Religious Attendance in 2025: Lower Participation and the Duty to Serve

Editorial correction — September 20, 2026. We correct the share of adults 65 and older who seldom or never attend from 24% to 51%, retain the weekly-or-nearly-weekly definition, and remove unsupported claims of uninterrupted decline and motives assigned to unaffiliated adults. Expanded September 20, 2026 with additional primary-source data and a Sources list.

Fewer than a third of American adults attend religious services weekly or nearly weekly, and a majority say they seldom or never go. Those two facts are well documented. What they mean is a different question, and the two best-known surveys do not measure it identically. This article lays out what Gallup and the Pew Research Center found, where they agree, where they differ, and what neither can tell us.

What Gallup found for 2025

Gallup’s March 3, 2026 analysis, written by Megan Brenan, reports annual averages from surveys conducted in 2025. In that year 31% of U.S. adults said they attended religious services weekly or nearly weekly, about 10% said they attended roughly once a month, and 57% said they seldom or never attend. Attendance about once a month has stayed relatively steady across Gallup’s trend line, so the movement has come from the other two categories.

The long view matters here. Gallup reports that from the early 1990s to 2008, majorities of U.S. adults said they attended services at least monthly. Since 2018, majorities have said they rarely or never attend. That is a change in the answer most Americans give, and it unfolded over roughly a generation.

Three different measures, not one

Religious life is measured in several ways, and they do not always move together. Gallup tracks three in this release.

Gallup reports that the importance measure has leveled off below 50% in recent years. The “Nones” figure rests on interviews with more than 13,000 adults across Gallup’s monthly 2025 surveys.

The age gap

Gallup’s age breakdown is the most useful part of the release, because it separates the trend from the average.

Gallup also reports no-religion shares of 29% for ages 30 to 49 and 18% for ages 50 to 64.

Two details deserve attention. The oldest adults are not uniformly devout: half of them, 51%, say they seldom or never attend. And the youngest adults are not uniformly absent: one in four attends weekly or nearly weekly. An age average that hides a split in each group is a poor guide to any individual congregation or family.

Who remains highly religious

Gallup compared 2001 to 2005 with 2021 to 2025 and found that most demographic groups have become less likely to say religion is very important. The exceptions were Catholic and Jewish Americans and Republicans. In the recent period, six groups still had between 55% and 67% saying religion is very important: Latter-day Saints, Republicans, Protestant or nondenominational Christians, Black adults, adults 65 and older, and Southerners.

The largest declines over the two periods were among Democrats, down 23 points to 37%, and Black adults, down 22 points to 63%. Jewish adults were the only group more likely to say religion is very important, up 10 points to 32%. The groups least likely to call religion very important were Jewish adults (32%) and adults 18 to 29 (33%). These are shifts within groups over time. They do not describe every member of any group.

Generations, not just changed minds

Gallup attributes much of the drop in affiliation to generational change: younger adults, who are far more likely to have no religion, are gradually replacing older generations that had fewer unaffiliated people. Pew reaches a compatible conclusion from its own data, and adds that between 2007 and 2023-24 each birth cohort became somewhat less religious as it aged, by several of its measures.

What Pew adds

Pew’s 2023-24 Religious Landscape Study surveyed 36,908 adults and was published February 26, 2025. It found that 62% of U.S. adults identify as Christian, down 9 points from 2014 and 16 points from 2007, and that 29% have no religion, up from 23% in 2014 and 16% in 2007. Its headline finding was one of stabilization: the Christian share has hovered between 60% and 64% since 2019, and 33% of adults say they attend at least monthly, close to what Pew measured in 2020 and 2024.

The same study shows why stability would not be the whole story. Adults 18 to 24 were less likely than those 74 and older to identify as Christian (46% versus 80%), to pray daily (27% versus 58%) and to attend at least monthly (25% versus 49%). Congregation membership was 27% among the youngest adults and 57% among the oldest. Pew’s reasoning is that older, more religious generations will shrink as their members die, so lasting stability would require something to change, such as today’s young adults becoming more religious as they age.

One more finding cuts against a simple story of handing down belief. Pew reports that 40% of adults attend services less often than they did as children, and 5% attend more often. Among young adults raised in highly religious homes, 28% are highly religious themselves.

Why the two surveys do not match exactly

Gallup puts “Nones” at 24% and Pew at 29%. Adding Gallup’s weekly-or-nearly-weekly and monthly groups gives about 41% attending at least monthly, compared with Pew’s 33%. This article does not try to reconcile those gaps. Different question wording, sampling and survey mode commonly produce such differences, and Pew notes directly that its older telephone surveys and its newer online and paper surveys give significantly different answers to attendance questions. That is why Pew says it lacks a clean long-term attendance trend across its own studies. The safe rule is to compare Gallup with Gallup and Pew with Pew, and to treat a difference between them as a difference in method until shown otherwise.

What these numbers cannot tell us

Attendance is self-reported. It counts how people describe their own habits, not how many people are inside a building. Non-attendance does not measure belief, and it says nothing about motives. Nor do these surveys measure moral conduct, charity or service, so they cannot show that a more or less religious country is a better or worse one. Anyone who says they do is asking one statistic to carry an argument it was never built for.

What they do establish is narrower and still important. Regular attendance is a minority practice, it is lower among younger adults, and the trend across recent decades has been downward, though both surveys show a recent flattening in some measures. Institutions that once relied on weekly gatherings to bring people together, feed the hungry and visit the sick may have a smaller regular base to draw on. That is an inference, not a survey finding, but it is a fair question for any community that values those functions: who will do that work?

How to read the next release

• Note which measure is being reported: attendance, importance, affiliation or belief.
• Keep the category exactly as written: “weekly or nearly weekly” is not “weekly”.
• Look at the age and group breakdowns before quoting a national average.
• Compare a survey with its own earlier results, not with a different survey.
• Separate what the data measure from what a writer infers about motives or morality.

Sources

1. Gallup, Megan Brenan, “Americans’ Religious Engagement Holds at Lower Levels,” March 3, 2026: news.gallup.com/poll/702572
2. Pew Research Center, “Decline of Christianity in the U.S. Has Slowed, May Have Leveled Off: Findings from the 2023-24 Religious Landscape Study,” February 26, 2025: pewresearch.org, Religious Landscape Study overview

Editorial standards · Report an error · Subscribe