Barna Group found that 26% of self-identified Christians report ongoing
unresolved spiritual doubt — not past doubt they worked through, but current,
persistent questioning — the dominant inaction-side rate. On the action side,
PRRI’s 2024 report on religious change in America found that only 9% of
religiously unaffiliated US adults say the statement “I am looking for a
religion that would be right for me” describes them well, making active seeking
the closest available attitudinal proxy for action-side regret. Pew Research
Center’s December 2025 survey of 8,937 US adults adds context: among
those who kept their childhood religion, only 64% rated belief in its
teachings as extremely or very important, leaving a substantial minority
whose reasons for staying center on community, habit, or family rather
than conviction.
The 17-point gap favoring inaction regret aligns with Gilovich and
Medvec’s temporal framework, in which regrets of inaction tend to
compound over time. Identity decisions like religious affiliation fit
this pattern: leaving a religion is a concrete, time-stamped action
whose consequences become familiar, while staying despite doubt is an
ongoing omission that can accumulate as a “what if I had been braver”
narrative. Bleidorn et al.’s 2024 longitudinal study of roughly 20,000
Dutch adults found no significant change in life satisfaction after
deconversion, suggesting that the action itself is neither reliably
beneficial nor harmful to well-being. The size of the gap here likely
overstates true regret asymmetry because the two proxies measure
different intensities of dissatisfaction: actively seeking a new
religion is a higher bar than reporting unresolved doubt.
Both proxies stretch the meaning of “regret rate.” Seeking a religion
is not the same as regretting that one left; some seekers may want a
novel spiritual practice rather than a return to what they abandoned.
Ongoing doubt is not regret; many theologians treat doubt as intrinsic
to mature faith rather than a sign that staying was the wrong call.
Denomination-level variation is likely large, with high-demand groups
(LDS, Jehovah’s Witnesses) producing higher social costs for leaving
and thus potentially more seekers whose motivation is community re-entry
rather than spiritual conviction. The PRRI and Barna surveys differ in
date (2023 vs 2017) and population framing. All sources are US-centric;
in societies where religion is legally or socially compulsory, the
inaction/action frame barely applies.
Sources: action
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
[1]PRRI (Public Religion Research Institute) — Religious Change in America
Primary study
Only 9% of religiously unaffiliated Americans say 'I am looking for a religion that would be right for me' describes them very or somewhat well
Excerpt
“"Only 9% of religiously unaffiliated Americans say the statement 'I am looking for a religion that would be right for me' currently describes them very or somewhat well."
”
Source data from
2024-03-27
Accessed
2026-04-27
Calculation
PRRI surveyed US adults in 2023 as part of its American Values Atlas. Among the religiously unaffiliated — the majority of whom left a childhood religion — only 9% report actively seeking a new religious home. We treat this as an attitudinal proxy for action-side regret: wanting to return to religious affiliation implies dissatisfaction with the decision to leave. The 91% who are not looking suggest most leavers are content with disaffiliation. This is a proxy, not a direct regret measure: some of the 9% may be curious rather than regretful, and some who are not looking may still harbor private regret. regret_rate = 0.09.
[2]Journal of Personality (Wiley) — Psychological change before and after religious conversion and deconversion
Peer-reviewed
Neither converts nor deconverts showed significant changes in life satisfaction
Excerpt
“"We assessed changes in personality traits and subjective well-being before and after religious conversion and deconversion. Neither converts nor deconverts showed significant changes in self-esteem, life satisfaction, or depressive symptoms."
”
Source data from
2024-02-01
Accessed
2026-04-27
Calculation
Bleidorn et al. (2024) used 11-wave longitudinal data from a nationally representative Dutch sample (N ~ 20,000) to assess changes before and after conversion/deconversion. The null finding on well-being supports the low action- regret rate: leaving religion does not systematically worsen life satisfaction, consistent with few leavers seeking to return. This source is Dutch, not American, limiting direct comparability but providing the best available longitudinal well-being data on deconversion.
Sources: inaction
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
[1]Barna Group — Two-Thirds of Christians Face Doubt
Primary study
26% of self-identified Christians still experience ongoing spiritual doubt
Excerpt
“"Just over one-quarter (26%) say they still experience spiritual doubt, while four in 10 (40%) say they have experienced it in the past but have worked through it."
”
Source data from
2017-07-25
Accessed
2026-04-27
Calculation
Barna surveyed US adults who self-identify as Christian (or have in the past). Of those, 65% have questioned their beliefs at some point; 26% report ongoing unresolved doubt. We treat persistent unresolved doubt among those who stayed as an attitudinal proxy for inaction regret: it captures the fraction who remain affiliated but experience sustained questioning about the faith they chose to keep. This is a proxy, not a direct regret measure: many doubters view doubt as spiritually productive rather than as evidence they should have left. regret_rate = 0.26.
[2]Pew Research Center — Why Do Some Americans Leave Their Religion While Others Stay?
Primary study
Among adults who kept their childhood faith, 64% cite belief in teachings as an extremely or very important reason for staying
Excerpt
“"Most U.S. adults who still identify with their childhood religion credit the following as extremely or very important reasons: They believe the religion's teachings (64%), their religion fulfills their spiritual needs (61%), and their religion gives their life meaning (56%)."
”
Source data from
2025-12-15
Accessed
2026-04-27
Calculation
Pew surveyed 8,937 US adults (May 5-11, 2025). Among those who stayed in their childhood religion, 36% did not rate belief in teachings as extremely or very important. This complement provides context for the Barna doubt figure: a substantial minority stays despite tepid conviction. Used as contextual support for inaction_side, not as the primary regret_rate source.
Caveats
Neither side has a direct regret survey; both rates are attitudinal proxies. The action-side proxy (9% seeking a new religion) conflates curiosity with regret — some seekers may want a different religion, not a return to the one they left. It also covers all unaffiliated adults, not only those who disaffiliated (some were raised without religion). The inaction-side proxy (26% ongoing doubt) conflates cognitive questioning with decisional regret; many theologians treat doubt as intrinsic to mature faith, not a sign that staying was the wrong call. Both proxies use the same construct type (attitudinal self-report about current religious satisfaction), making them more comparable than the previous version of this entry, which mixed a behavioral measure (reconversion) with an attitudinal one (doubt). The PRRI and Barna surveys differ in date (2023 vs 2017) and population framing (all unaffiliated vs self-identified Christians). The Bleidorn et al. longitudinal study is Dutch, not American, limiting direct comparability. Denomination matters enormously: leaving a mainline Protestant church carries different social costs than leaving a high-demand group (LDS, Jehovah's Witnesses, ultra- Orthodox), which affects both action regret and willingness to seek a new faith. US-centric data throughout; the pattern likely differs in societies where religion is more or less socially compulsory.
