What do 2025 surveys (Pew, Gallup, PRRI) say about U.S. church attendance and religious affiliation trends?
Executive summary
Pew (2023–24 Religious Landscape Study) finds the long decline in Christian affiliation has slowed and monthly in-person attendance is roughly flat at about one‑third of adults; 37% say they are members of a house of worship [1] [2]. Gallup reports that weekly attendance and measures of religiosity remain well below early‑2000s levels and that fewer Americans now say religion is “very important,” with attendance and perceived religiosity lower than pre‑pandemic [3] [4]. PRRI’s 2023–24 census documents continued growth in the religiously unaffiliated (around the high‑20s percent) and a long‑term decline in white Christian share, while non‑Christian minorities remain small [5] [6].
1. What Pew found: stabilization, not a rebound
Pew’s large 2023–24 Religious Landscape Study — a 36,908‑person survey — reports that the steep multi‑decade drop in Christian identity appears to have slowed and may have stabilized: about 62% identify as Christian and roughly one‑third of adults attend in‑person services at least once or twice a month; 37% report formal membership in a house of worship [2] [1]. Pew explicitly warns that changes in survey mode limit long‑term comparisons on attendance trends, but the short‑term monthly attendance figure is “pretty flat” compared with the 2020 NPORS [7] [8].
2. What Gallup reports: lower attendance, lower religiosity, and mixed short‑term trends
Gallup’s work continues to show long‑term declines in weekly attendance (e.g., only about 20% attend weekly if that threshold is used) and finds U.S. church attendance and many measures of religiosity are still below pre‑pandemic levels [9] [3]. Gallup also emphasizes that religious preferences have been “largely stable” since 2020 in some topline affiliation measures, even while questions like “is religion important in your life?” have fallen considerably — Gallup put that drop from 66% in 2015 to 49% in 2025 on record as a major decline [10] [11].
3. What PRRI documents: composition shifts and the rise of the “nones”
PRRI’s 2023 PRRI Census of American Religion and its 2024 update show continuing demographic shifts: the share of white Christians has fallen substantially since 2013 while the religiously unaffiliated reached about 28% in 2024; non‑Christian religious groups are growing but together still account for less than 10% of adults [5] [6]. PRRI also provides county‑level detail and highlights racial/age patterns — younger cohorts and many white Americans drive much of the unaffiliation trend [12] [13].
4. Numbers that matter and why thresholds change the story
Different thresholds give different impressions: Gallup notes weekly attendance (about 20% by one count) versus once‑a‑month or more (about 41% under a looser definition) produce very different “regular attendance” rates [9]. Pew’s RLS reports one‑third attending in person at least monthly and separate figures for virtual attendance [1]. This underlines that “attendance” is not a single agreed metric across surveys [9] [1].
5. Age, race and denominational nuances
All three organizations find important subgroup variation. Younger adults are less likely to identify as Christian and more likely to be unaffiliated (Gallup and PRRI), yet Pew notes some stability in cohort shares since 2020 and PRRI shows changing racial composition within Catholic and Protestant populations [4] [5] [2]. Retention and growth patterns differ by tradition: PRRI and Gallup point to higher retention among some conservative and Black Protestant groups and losses among white mainline Protestants and some Catholic segments [14] [13].
6. Pandemic, polling mode, and why experts caution about comparisons
Pew explicitly warns about a “mode switch” (telephone to online/paper) affecting long‑term attendance trends, making direct historical comparisons hazardous [7]. Gallup and others flag that pandemic disruptions and measurement changes complicate attributing declines to single causes; Gallup shows attendance remains below pre‑pandemic levels but does not conclude the pandemic is the sole driver [3] [10].
7. Competing narratives and political context
Media and interest groups draw different narratives: some outlets emphasize stabilization or a possible “pause” in Christian decline citing Pew and Gallup [8] [15], while others highlight continued large declines in religiosity and importance measures [11] [16]. PRRI’s political breakdowns also show unaffiliated Americans cluster heavily among Democrats — a fact often cited when interpreting religion’s role in politics [13]. Each source has methodological choices and audiences that shape framing: Pew focuses on a large academic study, Gallup uses repeated annual tracking, and PRRI emphasizes county‑level and demographic detail [2] [10] [12].
8. Bottom line for readers
Available sources show the broad picture: Christian affiliation’s long slide has slowed or plateaued in recent surveys, monthly attendance measures are roughly stable at around one‑third of adults per Pew, weekly attendance and measures of religion’s personal importance remain much lower than decades ago per Gallup, and PRRI documents continuing compositional change with rising unaffiliated shares [1] [3] [5]. Differences in question wording, attendance thresholds, and survey modes explain much of the apparent disagreement — treat single headline percentages as entry points, not final judgments [7] [9].
Limitations: this summary uses the provided reports and articles; available sources do not mention direct membership or attendance counts beyond the cited percentages and methodological notes.