent parts of mainline Protestantism.

I can see here that Methodists, within the mainline tradition, account for about 3.9% of the total population. And to see more detail, to find the United Methodists, I’m going to click on this yellow arrow to see the composition of Methodists within the mainline tradition. I can see now that the United Methodist Church is the largest Methodist denomination and because its name is in blue; that signals to me that I can click on it to learn more about that group.

When I do that, the page it takes me to is a profile of pretty much all the information we have about United Methodists in this study. It starts out with their demographics. You can read about their age distribution, about their gender, about the racial and ethnic composition and about any number of other demographic characteristics.

Below the demographics, we come to religious beliefs and practices. We can see that 71% of United Methodists in the United States tell us they believe in God with absolute certainty, and 61% tell us that religion is very important in their lives.

Scrolling further still, there’s a lot more information on religious beliefs and practices, and below the religion material we come to information on the social and political values of United Methodists, including things like their party identification, their ideology, their views about the proper size and scope of government, their attitudes about homosexuality, abortion, immigration, the environment, you name it.

The data also allow you to look at trends. You can click on the trends tab and see how the characteristics of United Methodists have changed since the first time we conducted this kind of study. You can even click on the compare option to see how the attributes of United Methodists compare to those in other denominations within the mainline tradition.

So you can really do an awful lot with this interactive database. In addition to exploring data on specific denominations, you can look at the religious characteristics of states and cities and demographic groups. You are really only limited here by sample size.

If we were successful in obtaining 100 interviews with members of the group you are interested in, whatever it is, chances are that you will be able to find it on this website. So we really hope that you will find this resource helpful moving forward, and even after this presentation.

I also hope you will stay tuned for some of our forthcoming reports. We do plan to continue to put out reports on the Religious Landscape Survey moving forward. We plan to explore things like lived religion – how do people incorporate their religion in their day-to-day lives, over and above their participation in typical religious activities?

We plan to look at topics like people’s religious background, how many come from a mixed religious background in the United States. We plan to look at the link between geographic mobility and religion. How does geographic mobility interact with people’s religion? And we intend to continue to follow up with profiles of lots of interesting religious and demographic groups.

So there’s a lot here and there’s a lot still to come. With that, I think I’ll stop. I thank you for your attention and I’m looking forward to our discussion.

Alan Cooperman: Greg, thank you very much.

Let’s bring John Green in the conversation. John, you want to come have a seat? John, as I mentioned, is a professor of political science at the University of Akron.

John, do you want to just share some of your thoughts about the most interesting findings here?

John Green: Certainly. Thank you, Alan. It’s wonderful to be here today to talk about this interesting study which we have done here at the Pew Research Center twice in seven years, and hopefully will do it in the future again.

I think the most important overall finding of this study is the increasing religious diversity of the United States. Greg has talked eloquently about one aspect of that, which is the growing group of unaffiliated Americans, many of whom are consistently secular in their religious beliefs and behaviors. But there is much more to the increasing diversity of the American population than just this one very important trend.

For instance, there is an important increase in racial and ethnic diversity in the United States. The country as a whole is becoming more diverse in those ways, and every religious group shows an increase of racial and ethnic diversity, whether those are Christian groups or non-Christian groups.

This study also shows a continuing increase in the number of people affiliated with non-Christian faiths, something that’s been happening for a long time but continues to happen, as amply demonstrated in these data.

But it’s also important to remember that within the religious stability that Greg talked about, the fact that religiously affiliated people have about the same level of belief and religious behavior as they had seven years ago, there is an enormous amount of diversity. Tremendous diversity within Christianity. And Greg touched on some of those differences in his last description of the interactive website.

Many of those differences persist in the present, and in some cases they have actually grown a little bit bigger. And within those religious communities, there are lots of differences by demographic characteristics. The United States, seven years ago, was arguably the most religiously diverse country in the world and certainly the most religiously diverse among the advanced industrialized nations. And seven years later, it has become even more so.

Alan Cooperman: Thank you.

Let’s open it up for questions. And for those who are watching on the live stream, feel free to join the conversation. Please send your comments or questions to One more time,

For those in the room, we have a microphone floating around. If you could just wait until you get the microphone, and if you’d like to identify yourself that would be helpful to others.

Thank you.

Kip Banks: Kip Banks, Progressive National Baptist Convention. And certainly on behalf of our convention, we want to commend you for the research. We find it very useful in our attempts to reach, in particular, more of our younger persons that are out there.

Back in August, the Religion News Service, Adele Banks reporter, wrote an article on some of the findings. And the headline said African-American churches are bucking the decline with respect to the “nones.” I’d be interested in the insights that you may have on that, Greg, or others.

Gregory A. Smith: Yes. I read the article. I thought – and I think, if I remember correctly, I think we talked with Adele about that and tried to help with it.

It’s interesting, and I’d say it kind of depends on how you look at the data. If you look at the share of Americans who identify with denominations within the historically black Protestant tradition, there we see a picture of real stability. We see declines in the share of Americans who identify with Catholicism and the share of Americans who identify with mainline Protestantism, even a very tiny downturn in the share who identify with evangelical Protestantism.

But the share of Americans who identify with historically black Protestant denominations is absolutely steady. If I remember correctly, I think it was 6.9% in 2007 and it’s 6.5% today, very much within the survey’s margin of error. So in that sense it’s true, the growth of the “nones” in the population does not seem to be occurring at the expense of churches within the historically black Protestant tradition.

However, there’s another way of looking at the data, and that’s if we look at African-Americans as a whole. One thing that’s really interesting about the growth of the “nones” is that it is very demographically broad-based. The “nones” are growing among men and women, they are growing in every region of the country. They are growing among college graduates and those with less education. They are growing among African-Americans as well as among whites and Latinos and Asians.

So in that sense, it’s not a case that the African-American community is sort of immune, so to speak, from the growth of the “nones.” So it depends on how you look at it. It’s not occurring at the expense of the size of the historically black Protestant tradition, but it is occurring within the African-American community.

John Green: If I might just add one point to that. There’s also a larger number of African-Americans outside of the historic black Protestant churches. For instance, this survey showed that almost a quarter of members of evangelical Protestant churches are African-Americans. So there’s been a sort of dispersion of the African-American population into a variety of different categories.

Alan Cooperman: Questions? Folks, don’t be shy. Yes, sir?

Daoud Nassimi: Regarding the number of Muslims in the United States of America, I’ve seen estimates 7 to 8 million, 6 million, and most of them Muslim organizations in the country have given those kind of figures. I have seen also statistics such as there’s like 3 million or so and 1.9% of the country’s population.

Can you give us some explanation if Muslims came up with 7 to 8 million? What is that based on and how that can be denied or 3 million can be confirmed?

Gregory A. Smith: Well, if it’s OK, I might put my colleague Besheer Mohamed on the spot and ask him if he can remind me – I always forget this – what our official estimate of the number of Muslim American adults is. Do you remember that?

Besheer Mohamed: I think it’s in the 2 to 3 million range. I don’t know that we have – we don’t have an updated 2015 number.

Gregory A. Smith: Right. And that’s from 2011?

Besheer Mohamed: Our 2011 number was 2.75 million and all our projections suggest it’s been going up.

Alan Cooperman: And that included children – 2.75 million included children.

Gregory A. Smith: Includes children, right.

I cannot speak to how other organizations might come up with estimates of the Muslim population or any number of other religious groups, for that matter. There are different ways of counting. I know that sometimes efforts are made to count congregations, the members within congregations, or members within mosques in this case. Or other methods of counting.

I guess what I can say is that every indication that I have seen, every estimate that I’ve seen of the size of the Muslim population that comes from scientific surveys that are done among a representative sample of the public and ask people about their religious identity. Those tend to come up with smaller estimates than those 7 to 8 million range.

We consistently find that a little less than 1% of American adults self-identify their religion as Islam. Now, the estimate does go up a little bit when you do surveys in more languages than in English. We have done two surveys of Muslim Americans, one in 2007, one in 2011, and we’ve done those surveys in Arabic, Farsi and Urdu in addition to English. When you do that you do – not surprisingly – get a higher estimate of the share of Muslims. But it’s marginally higher. It doesn’t get you up to that 7 to 8 million range.

We do see evidence, no matter how we look at it, that the Muslim American population is growing. We see that occurring steadily in our surveys when we ask people about their religion. We also know that the Muslim American population – I think I’m going to be correct in saying this – is younger on average, and so as Muslim American children enter into adulthood, then the share of adults who identify with Islam is going to continue to grow.

But every indication that I’ve seen that comes from a representative survey puts that at a little bit lower than that 7 to 8 million range.

Alan Cooperman: I can add a little bit more detail. In fact, our estimates are based both on survey incidence rates – that is, how many of the people that we call self-identify as Muslim – and we’ve also taken a look at U.S. immigration records, which do not track religion, but they do track country of origin. Then we can cross-pollinate with surveys, for example, Princeton University’s New Immigrant Survey, which asks people of various countries of origin about their religion.

Using a combination of those two techniques, which we explain in our reports, we’ve come up with our estimate, which is today in the range of 3 million Muslims of all ages in the United States.

Shirley Hoogstra: Shirley Hoogstra from the Council for Christian Colleges and Universities. I’d like to know if you have any advice about drawing fair conclusions from your research. So, for instance, the stability around religiously affiliated, but also perhaps you could conclude that religious practices actually enhance or keep stable religious affiliation. Would that be a fair conclusion?

John Green: To some extent, yes. The problem is that the countertrend also happens. It could very well be that – and we see evidence of this not so much in this survey but in other surveys that have re-interviewed the same people over time – it’s called a panel study. And what those show is that many people who are actively engaged in their faith, whether it’s worship attendance or prayer or whatever, are much more likely to maintain their religious affiliation.

There’s a small number of those people that change their affiliation but they don’t become unaffiliated. So there is an argument that – there is evidence for what you suggest, that religious activity maintains affiliation.

But there is some counterevidence as well that when – Greg illustrated that people move – we have seen a large number of people moving from some kind of affiliation to non-affiliation, but there is a countermovement. There are people who were raised or who at one point became unaffiliated who became affiliated with a religious group. Typically, there, the affiliation precedes religious behavior. So it kind of goes both ways.

One way to think about it is, it depends for many people at what point you observe them. If you observe them while they are involved, where activity helps maintain affiliation, or do you observe them before they become involved. And becoming a convert or becoming recruited to a religious affiliation often then leads to behaviors and beliefs. So it works both ways.

Gregory A. Smith: Maybe just to follow up. One question that we’ve gotten a lot about these data, and as we’ve talked about these data is, so what – given the importance of generational change and generational replacement – what are we to expect for the future? Will the young adults who today are not particularly religious, will Millennials who are unaffiliated today, will they come back to religion? Will they become more religious as they have families, as they have children, as they settle down?

That’s a very hard question. It’s a very interesting question, and to be honest, we can’t answer it. We don’t have a crystal ball. The future could hold any number of possibilities.

If we look back at existing data, we do see that there are many ways in which people tend to become more religious as they age. They tend to become more prayerful as they get older. People tend to become more likely to attend religious services as they get older. So there are ways in which people tend to become more religious as they age.

But people do not become more likely to identify with a religion as they get older. You can see that here in this chart. These data come from the General Social Survey, which is a high-quality academic survey conducted for the last four decades or so by the National Opinion Research Center at the University of Chicago.

What these data show are the percentage of people who identify as religious “nones” – as religiously unaffiliated – over time, by generational cohort. So check out that middle line there, the blue line. That shows Baby Boomers.

When Baby Boomers were just entering adulthood in the late 1970s, 13% of them said they were religiously unaffiliated, that they were “nones.” Today, with Baby Boomers entering retirement, 15% are religiously unaffiliated. So the number of Baby Boomers, the share of Baby Boomers, who are “nones” is almost perfectly flat. If anything, maybe it’s even ticking up modestly in recent years.

You see the same thing for Generation X. When they entered adulthood, 20% were religiously unaffiliated. Now that they are in their late 30s and 40s – I don’t know if any of them are in their 50s yet or not – 21% are religiously unaffiliated.

Then look at the top. Look at Millennials. They are entering adulthood with a rate of dis-affiliation from religion that far exceeds anything we have seen in the era of scientific survey research about religion. They, if anything, have become more unaffiliated as they have moved from their late teens and early 20s to their late 20s and early 30s.

Does this mean that there is no prospect that they are going to come back? No. We can’t predict the future. Anything is possible. But the data suggest that this is not one of the ways in which people tend to become more religious as they age.

And in our data, we only have seven years’ worth of trend data, but in our data over those seven years we see very little evidence that older Millennials, those who were just entering adulthood in 2007 and are now in their late 20s and early 30s, there’s no evidence that that group has become more religious in the last seven years. If anything, it might be just the opposite. They might actually be growing less religious as they get older.

So what will the future hold? I don’t know, but these are some of the things that are suggested by the data we have on hand.

Jessica Hamar Martínez: A question from email here: Are the “nones” growing because of the entanglement of religion in politics?

John Green: There is some evidence that the Millennial generation in particular, but also older individuals as well, have become less affiliated with religion because of the broader involvement of religion and politics. So yes, there is some evidence to that.

There are other factors, though, that make a big difference as well. For instance, younger people, the Millennial generation, are more likely to go to college and move away from home. They are more likely to be geographically mobile after college or after they enter the workplace. They are also more involved in social media and in popular culture.

So there are many, many factors that may be contributing to this, but there is no doubt that one of the factors is the perception that religious institutions of all kinds, on all sides of the political issues, have been too involved in the political process.

On one of the slides that Greg just showed you can see real good evidence of that. The unaffiliated population are particularly critical of religious institutions because of their involvement in politics.

Gregory A. Smith: Just following up on that a little bit. Part of the thinking is that the reason that politics are driving young people in particular away from religion is that religion has become caught up in the popular imagination with a particular kind of conservative politics.

So you have young people and others who may not share those politics and they think of themselves, well, if being a religious person means being a political conservative, then I must not be a religious person. That’s the idea behind this connection between politics and disaffiliation from religion.

You know, our data can’t prove that that’s the main thing or the only thing that’s happening, but there is some evidence to suggest as part of the picture, as John mentioned, just a couple things I would highlight, are changing attitudes about homosexuality.

Our data clearly show that acceptance of homosexuality is increasing pretty much across the board, including among religious groups that have traditionally been least accepting of homosexuality. We have seen a 12-point increase, for example, in the share of Mormons who say that homosexuality should be accepted by society and a 10-point increase in the share of evangelicals who say that homosexuality should be accepted by society.

Now don’t get me wrong. Evangelicals and Mormons are way more conservative on this issue than any number of other religious groups, but they are moving in a more accepting direction.

The data will also show one more point, that religious “nones” are quite socially and politically liberal, and they vote overwhelmingly for Democratic candidates for public office. In fact, the data show – and this is really interesting. John and I are both political scientists and I think we find this fascinating – that religious “nones,” the religiously unaffiliated now account for 28% of all Democrats and those who identify with the Democratic Party. That makes them the single largest religious group within the Democratic coalition.

They are more numerous among Democrats than Catholics, they are more numerous than mainline Protestants. They are more numerous than members of the historically black Protestant tradition. They are more numerous than any other single religious group among Democrats. That’s a big change, and an important one, I think, for American politics.

Alan Cooperman: Although, if we were to amalgamate all Protestants, all Protestants would be larger. And still, it’s a minority of all Democrats and Democratic-leaning voters who are religiously not affiliated. So we are not trying to – to tar the Democratic Party as a party of non-religion. Nonetheless, OK, some people have unfortunately read that into some of these findings.

Greg, some people have wondered whether the rise of the “nones” is actually a change in anybody’s behavior or whether it is merely a change in the social desirability of identifying with a religion. Essentially, whether people who were nominally affiliated but actually weren’t going to church, actually were not very religious, have simply given up the pretense of still calling themselves X, Y or Z.

What do you think?

Gregory A. Smith: Well, I think that’s part of what’s happening. That’s probably part of what’s happening. You’ve always had people who identify as Catholics or Lutherans or Methodists who, in fact, are not particularly religiously observant. And it may be that those people increasingly describe their religious identity in a way that more closely matches their religious practice than they might once have done. That could be part of it.

But I don’t think that’s all of it, or even the biggest part of it, and there’s two reasons for that. No. 1 is, at the same time that the “nones” are growing, the share of people who are not particularly religiously observant, by any number of ways of measuring that, are also growing.

So it can’t be just that not particularly religious folks are changing their identity. It’s the share of folks who are not particularly religious is itself growing. So that’s a big part of what’s happening. There are fewer religious people as a share of the population. It’s not just a matter of identity.

There’s one other important thing to keep in mind, though. I think sometimes when people think about the growth of the “nones” as a function or as the product of nominally religious people changing their identity, they are thinking about individuals who are changing their identity or who are changing as individuals. And no question, that’s part of what’s happening.

But a major part of what’s happening, we have to remember, and if there’s one thing that you leave this room with, I might suggest – I hope – it’s this. So much of this is generational. It’s not that people have changed their religious identity in the last seven years. It’s not that people in their 40s and 50s and 60s are now changing the way they describe themselves.

It’s that people in their 20s describe themselves in ways that are far different than the way their parents and their grandparents did before them. So this is generational as much or more so than it is about individuals changing their identity.

John Green: Let me just add something very briefly there. There are still lots of nominal affiliators, and part of the stability that Greg talked about is within all the religious groups, there is still a significant, almost the same number as in 2004, number of people who identify with a religious group but are not particularly observant.

Lyman Smith: Lyman Smith, Military Chaplains Association. You are looking at this – if we look at it from an institutional perspective, how much does the lack of affiliation of the Millennial generation with the institution of religion correspond with lack of affiliation with other institutions?

I say that specifically thinking about veterans organizations and others that have traditionally had major populations, and we are having significant issues in trying – in terms of trying to get younger veterans to affiliate with some of the same sorts of things.

Alan Cooperman: Oh, that’s interesting. You know – Lyman, thank you. That’s a great question. One of the additional theories about the rise of the “nones” is the theory that I would encapsulate with two words: “Bowling Alone” – the title of Harvard Professor Robert Putnam’s very well-known book from a few years ago that illustrated a decline in social capital in America writ large.

And part of what one of the theories is saying is that, folks, this isn’t just about religion. This is about a drawing back of Americans from all kinds of associations with all kinds of bricks and mortar institutions. So it’s not only veterans clubs that are down, it’s also fraternal organizations. It’s also all kinds of – other than maybe health clubs – all kinds of places where people used to go to gather.

And the bowling alone example is that bowling leagues, which used to be very popular, have declined. Bowling has not gone down. Many people still bowl, but they don’t bowl in bowling leagues the way they used to.
Part of this could be – part and parcel of this could be the rise of the internet, which could be giving people ways to have virtual relationships that are taking the place of going physically to join others.

But it’s interesting to note, in the GSS, General Social Survey, data that Greg showed earlier, if we go back all the way to 1972 and we start looking at affiliations, we see the rise of disaffiliation beginning in the late 1980s and early 1990s, even before most Americans had access to the internet.

So it’s a phenomenon that may have preceded the internet. It may – it’s possible that it is a phenomenon also that is accelerated by the virtual media.

John Green: I might add that the Millennial generation is also not as engaged in political organizations as older generations. So there’s clearly a generation effect there where it comes to a wide variety of institutions.

Shirley Hoogstra: So this question kind of follows up from the chaplain here about the leading and the following piece. So as we have a rise in suspicion around institutions, or as you just talked about the ’80s and ’90s, it seems to me that, you know, starting with Nixon and Watergate and then starting with the intrusiveness or the publication of much more detail around people’s lives, you – you tend – there’s a population that is just highly – we don’t even believe that anymore.

You’ve got movies, you’ve got – what they say about President Kennedy, what kind of information would have come out about President Kennedy. And now any politician is – anything in your life is up for grabs.

How does all of that affect how people connect with religion, which is a faith-based, or trust-based concept?

Alan Cooperman: I think the short answer is you are correct. In Pew Research Center polling about trust in institutions, we find a general decline in trust in most institutions – major institutions in American life, declining trust in many professions as well.

Interestingly, the military is one big exception to that – there is steady, or even rising, trust in the military, but declining trust in religious organizations, as well as in the medical profession and many other realms of American life, including Congress, of course.

I’m afraid that will have to be the last question. I want to thank you all again for being here today. I want to also let you know that if you are interested in a tailored briefing for your particular denomination or religious group, please contact us. We will try to provide the information you would like.

And in general, please look at our website,, click on that tab on religion. Sign up to receive our weekly newsletters. And be in contact with us. We are here as a resource for the American public, and that very much includes religious leaders of all varieties. So thank you very much for coming.