Religion in America
updated March 2021
About Religion in America. In 2012 Stu Johnson was the lead author of a report on the Religious Publishing Marketplace in America, produced by Simba Information for Zondervan/HarperCollins Christian Publishing. The chapter on Religion in America, a summary of survey research, some of it going back many decades, was compiled solely by Stu and subsequently extracted as a separate report.
About this page. The graphs on this page are selected slides from a more detailed PowerPoint® presentation that illustrates major trends in the American religious landscape. You can click on a graph to enlarge it. (It will open in a new tab.)
The primary data sources used in our research include Gallup, Pew Research, and Barna Group. Data sources are shown on each chart. For some charts, the survey is repeated annually, others more sporadically, Each chart shows the latest data available at the time of the last update of this page.
- In the U.S. the number of Christian adherents continues a decline that became noticeable around 1970 and is accelerating in the early decades of the 21st century.. Those claiming Christian affiliation (Protestant, Catholic and non-denominational Christians) accounted for 93% of the adult population in 1964. It was 83% in 2000, then 68% in 2020. Meanwhile, those claiming no affiliation have been steadily climbing from 2% in 1964 to 20% in 2020 according to Gallup and 28% according to Pew.
- While Christian adherents are declining in the U.S., those who describe themselves as "born again" or "evangelical" remain more constant. Gallup starting asking the question in 1992, when 36% of adult respondents claimed to be born again or evangelical. The number peaked at 47% in 1998, then began moving down, ending at the low point of 34% in 2020, but that is only 2 points lower than where it started 28 years ago.
- Breaking the numbers down, it appears that most of the loss in Christian adherents is occurring among those who might be identified as "cultural" or "nominal" Christians, whose adherence has been driven by tradition more than personal belief and practice. This is clearly seen in the shrinkage of mainline denominations.
- The significant "rise of the nones" in the United States—those who claim no religious affiliation—has been accelerating since 2000 and is most noticeable among the youngest generations. A study in January 2018 by Barna, which includes 13-18 year-olds in Generation Z, shows a doubling of those identifying themselves as atheist (13%, compared to 6-7% for all older cohorts, including Millennial and Gen X).
- Americans have shown a steady erosion of confidence in major institutions—including organized religion— tracked by Gallup since 1973, but 2020 saw several changes that could be attributed to the year of the COVID-19 pandemic, the aftermath of the Floyd Brown death at the hands of Minneapolis police, and the 2020 presidential election.
Organized religion was number 1 from 1973 through 1985. In 2020, it bounced up to 42% positive response from a low of 36% in 2019 and rose one rank to number 5 among the 16 institutions measured. That was because Medical Systems—representing some of the heroes of the pandemic—shot up from 36% and Nr. 7 in 2019 to 51% and Nr. 3 in 2020 (its highest rank and confidence level since Gallup introduced it in 1975). In response to the George Floyd-inspired protests, Police dropped one rank to Nr. 4 and hit its lowest confidence level (48%) since being introduced to the survey in 1993.
- Globally, Christian adherents represent the largest group, at 31%, followed by Muslim, Unaffiliated, Hindu, Buddhist, Other and Jewish. According to a 2018 Pew study,by 2050 Muslims will catch up with Christians as a proportion of world population, together claiming six in ten adults. The growth of unaffiliated will impact the U.S. more than the global population, according to the study.
The U.S. compared to the World
While Christian adherents have been declining, especially in the last two decades, the United States remains a predominantly Christian country—with 65% of adults identifying themselves as Christian in the most recent Pew polling. (Gallup shows 68%).
Around the world, according to various sources, Christians represented about half that level, with roughly three in ten of adults in 2020.
Pew suggests 28% unaffiliated in the U.S. in 2020, while Gallup puts it at 20%. The difference cannot be explained simply by margin of error, but by differences in survey methodology, question wording and interpretation of results. More on that as we dig deeper.
Though declining, Christian adherents are expected to dominate in the United States into the foreseeable future. By 2050 Pew predicts a drop from three-quarters to two-thirds of adults. In reality, we reached that level in 2018, according to Gallup and by 2020 it dropped to 65%.
Non-Christian religious adherents in the U.S. will increase some, but the most noticeable shift will be the continued increase of those with no stated affiliation. However, when looking at global projections, notice that Muslim adherents could match Christian by 2050. Most others, including unaffiliated, are seen as remaining the same or declining.
The changing mix
A 55-year trend
For much of the 20th century, Protestantism dominated the American religious landscape, with its adherents representing 70% or more of the adult population. Catholic adherents accounted for another one-quarter of adults, putting the combined total of Christian adherents well over 90% for many years.
Following World War II, Protestants and Other (non-Catholic) Christians declined from a steady 70% majority to just below 50% for the first time in 2012, according to Gallup. In 1998 Gallup began including "Other Christian" with Protestant. This reflects the rise of non-denominational churches as well as churches that do not promote their own affiliation with a denominational organization, so congregants may not know there is a denominational connection. Denominational Protestants fell below 50% of the population in 2006, while all Protestants hit that mark in 2012. Total Protestants fell to a low of 45% in 2019, but came back to 46% in 2020. Whether that is related to COVID, the presidential election, or other factors is simply too close to call when the margin of error alone could produce a swing of a percentage or two from year to year. It is interesting, however, that a number of measures that had been trended down saw a noticeable upturn in 2020.
Catholics have held steadier, at roughly a quarter of the adult population, but show a steady decline from a peak of 29% in 1982 to a low of 21% in 2019, followed by a slight bump up to 22% in 2020.
"Nones" have been steadily growing from practically non-existent status in the 1960s to breaking 10% in the first decade of the 21st century, then approaching or exceeding 20% in the second decade, depending on the survey. (Pew has shown Unaffiliated as high as 28%, though its estimate of total Christians (Catholic and Other Christian) is fairly close, at 65%, compared to Gallup's 68%. The difference, while notable, may be attributed to such factors as polling methodology, wording of questions and interpretation of responses.
Initially, the increase in unaffiliated (commonly called "the rise of the nones") reflected a rejection of organized religion, with some "nones" remaining spiritually open. More recently, however, especially among younger generations, we are seeing growth in those who identify as agnostic or atheist. (See the charts on Generation Z below).
Belief in God
The trend indicated by this chart, developing since 2000, is very obvious (and corresponds to the "rise of the nones" discussed throughout this report). This is the longest-running of the Gallup questions on religion, going back to 1945. As you can see, belief in God was nearly unanimous from the end of World War II to the last decade of the 20th century. Since then, it has begun to erode noticeably. When first asked, the question referred only to "God," but in more recent years it has referred to "God or a Universal Spirit."
"Born again" and "evangelical"
Since 1992, Gallup has asked people if they describe themselves as "born again" or "evangelical." As a self-description there is plenty of latitude and I believe the term "evangelical" has been co-opted by the media in political analysis since the 2000 presidential election, applied to what in the 1990's was called the Religious Right. That designation was represented by Jerry Falwell's Moral Majority and similar movements growing out the Fundamentalist stream of American religion.
George Barna developed a strict definition of evangelical, based on beliefs and behaviors, which narrows the group down to about 8% of the American adult population. Beyond those who fit this narrow definition, a third or more in Barna surveys have described themselves as "born again."
During the early 2000's, I was involved in research for Tyndale Publishing House during the height of the popularity of the Left Behind end-times series. During that time, Tyndale commissioned research, part of which indicated that Catholics compromised a good part of the audience for that series and many of them used "born again" to describe themselves. Therefore, while "evangelical" generally refers to a subset within Protestantism, those who identify themselves as "born again" can be found across Christian adherents.
Needless to say, it is interesting to note the rise of this group from the start of the question in 1992 at 36% of all respondents) to a high of 47% in 1998. From there, the trend has been slowly downward, but the low of 34% in 2020 is barely below the starting point back nearly 30 years ago.
Compared to other indicators with far more significant change, the trend here is remarkably level. In fact, the proportion of born-again/evangelical among Christian adherents (including Catholics), at 52% in 2020, has actually grown, from 44% in 1992 to 52% in 2020. This could be attributed to the idea expressed by Ed Stetzer and others that "the rise of the nones" actually represents a winnowing out of nominal Christians for whom faith was more form and tradition than substance.
The changing mix: Other religions
The data comes from different sources. Because the numbers are small (within the error of margin for most surveys), we can only make a few general observations.
Clearly, there has been some increase in Other religions, yet they still represent a very small portion of the adult population in the U.S., between three and five percent.
Jews have remained between one and two percent in surveys since 1990.
Muslim adherents remain a smaller segment than many people might assume, given exposure in news and politics (and confusion between religion, ethnic background, and national origin). The growth shown by 2010 seems to have settled back to just under two percent.
Adherents of Other non-Christian religions (which represent many groups) are growing, up three-fold since 1990, and greater than the combined total of Jews and Muslims.
None does not equal Atheist, but does mean young
Prior to 2018, a breakdown of those who claimed no religious affiliation showed some growth of those who identify themselves as Atheist and Agnostic, but the greatest growth was among those who simply have no preference. (Compare this chart with the earlier one showing Gallup results, which shows unaffiliated ("None") about 5-8 percentage points below Pew throughout the time covered here).
From 2008 through 2016 Pew divided unaffiliated ("nothing in particular" in this chart) into Secular and Religious Unaffiliated. Starting with a nearly even split in 2008 (close to 6% each), by 2018 the gap widened to 14% Secular and 3% Religious.
The presence of the seemingly paradoxical "Religious Unaffiliated" is an indicator of the discomfort that a number have had with organized religion, especially as the rise of the Nones began to explode. In 2008, Pew found that when Unaffiliated were asked a number of questions about spiritual issues, they could be divided between Religious—those who claim that spiritual factors are important to them—and Secular—those who claim little or no importance for spiritual things.
It appears that those with spiritual interest began to decline, so the most recent division between Atheist, Agnostic, and Nothing in Particular shows how significant has been the move away from organized religion and religious belief, especially in the second decade of the 21st century.
When viewed by generation, it is striking to note the accelerating growth of the unaffiliated with each succeeding generation. The proportion moves from 10-percent or less unaffiliated for those born before the end of World War II in 1945, to more than a third of the youngest generation.
Note that Pew shows unaffiliated Millenials in 2018 ten-percent above a similar breakdown by Barna in the next chart. In the other age cohorts, Barna shows higher unaffiliateds than Pew by a smaller margin. The lesson here is that survey research is most useful to show trends over time than a high degree of precision at any one time. As already pointed out, this is often due to differences in research methodology and how questions are framed and interpreted.
Meet Generation Z
In a study released in January 2018, Barna Group revealed a startling indicator of the depth of age-related disconnect with organized religion.
The youngest age cohort, known as Generation Z, or Gen Z, shows a 5-point increase over Millenials in not claiming a religious affiliation. Most significant: all of the increase and more came from those claiming to be atheists.
The Barna report reveals much more detail, which was the subject of an InfoMatters blog, "Generation Z," posted in February, 2018.
The Gen Z study asked about respondents' views of science and the Bible. In general, younger generations side with science or view science and the Bible as independent, expressing different aspects of reality. A diminishing proportion see the Bible and science as complementary, while the number who side with the Bible when there is apparent conflict increases with the youngest generations.
This may seem counter-intuitive, but could be due to the strong influence among some home schoolers and Christian academies, "young earth" groups and others taking exception with scientific assumptions about the origins of life and the age of the earth.
An interesting parallel to the Science and the Bible chart above is data from Gallup on views of the Bible.
What I find striking is that those who perceive the Bible as the inspired word of God have remained fairly steady since 1976. On the other hand, those who take a more literal view have declined by 12%. While it does not represent the movement of individuals, the decline is not matched by a growth in Inspired, but rather by an 11% increase in those viewing the Bible as Fables & Legends. The two lines converged in the latest survey in 2017. Unfortunately, the survey has not been done again, which would likely show those two lines continuing to move further apart, in opposite directions. [Unlike other charts, which represent annual surveys, the colored dots show the value in years the survey was conducted, with the solid lines showing the overall trend.]
The importance and influence of religion in America
How important is religion in your life?
Both "very important" and "somewhat important" show declines over 25 years. "Very important" is off 13% from the peak of 61% in 1998 and 2003. "Fairly important" fell 7% from its starting point of 29% in 1992 to its low of 22% in 2016. From there it has rebounded slightly, ending at 25% in 2020. "Not very important" moved up about 15% in nearly three decades, crossing the slowly declining "fairly important" line in 2016, matching it in 2019, then moving ahead again by two points in 2020 (remember, however that this could easily be within the margin of error for such research). This exceeds the "rise of the nones" in the same time period, which went up 10% (some earlier converts to none could have shifted from religious to secular unaffiliated).
Ability of religion to address today's problems
The decline for "yes" is greater than that of personal importance on the previous chart—a decline of 18% since the peak of 67% in 1999. However, Yes moved up 4% in 2019 to 53% and stayed there into 2020.
Those who felt that religion "is largely old-fashioned and out of date" ("No" on the chart) rose fairly steadily by 17 points, from 21% in 1994 to 38% in 2018. As with Yes, it reversed course in the next two years, ending at 35% in 2020—a likely temporary change of direction in the overall steady climb. What is interesting is that the low for Yes and the high for No both appeared in 2018, with the change of direction occurring before COVID and other events of 2020. Over the next few years, that may level off into a blip in time. How long now before the No's outnumber the Yes's?
The influence of religion
Here, the change is much more dramatic, with strong belief of increasing influence through 2001 followed by an equally dramatic decline to 2009, when it leveled off in the low 20% area, hitting the low of 19% in 2019. Then, an up-tick of 14% to 33% in 2020 mimics a similar jump in 2000-01. While the larger spikes may be related to 9/11 and COVID-19 as indicated by the markers, the overall trend is likely due to much more complex cultural and organizational associations than the more personal "problems" question. The trend for "losing" influence is a mirror image given the binary choice. (Also seen more subtlety in the previous chart on the ability of religion to answer questions where there is a stronger presence of "other" responses).
Membership and attendance
While membership has fallen faster than attendance, the importance of membership (which requires a formal process in most religious bodies) has itself been in decline. Some are content to attend without being official members. Furthermore, while most churches may still distinguish between members and attenders, it is unclear whether respondents make that distinction. At any rate, it is clear that immediacy of attendance (in the last seven days) is lower than having a sense of belonging since it declined by less than half (10%) that of membership (23%).
Confidence in religion and other American institutions
In 1973, the first year Gallup conducted its annual measure of confidence in major institutions, "the church or organized religion" was rated the highest among 15 institutions. Religion continued to rank first through 1985, then fell below 60%. Among reasons: Televangelist scandals, child molestation by priests; then a general decline in religiosity continued the downward trend (as scandals in Catholic and Protestant churches continued). Religion dropped 47% from its high of 68% in 1975 to 36% in 2019). For all institutions, confidence fell 33% from a high of 48 in 1979 to a low of 32% in 2014. As we have seen with other measures, there was a noticeable upturn in 2020, most likely attributable to responses to the COVID-19 pandemic, as evidenced in the following charts (which also show a downturn for Police and TV News, which reflect more on the turbulence of the election year and response to the George Floyd killing by Minneapolis police. Because the lines are quite erratic from year to year, it will not be surprising to see the downward trend continue in future years.
In 2019, religion ranked sixth among 15 institutions, even though it had fallen further than any of the institutions included. In 2020 it regained six points, to 42% confidence, bringing it up to fifth rank.
In 2020, Large Tech was added, bringing the list to 16 institutions. (It has always had 15 or 16, with half of them on the list since 1973).
Reflecting the events of 2020, Small Business and the Medical System reach their highest level of confidence, while Police and TV News hit their lowest levels. While 10 of the 16 have had 50% or more confidence at some point, in the 2020 survey only 3 of 16 had the confidence of 50% or more of those surveyed.
Falling more than 20 points from their high mark to 2020 are seven of the 16 institutions. Except for TV News, which hit its low in 2020, the others have been lower than they were in 2020. All but two (TV News and Congress) did hit highs above 50%. None are over 50% now.
- The Presidency - 33 points since high of 72% in 1991 - - average 42% - 39% now
- Congress - 31 points since high of 42% in 1973 - average 24% - 11% now
- TV News - 28 points since high of 46% in 1993 - average 28% - 18% now
- Religion - 26 points since high of 68% in 1975 - average 55% - 42% now
- Newspapers - 24 points since high of 51% in 1979 - average 33% - 24% now
- Banks - 22 points since high of 60% in 1980 - average 40% - 38% now
- Public Schools - 21 points since high of 62% in 1975 - average 41% - 41% now
What does the overall decline of confidence say about the future of our society when we have so little trust in our bedrock institutions? What can the Christian church (and other faith communities) do at the individual and local level to restore trust when organized religion itself has fallen so far?
Defining religious cohorts
2009: George Barna's Seven Faith Tribes
Casual Christians (66%) - Rather than allowing the Christian faith to shape their minds and hearts, Barna suggested, they have chosen to fit Christianity within the box they have created for it.
Captive-Christians (16%) - They see themselves first and foremost as spiritual beings. They believe their lives are to be lived out in the flesh, but for spiritual purposes and on the basis of biblical principles. [The name Captive Christians the name comes from the idea expressed by the Apostle Paul that he voluntarily took on the role of a slave of Jesus Christ.]
One quarter of the Skeptics adopt the label “deeply spiritual” (which is distinct from being religious). Nearly half of the adults in this tribe describe themselves as faith seekers who are unattached to an organized body of faith or theological perspective.[If these seekers represent the religious unaffiliated described by Pew, their significance has gone down dramatically since 2009.
2018: Pew Research Religious Typology
Research conducted extensive interviews using a battery of questions that covered a wide range of religious and non-religious behaviors and beliefs. Cluster analysis was used to look for significant groupings that represent the American adult population.
HIGHLY RELIGIOUS (39%) are represented by three subgroups:
- Sunday Stalwarts (17%) - Religious traditionalists actively involved with their faith and engaged in their congregation.
- God-and-Country Believers (12%) - Socially and politically conservative, most likely to view immigrants as hurting American culture.
- Diversely Devout (11%) - Traditionally religious, but majorities also believe in psychics, reincarnation, and that spiritual energy can be located in physical objects.
SOMEWHAT RELIGIOUS (32%) are represented by two subgroups:
- Relaxed Religious (17%) - Say it’s not necessary to believe in God to be a moral person. Religion is important to them, but few engage in traditional practices.
- Spiritually Awake (15%) - Few practice religion in traditional ways, but most believe in heaven and hell, and subscribe to New Age beliefs.
NON-RELIGIOUS (29%) are represented by two subgroups:
- Religious Resisters(12%) -
Most think organized religion does more harm than good, politically liberal and Democratic.
- Solidly Secular (17%) - Hold virtually no religious beliefs and reject New Age beliefs.
Walking the talk
When asked about weekly habits of attending religious services and reading the Bible or other holy book, only the Sunday Stalwarts showed a high degree of connection between faith and practice (both in connection to organized religion and personal devotion).
One might expect God & Country Believers to be close behind, but they and all other groups fall off rapidly from this high degree of personal involvement.
It is interesting that the Sunday Stalwarts, at 17% in 2018 are remarkably close to Barna's Captive Christians, at 16% in 2009. This pattern is consistent with the observation of Ed Stetzer, president of Lifeway Research in 2013 when he reported on "The State of Evangelicalism":
[T]here is decline—in self-identified Protestants, primarily in mainline churches. Many who once identified themselves as nominal mainline Protestants now identify as nothing. The nominals have become the nones.
As such, we see that Christianity isn't dying—cultural Christianity is. . . (emphasis added).
For Christians who take their faith seriously, it is not time to simply sigh and be thankful that the least-faithful are falling away. It is time, as never before, for reassessment, prayer and re-commitment to the mission Christ called his followers to:
"'You must love the Lord your God with all your heart, all your soul, and all your mind.' This is the first and greatest commandment. A second is equally important: 'Love your neighbor as yourself.'"
Jesus quoting The Great Commandments, Matthew 22:37-39 NLT
"Therefore, go and make disciples of all the nations, baptizing them in the name of the Father and the Son and the Holy Spirit. Teach these new disciples to obey all the commands I have given you. And be sure of this: I am with you always, even to the end of the age."
The Great Commission, Matthew 28:19-20 NLT
Which type are you? Take the Religious Typology quiz. (This is a short version of the much longer survey questionnaire, but produces a fair guesstimate of which cohort you would be identified with.)
Return to top