For years, Gallup polls3 and other surveys have found that roughly 40 percent of Americans say they are in the pews on any given Sunday. Scholars often reacted with incredulity4, and not a few clergy have as well—unable to reconcile such figures with their own observations of more modest attendance at worship services.
Social scientists attribute inflated rates of church attendance to a phenomenon called social-desirability bias, or the tendency to make oneself look or sound better than is actually true.5 Some speculate that people think it will make them look better if they say they went to church, which is a big problem for survey-takers who are primarily interested in accurate estimates.
To gauge the magnitude of this problem, a few enterprising researchers found ways to estimate church attendance without asking about it directly. Three scholars spent a month counting attendance on weekends and estimated that church attendance rates were roughly half of what would be expected if people were taken at their word on a survey.6 They estimated that only about 20 percent of Americans attended a worship service in any given week. Other researchers employed “time use diaries” where respondents are asked to record their activities over a time period, and estimated that just over one quarter (29 percent) of Americans attend religious services in any given week.7
In the Relationships in America survey, we find that 27 percent of Americans report attending church weekly (on average). Yet this may remain an undercount of the number of people that attend worship services in any given week because it misses those who attend occasionally, but less often than once a week. According to self-reports, when occasional attenders are accounted for, we find that 35 percent of Americans attend religious services in any given week.8
Which faiths are most likely to pack the pews?
Who are the most and least likely to be at religious services? Very few of the religiously unaffiliated attend religious services—as anyone could have guessed—while Mormons report much higher attendance than any other group, at just over 80 percent weekly. Protestants are slightly more likely to say they attend services than Catholics, a recent change.9 At just 22 percent attendance in a given week, Jewish Americans are notably less likely to show up to the synagogue than their Christian peers are to be in church. Other world religious traditions also have varied attendance patterns, with Muslims more likely than all other major religious groups (with the exception of Mormons) to attend services at least weekly, while Hindus and Buddhists rate relatively low on attendance in comparison with Christian groups. Given that regular services at specific locations are less central to those two religions, this should not surprise.
There is also considerable diversity between groups under the same religious umbrella. For example, among Catholics, those who consider themselves “traditional” Catholics are the most likely to say they were in church recently: nearly 3 in 5 (58 percent) report being there in any given week. Meanwhile, liberal Catholics report church attendance at about one-third of that rate (21 percent). Similar variation can be seen among Protestants: evangelicals and Pentecostals lead all Protestant groups at 74 percent and 70 percent weekly attendance, respectively. Meanwhile, self-identified liberal Protestants have the lowest attendance numbers among Protestants, at less than half that rate—around 33 percent.
Keep in mind that some of these figures may remain inflated, since people have a tendency to report that they attend church more often than they actually do. So it’s difficult to know if evangelicals actually attend church more regularly than mainliners and if traditional Catholics really are three times as likely to be in church as their liberal Catholic peers, or on the other hand if certain groups are just more likely to overestimate their attendance. But the general trends noted here certainly remain, and who knows but that liberal Protestants or Catholics may perceive themselves as liberal in part because they don’t attend very often. That’s a plausible theory best left to be tested in another study.
Where are the men?
It’s long been noted that significant gender gaps exist in American Christian congregations, with women much more likely to attend than men.10 The Relationships in America survey finds that the gender gaps in religious affiliation for most Christian groups are small, usually statistically indistinguishable from 50/50. Overall, 64 percent of men and 68 percent of women claim an affiliation with some Christian church. But when you factor in that there are more women in the US than men (the gender ratio for the populations as a whole in 2014 was 97 males per 100 females), and that women report attending church more often than men, these small differences compound to cause skewed gender ratios.11 We estimate that there are 115 women for every 100 men (ages 18-60) at the average Christian worship service.12 The numbers are fairly similar between Protestants and Catholics at 111 and 108 women per 100 men, respectively. Mormons are closer to numerical equality between men and women with 105 women at church for every 100 men.
Is going to church a Southern thing?
Church attendance varies quite a bit by region.13 Where are you most likely to see your neighbors in church, and where are they more likely to join you for a Sunday morning on the back nine of the local golf course instead? New England and the surrounding region reports the lowest church attendance numbers, with five of the six New England states in the bottom 10, and the remaining state, Vermont, at 35th. New Hampshire exhibits the lowest attendance: about 12 percent of its respondents reported attending worship services in a given week.
In contrast, the South is called the “Bible Belt” for good reason. Additionally, the “Mountain West” also exhibits high rates of church attendance. Of the top 10 states in church attendance six are in the south, and three are in the Mountain West region. Utah, with its majority Mormon population, leads the nation in attendance—by far—at 65 percent in any given week. The second state, Arkansas, is 14 percentage points back, at 51 percent.
Are churches filled with old people?
Traditionally college-aged adults are less religious than older adults. Even 40 years ago polls showed lower religiosity and church attendance among young adults. A 1970 Gallup report on the subject expressed the sentiment:
“An accumulating mass of data suggests that organized religion is currently a significant object of commitment for only a minority of young people. Gallup polls conducted in 1970 and 1971, for example, reveal that only 28 percent of those age 21 to 29 have attended church during the previous week and that a striking 80 percent in this age category perceive religion as losing its influence in American life (Gallup Opinion Index, January 1970; February 1971).”14
Similarly, the RIAdata find that church attendance rates are lower among young adults in their 20s: 30 percent of adults ages 21 to 29 say they attend church in any given week, statistically indistinguishable from the 1970-71 Gallup poll. Church attendance rates increase moderately among older adults. If self-reports of attendance are to be believed, in any given week you can expect 29 percent of adults ages 20-25 to attend a worship service, while among the oldest adults in the survey, those 55-60, you would expect to see 37 percent in attendance that week. The data suggests an age effect, but not as profound of one as many may expect. While many popular accounts of religious behaviors suggest a lack of religious zeal among Millennials, it is not clear whether this is indicative of a secularizing trend among young people or if it is simply reflective of longstanding patterns of religiosity over the life course.
We reported earlier that those who are more educated are moderately less likely to be religiously affiliated, but the same is not true for religious service attendance. The groups with the highest church attendance are the two extremes of the education distribution—those with less than a high school education and those with a bachelor’s degree or more (at 37 percent attendance each in any given week). The figures for those with only a high school diploma or some college education are 32 percent and 33 percent, respectively.
While the most educated Americans are the most likely to be unaffiliated, they are also the most likely to attend church if they have a religious affiliation. Just over half (51 percent) of those with a bachelor’s degree who claim a religious affiliation report attendance in a given week, compared with 40 percent of affiliated people with a high school diploma. Education, then, has neither a linear increasing nor decreasing effect on religiosity, but rather a mildly polarizing effect: those with more education are modestly less likely to subscribe to a faith tradition, but those who do also say they’re more active in their faith than less-educated believers.
3 “Religion.” Gallup Historical Trends. September 16, 2014. Retrieved August 8th, 2014.
4 Brenner, Philip. “Exceptional Behavior or Exceptional Identity? Overreporting of Church Attendance in the U.S.” Public Opinion Quarterly 75 (2011): 19-41; Woodberry, Robert. “When Surveys Lie and People Tell the Truth: How Surveys Over-Sample Church Attenders.” American Sociological Review 63, no. 1 (1998): 119-122.
5 Regnerus, Mark and Uecker, Jeremy. “Religious Influences on Sensitive Self-Reported Behaviors: The Product of Social Desirability, Deceit, or Embarrassment?” Sociology of Religion 68 (2007): 145-163.
6 Hadaway, C. Kirk et al. “What the Polls Don’t Show: A Closer Look at U.S. Church Attendance.” American Sociological Review 58, no. 6 (1993).
7 Presser, Stanley and Stinson, Linda. “Data Collection Mode and Social Desirability Bias in Self-Reported Religious Attendance”. American Sociological Review 6, no. 1 (1998).
8 To calculate the percentage in church in any given week we assign each respondent a probability of attending religious services in any given week based on their self-reported religious service attendance. (i.e., those who attend about twice per month have a probability of 0.5 of attending in any given week.) We then find the mean of this probability measure to find the percentage of people attending church in a given week.
9 Gallup, George. “Catholics Trail Protestants in Church Attendance.” Gallup. December 16, 2003. Retrieved August 26, 2014.
10 “U.S. Religious Landscape Survey.” The Pew Forum on Religion and Public Life. 2007. Retrieved August 26, 2014 ; “Gender Gap in Church Persists; Worse Among Evangelicals.“ Austin Institute for the Study of Family and Culture. August 20, 2013. Retrieved August 26th 2014.
11 “Sex Ratio.” CIA World Factbook. Retrieved August 26th 2014.
12 Gender ratios in worship services were generated by assigning each person a probability of attending in a given week based on their self-report of how often they attend church. These probabilities are then summed for each gender, and then the sums are divided to calculate an expected gender ratio.
13 Figure 2.5 is organized by quintile with 10 states in each quintile. All states in the same quintile received the same coloring. Darker blues represent higher rates of church attendance.
14 Pahman, Dylan. “Are Young Millennials Less Religious or Simply Young?.” Acton Institute Power Blog. April 27, 2012. Retrieved August 26, 2014.
Questions, media inquiries, and comments should be directed to our research team. They can be reached by email at firstname.lastname@example.org. If you find this information interesting and helpful, we also invite you to visit our website, like us on Facebook, or share stories on social media.