Skip to content
Back to the question

Evidence record

Is Christianity growing or shrinking?

Every sentence on the exploration belongs to at least one claim below. Each says what kind of statement it is, what it rests on, what the evidence does not establish, and how far it was checked. Nothing here is a score, and none of it is a verdict on a belief or a theory.

Last checked 23 September 2026.

  • Source checked: 33
  • A judgement, not a fact: 19
  • Presented as open: 1

If something here is wrong, misquoted, out of date or unfair, write to support@unknownoceans.com and give the claim number, which does not change.

What the question takes for granted

A demographic question with a measured answer, as long as "growing" is pinned to one measure. The count and the share point in different directions, and both are estimates of self-identification built from censuses and surveys.

  • Christians can be counted as one group

    Contested

    Pew counts every denomination as one group and counts what people say they are. Churches count baptisms, members or attenders, and mean different things by each; the page says so in its counting passage.

  • The counts are good enough to answer it

    Established

    For the decade 2010 to 2020 the estimates rest on more than 2,700 censuses and surveys. They are estimates, some delayed by the pandemic, and Pew revised its own 2010 baseline in 2025, which is the size of error a reader should allow for.

  • “Growing” means one thing

    Contested

    A count, a share of the world and a rate of growth against other groups give three different answers, which is the page’s central point.

  • Present trends say what comes next

    Open

    The projections assume current fertility, mortality, migration and switching continue, start from a baseline since revised, and leave switching unmodelled in China and India.

How the question is labelled

Claim: The demographic turn

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Names the shift the page reports from Pew’s figures: decline in Europe and North America, growth elsewhere, and sub-Saharan Africa now the largest regional Christian population.

Claim 61 · checked 23 September 2026 · AI-assisted audit

Claim: The numbers

Kind
Framing. Signposting. It asserts nothing a source could check.

The question

Claim: Both, in different places, and the centre of it has moved.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    How the global religious landscape changed from 2010 to 2020, chapter 2: Christian population changeRegional change; Regional distribution of Christians

    Cited · Link answered, 19 October 2026 · Read in full, 23 September 2026

    The source reads: In Europe, Christians declined to 505 million (down 9%). In North America, they shrank to 238 million (down 11%). In every other region, the number of Christians grew. ... sub-Saharan Africa surpassed Europe as the region with the largest Christian population.

Claim 1 · checked 23 September 2026 · AI-assisted audit

Claim: The number of Christians is rising and the share of the world that is Christian is falling.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    How the global religious landscape changed from 2010 to 2020Overview, opening findings

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: The number of Christians rose by 122 million, reaching 2.3 billion. Yet, as a share of the world’s population, Christians fell 1.8 percentage points, to 28.8%.

What this does not establish

Measured for 2010 to 2020. Anything after 2020 is a projection.

Claim 2 · checked 23 September 2026 · AI-assisted audit

Claim: Where they live is changing fast

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

Claim 3 · checked 23 September 2026 · AI-assisted audit

Claim: and much of what else is claimed about it depends on which country the claimant is looking from.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Follows from the regional figures: the count fell in Europe and North America and rose everywhere else, so an observer’s own country is a poor guide to the world.

Corrected

23 September 2026. Previously: “and almost everything else claimed about this is an artefact of which country the claimant lives in.” Too sweeping. The fall in the Christian share is global, 1.8 points in a decade, and is not an artefact of anyone’s vantage point. What does depend on the observer’s country is whether the count looks like it is rising or falling, which is what the corrected clause says.

Claim 4 · checked 23 September 2026 · AI-assisted audit

Claim: Ask this in London or Boston and the answer looks obvious: emptying churches, a collapsing share of the young.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Describes how the question looks from two Western cities, not a measurement of either; the UK lost its Christian majority in the decade and the US age gap in affiliation is 17 points.

  • Context

    How the global religious landscape changed from 2010 to 2020Most countries still have Christian majorities

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: Christians dropped below 50% of the population in the United Kingdom (49%), Australia (47%), France (46%) and Uruguay (44%).

  • Context

    The age gap in religion around the worldOverview

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: the share of U.S. adults under age 40 who identify with a religious group is 17 percentage points lower than the share of older adults who are religiously affiliated.

What this does not establish

“Collapsing” is the impression, and the page itself later says the age gap is narrower than it is usually made to sound.

Claim 5 · checked 23 September 2026 · AI-assisted audit

Claim: Ask it in Lagos or Kinshasa and it looks equally obvious the other way.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

What this does not establish

The figures read are regional. No country figure for Nigeria or the Democratic Republic of the Congo was read, and Pew notes Nigeria is split between a Muslim north and a Christian south.

Claim 6 · checked 23 September 2026 · AI-assisted audit

Claim: Both observations are accurate. What makes the question answerable is that somebody counts.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The count fell in Europe and North America and rose in sub-Saharan Africa, so both local impressions are right about their own region.

  • Context

    How the global religious landscape changed from 2010 to 2020How we did this

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: This analysis is based on more than 2,700 sources of data, including national censuses, large-scale demographic surveys, general population surveys and population registers.

Claim 7 · checked 23 September 2026 · AI-assisted audit

What the counts show

Claim: What the counts show

Kind
Framing. Signposting. It asserts nothing a source could check.

The absolute number is going up

Claim: The absolute number is going up

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Measured to 2020; the present tense rests on the projections after that.

Claim 9 · checked 23 September 2026 · AI-assisted audit

Claim: Between 2010 and 2020 the number of Christians rose by 122 million, to 2.3 billion.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

Claim 10 · checked 23 September 2026 · AI-assisted audit

Claim: That is 6 per cent, against 15 per cent for everyone else, so the Christian share of the world fell 1.8 points, to 28.8 per cent.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

Claim 11 · checked 23 September 2026 · AI-assisted audit

Claim: Pew names disaffiliation as the main driver of that fall: people raised Christian who now say they have no religion.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Pew also notes that some who leave Christianity join another religion rather than none, and that fertility differences between regions play a part.

Corrected

23 September 2026. Previously: “Pew puts that fall down to disaffiliation: people raised Christian who now say they have no religion.” Pew calls disaffiliation “the main driver” of the fall in share, not its whole cause; natural increase differs between regions too. The sentence now says what Pew says.

Claim 12 · checked 23 September 2026 · AI-assisted audit

Claim: The projections, which still run forward from a 2010 baseline, have Christians growing by about 34 per cent between 2015 and 2060 and Muslims by about 70, which brings the two close to parity: roughly 3.1 billion Christians and 3.0 billion Muslims, about 32 and 31 per cent of the world.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The changing global religious landscapeGlobal population projections, 2015 to 2060

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: is projected to increase by 70%. The number of Christians is projected to rise by 34% ... by 2060, the count of Muslims (3.0 billion, or 31% of the population) will near the Christian count (3.1 billion, or 32%).

  • Supports

    The changing global religious landscapeNote (June 2025) at the head of the report

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: Religious composition estimates in this report are based on 2010 baseline numbers that have since been updated.

What this does not establish

A projection made in 2017 from a baseline Pew has since revised. The 2015 starting figures in it are themselves estimates, and the revised 2025 report puts the Christian share at 28.8 per cent in 2020, below the path these numbers assume.

Claim 13 · checked 23 September 2026 · AI-assisted audit

Claim: Christianity is not shrinking in absolute numbers.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

True of the world. The count did fall in Europe and North America.

Claim 14 · checked 23 September 2026 · AI-assisted audit

Claim: It is losing share, and it is growing more slowly than the fastest-growing alternative.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    How the global religious landscape changed from 2010 to 2020Overview, opening findings

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: Muslims were the fastest-growing religious group over the decade. The number of Muslims increased by 347 million – more than all other religions combined.

Claim 15 · checked 23 September 2026 · AI-assisted audit

Claim: Those are three different statements.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Count, share and relative rate, each checked in the claims just before.

Corrected

23 September 2026. Previously: “Those are three different statements and the last is usually the one being made.” The second clause is a claim about what people usually mean when they say Christianity is shrinking. Nothing cited, or found, measures that.

Claim 16 · checked 23 September 2026 · AI-assisted audit

The centre of gravity has moved south

Claim: The centre of gravity has moved south

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

Claim 17 · checked 23 September 2026 · AI-assisted audit

Claim: It has already happened once.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Points at the measured change that follows, as distinct from the projection at the end of the paragraph.

Claim 18 · checked 23 September 2026 · AI-assisted audit

Claim: Sub-Saharan Africa passed Europe during the 2010s as the region holding the most Christians: 31 per cent of them by 2020, up six points in a decade, against 22 per cent in Europe.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    How the global religious landscape changed from 2010 to 2020, chapter 2: Christian population changeRegional distribution of Christians

    Cited · Link answered, 19 October 2026 · Read in full, 23 September 2026

    The source reads: Between 2010 and 2020, sub-Saharan Africa surpassed Europe as the region with the largest Christian population. As of 2020, sub-Saharan Africa was home to 31% of all the Christians in the world (up 6 points since 2010), while Europe held 22% of the global Christian population (down 4 points).

Claim 19 · checked 23 September 2026 · AI-assisted audit

Claim: The European count fell 9 per cent over those ten years, to 505 million, and the North American count 11 per cent, to 238 million.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

Claim 20 · checked 23 September 2026 · AI-assisted audit

Claim: The projections take the African share to 42 per cent by 2060.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The changing global religious landscapeGlobal population projections, 2015 to 2060

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: the share of Christians worldwide who live in sub-Saharan Africa is expected to increase dramatically between 2015 and 2060, from 26% to 42%, due to high fertility in the region.

What this does not establish

A projection from the 2010 baseline Pew has since revised; the measured share was already 31 per cent by 2020.

Claim 21 · checked 23 September 2026 · AI-assisted audit

Claim: In 1910 about two thirds of the world’s Christians lived in Europe.

Kind
Historical claim. A statement about the past that historical evidence can test.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Global Christianity: a report on the size and distribution of the world’s Christian populationExecutive summary

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: In 1910, about two-thirds of the world’s Christians lived in Europe, where the bulk of Christians had been for a millennium, according to historical estimates by the Center for the Study of Global Christianity.

  • Supports

    How the global religious landscape changed from 2010 to 2020Geographic distribution of religious groups in 2020

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: This is a major geographic change since the early 1900s, when Christians in sub-Saharan Africa made up 1% of the global Christian population and two-thirds of Christians lived in Europe.

What this does not establish

A historical estimate by the Center for the Study of Global Christianity, not a census count.

Claim 22 · checked 23 September 2026 · AI-assisted audit

Claim: The institutions, the arguments and the music follow the people.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s own judgement about what a shift in numbers carries with it. None of the reports read measures institutions, arguments or music.

Claim 23 · checked 23 September 2026 · AI-assisted audit

Where the decline is real, it is steep

Claim: Where the decline is real, it is steep

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

Claim 24 · checked 23 September 2026 · AI-assisted audit

Claim: In 41 countries the Christian share of the population moved by at least five points between 2010 and 2020, and in all but Mozambique it moved down: Australia by twenty points, Chile by eighteen, Uruguay by sixteen, the United States and Canada by fourteen each.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Pew sets the five-point threshold because differences in data sources make formal significance tests hard.

Claim 25 · checked 23 September 2026 · AI-assisted audit

Claim: Pew attributes most of that to disaffiliation, adults raised Christian who now report no religion.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

Claim 26 · checked 23 September 2026 · AI-assisted audit

Claim: The age gap runs the same way

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The age gap in religion around the worldOverview

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: In the places where there is a difference, however, it is almost always in the direction of younger adults being less religious than their elders.

What this does not establish

The age-gap figures cover people of every religion in Pew surveys from the decade before 2018, not Christians alone and not the 2010 to 2020 decade exactly.

Claim 27 · checked 23 September 2026 · AI-assisted audit

Claim: though it is narrower than it is usually made to sound

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Follows from the counts that follow it: in 63 of 106 countries Pew found no significant difference in affiliation by age.

  • Supports

    The age gap in religion around the worldOverview

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: In many countries, the gaps are relatively small. Indeed, the average gap between younger adults and older adults across all the countries surveyed is 5 percentage points for affiliation

What this does not establish

How it is “usually made to sound” is the page’s impression, not a measurement.

Claim 28 · checked 23 September 2026 · AI-assisted audit

Claim: adults under forty are less likely than their elders to claim any affiliation in 41 countries of 106 and more likely in two, and less likely to pray daily in 71 of 105.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The age gap in religion around the worldOverview, “Same pattern seen over multiple measures of religious commitment”

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: In 41 countries, adults under 40 are significantly less likely than their elders to have a religious affiliation, while in only two countries (Chad and Ghana) are younger adults more likely ... less likely to say they pray daily in 71 of 105 countries

Claim 29 · checked 23 September 2026 · AI-assisted audit

Claim: The falls are scattered across every region, but many are in Europe and in other Western or English-speaking countries, where Christian majorities have been shrinking for decades.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

Corrected

23 September 2026. Previously: “This is not a global pattern being reported from a global vantage point. It is a regional pattern being reported by the regions that own most of the world’s newspapers.” Two problems. Nothing cited or found says which regions own most of the world’s newspapers, so the second sentence was an unsourced factual claim. And Pew finds the falls scattered through every region, including Benin, Namibia and Cape Verde, with the global share down too, so “not a global pattern” overstated it. The corrected wording is Pew’s own description of where the falls are, and the point about vantage is kept as the page’s reading.

24 September 2026. Previously: “The falls are scattered across every region, but most are in Europe and in other Western or English-speaking countries, where Christian majorities have been shrinking for decades.” Pew says many of the falls are in Europe and other Western or English-speaking places, not most. Its other count, most in the Americas and Europe, includes Latin America and so does not carry "most" for the Western and English-speaking group either.

Claim 30 · checked 23 September 2026 · AI-assisted audit

Claim: Seen from there, a regional pattern is easy to mistake for a global one.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s reading of the sentence before it; the correction is recorded on claim 30.

Claim 31 · checked 23 September 2026 · AI-assisted audit

What the counts are

Claim: What the counts are

Kind
Framing. Signposting. It asserts nothing a source could check.

What the projections can and cannot do

Claim: What the projections can and cannot do

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: A projection of this kind runs current fertility, mortality, migration and rates of switching forward.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The changing global religious landscapeAppendix B: Methodology

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: Pew Research Center has undertaken the first formal demographic projections using data on age, fertility, mortality, migration and religious switching for multiple religious groups around the world.

Claim 34 · checked 23 September 2026 · AI-assisted audit

Claim: It is arithmetic about trends, not prophecy.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Follows from the cohort-component method Pew describes.

  • Context

    The changing global religious landscapeAppendix B: Methodology

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: Each cohort is projected into the future by adding likely gains (immigrants and people switching in) and by subtracting likely losses (deaths, emigrants and people switching out) year by year.

Claim 35 · checked 23 September 2026 · AI-assisted audit

Claim: The input it handles worst is the one that moves fastest.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Switching is the input Pew itself calls hardest to model, and the one driving the measured decade’s change in the Christian share.

  • Context

    The changing global religious landscapeOverview

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: The projections attempt to take religious switching (in all directions) into account, but conversion patterns are complex and varied.

Claim 36 · checked 23 September 2026 · AI-assisted audit

Claim: Switching was modelled for 70 countries, holding 43 per cent of the world’s people, and the two most populous left out are China and India.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The changing global religious landscapeEndnote on the 70 countries

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: These 70 countries were home to 43% of the world’s population in 2010 ... The most populous countries for which switching was not modeled are China and India.

Claim 37 · checked 23 September 2026 · AI-assisted audit

Claim: Pew says what follows: if China moves toward affiliation, the global figure tilts with it, because most of the world’s unaffiliated live there.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The changing global religious landscapeEndnote on the 70 countries

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: If China experiences a net movement toward religious affiliation via switching in the decades ahead, that would tilt the needle toward a more religiously affiliated global population, particularly since China is currently home to a majority of the worldwide unaffiliated population.

Claim 38 · checked 23 September 2026 · AI-assisted audit

Claim: The 2060 run also still starts from a 2010 baseline that Pew has since revised, and nothing has yet replaced it.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    The changing global religious landscapeNote (June 2025) at the head of the report

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: Religious composition estimates in this report are based on 2010 baseline numbers that have since been updated. Please refer to our 2025 report “How the Global Religious Landscape Changed from 2010 to 2020” for the latest 2010 and 2020 religious composition estimates. New projections past 2020 are not yet available.

What this does not establish

True as of the report page read on 23 September 2026.

Claim 39 · checked 23 September 2026 · AI-assisted audit

What counting a Christian means

Claim: What counting a Christian means

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: The figures above count self-identification.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Pew’s Israel figure for Jews uses a population register, so the rule has an exception, though not for Christians.

Claim 41 · checked 23 September 2026 · AI-assisted audit

Claim: A census answer, a baptismal register and a Sunday attendance count produce three different populations.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Being Christian in Western EuropeOverview

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: Although the vast majority of adults say they were baptized, today many do not describe themselves as Christians. ... non-practicing Christians (defined, for the purposes of this report, as people who identify as Christians, but attend church services no more than a few times per year)

What this does not establish

Shown here for Western Europe, from survey answers about baptism rather than church registers themselves.

Claim 42 · checked 23 September 2026 · AI-assisted audit

Claim: In the United Kingdom, Pew found 55 per cent of adults calling themselves Christian while going to church a few times a year at most, against 18 per cent going monthly or more.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Being Christian in Western EuropeOverview

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: church-attending Christians (those who go to religious services at least once a month). In the United Kingdom, for example, there are roughly three times as many non-practicing Christians (55%) as there are church-attending Christians (18%) defined this way.

What this does not establish

A 2017 survey of adults.

Claim 43 · checked 23 September 2026 · AI-assisted audit

Claim: Traditions differ on which of those is the real number, and they are not being disingenuous about it.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Follows from the next sentence: churches that differ on who is baptised and who is a member will differ on which count describes them.

What this does not establish

No church document read here ranks census, register and attendance against each other. This is the page’s reading.

Claim 44 · checked 23 September 2026 · AI-assisted audit

Claim: A church that baptises infants and a church that baptises believers on their own profession of faith are measuring different things because they mean different things.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

One church of each kind was read: the Catholic Church baptises infants shortly after birth, and the Southern Baptist confession baptises believers and makes that baptism the door to membership.

  • Supports

    Catechism of the Catholic Church, paragraphs 1246 to 1255: Who can receive Baptism?Catechism of the Catholic Church 1250, 1252

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: the Church and the parents would deny a child the priceless grace of becoming a child of God were they not to confer Baptism shortly after birth. ... practice of infant Baptism is an immemorial tradition of the Church.

  • Supports

    The Baptist Faith and Message 2000Articles VI and VII

    Cited · Link answered, 19 October 2026 · Read in full, 23 September 2026

    The source reads: A New Testament church of the Lord Jesus Christ is an autonomous local congregation of baptized believers ... Christian baptism is the immersion of a believer in water ... Being a church ordinance, it is prerequisite to the privileges of church membership

What this does not establish

Two documents stand for two families. Orthodox, Lutheran, Reformed, Anglican and Methodist churches also baptise infants, and Anabaptist and many Pentecostal churches baptise believers; their documents were not read here, and each counts members in its own way.

Corrected

23 September 2026. Previously: “A church that baptises infants and a church that counts adult professions of faith are measuring different things because they mean different things.” Churches that practise believer’s baptism baptise believers, not adults as such: the Baptist confession speaks of “the immersion of a believer”, and children who profess faith are baptised in such churches. “Adult” was the wrong line to draw.

Claim 45 · checked 23 September 2026 · AI-assisted audit

Whether the pattern generalises

Claim: Whether the pattern generalises

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: The decline is real, and most of the countries where it shows are in Europe and the Americas.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    How the global religious landscape changed from 2010 to 2020Changes within countries

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: Christians experienced a substantial change, as defined above, in more countries (41) than any other religious group. In all but one case, Christians shrank as a share of the population. Most of the countries experiencing declines were in the Americas and Europe.

Corrected

23 September 2026. Previously: “The decline everybody reports is happening in the places that own the newspapers.” Newspaper ownership by region was asserted with no source, and the global Christian share fell too, so the decline is not only local. The corrected line says where Pew finds most of the falls.

24 September 2026. Previously: “The decline is real, and most of it is in Europe and the Americas.” Pew counts countries: most of the 40 countries where the Christian share fell by five points or more were in the Americas and Europe. "Most of it" read as most of the decline itself, which the source does not measure.

Claim 47 · checked 23 September 2026 · AI-assisted audit

Whether secularisation is one thing

Claim: Whether secularisation is one thing

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: A familiar theory holds that modernity dissolves religion

Kind
What scholars argue. A report of what specialists argue, or where they disagree.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    How the global religious landscape changed from 2010 to 2020Note to “Economic development and religious affiliation”

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: A popular theory is that rising economic and existential security leads to religious decline. Studies have tested and developed this theory in many contexts.

  • Supports

    The age gap in religion around the worldChapter 1

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: One common explanation is that new generations become less religious in tandem with economic development

What this does not establish

Pew describes the theory and its tests without endorsing it. No history of the secularisation thesis was read, so the page no longer dates it.

Corrected

23 September 2026. Previously: “The long-standing expectation was that modernity dissolves religion” The sources cited describe it as “a popular theory” and “one common explanation”, not as an expectation or as long-standing. No source on the history of the idea was cited or read.

Claim 49 · checked 23 September 2026 · AI-assisted audit

Claim: and the last century ran that experiment in several places at once with results that do not agree.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

Follows from the Western European, American and African contrasts in the next sentence.

Claim 50 · checked 23 September 2026 · AI-assisted audit

Claim: Western Europe secularised, the United States did not on the same schedule

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    Being Christian in Western EuropeOverview, “Western Europeans are less religious than Americans”

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: Yet Americans, overall, are considerably more religious than Western Europeans. Half of Americans (53%) say religion is “very important” in their lives, compared with a median of just 11% of adults across Western Europe.

  • Supports

    The age gap in religion around the worldChapter 1

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: On this measure, the U.S. (where 55% of adults pray daily) is a major outlier; of 102 countries studied, it is the only one with higher-than-average levels of both prayer and wealth.

What this does not establish

The US is now losing Christian share quickly, 14 points between 2010 and 2020, so “not on the same schedule” is the right strength: later, not never.

Claim 51 · checked 23 September 2026 · AI-assisted audit

Claim: and the one region where Christianity is still gaining share, narrowly, is the one with the highest fertility.

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

What this does not establish

Two facts about one region, not a cause: the Muslim share in the same region rose by about a point, more than the Christian.

Corrected

23 September 2026. Previously: “and the one region where Christianity is still gaining share is the one with the highest fertility.” The gain is under half a point in a decade, which “gaining share” alone lets a reader picture as large. The fertility claim needed chapter 14, now cited.

Claim 52 · checked 23 September 2026 · AI-assisted audit

Claim: Pew finds affiliation lower where the Human Development Index is higher

Kind
Present-day figure. A count, survey or report finding, as the named source gives it.
Review
Source checked. The cited passage was read, and it says this.

  • Supports

    How the global religious landscape changed from 2010 to 2020Economic development and religious affiliation

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: We find that countries with high HDI scores (i.e., those that are more developed) tend to have lower rates of religious affiliation.

Claim 53 · checked 23 September 2026 · AI-assisted audit

Claim: which is a correlation across countries, not a schedule each of them has to keep.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

A cross-sectional pattern does not say what any one country will do over time.

  • Supports

    How the global religious landscape changed from 2010 to 2020Economic development and religious affiliation

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: Economic development is not perfectly correlated with religion, and there are some countries with high rates of religious affiliation across the range of HDI scores.

Claim 54 · checked 23 September 2026 · AI-assisted audit

Claim: Whether the European pattern is the future arriving early or a regional peculiarity is genuinely open

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
Presented as open. The page says this is unresolved, and the review found that it is.

  • Supports

    The age gap in religion around the worldOverview, “Do age gaps mean the world is becoming less religious?”

    Cited · Link answered, 19 October 2026 · The cited passage was read, 23 September 2026

    The source reads: But even if parts of the world are secularizing, it is not necessarily the case that the world’s population, overall, is becoming less religious.

Claim 55 · checked 23 September 2026 · AI-assisted audit

Claim: and a confident answer either way is usually a statement about where the speaker lives.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s judgement, and the same point as its opening: an observer in London and one in Lagos see opposite trends.

What this does not establish

An observation about how the question is argued, not a measured finding.

Claim 56 · checked 23 September 2026 · AI-assisted audit

What a number settles

Claim: What a number settles

Kind
Framing. Signposting. It asserts nothing a source could check.

What the numbers are not

Claim: What the numbers are not

Kind
Framing. Signposting. It asserts nothing a source could check.

Claim: A count of adherents says nothing about whether the thing adhered to is true, in either direction. A shrinking religion is not thereby refuted and a growing one is not thereby confirmed.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

A point of logic about what demography can settle. It keeps the page from reading as an argument for or against the faith it counts.

Claim 59 · checked 23 September 2026 · AI-assisted audit

Claim: That is obvious when stated and routinely ignored in practice, in both directions, usually by whichever side the current trend suits.

Kind
This page’s reading. A judgement the page makes from the evidence. Others read it differently.
Review
A judgement, not a fact. Reviewed for whether it is fairly put and follows from the checked claims. No source can confirm it.

The page’s judgement, applied evenly to both sides.

What this does not establish

An observation, not a measured finding.

Claim 60 · checked 23 September 2026 · AI-assisted audit

What was read