Rage 2 Rebuild · Worldwide evidence library

What does the data show
about leaving religion?

Rage 2 Rebuild’s worldwide evidence library compares religious identity in 2010 and 2020 across 201 countries and territories. It provides 877 reproducible calculations from Pew’s rounded estimates, plus a separate adult switching table. These datasets describe populations and reported identity; they do not measure religious harm, personal safety or recovery.

Which worldwide religious-identity figures can I examine?

Rage 2 Rebuild presents Pew’s estimates for 201 countries and territories, six regions and the world. The table describes people of all ages in seven broad religious categories in 2010 and 2020. Its 877 calculated changes each identify a place, category and period. They compare rounded shares, not individual journeys.

The underlying research belongs to Pew Research Center. The calculations and explanatory presentation are our work. This is public research with original analysis, not a proprietary survey, a sample of our visitors, or proof that Elder X has helped a particular number of people.

Download the calculations, inputs, and source references (CSV) ↓

What does a change in religious population share mean?

Rage 2 Rebuild calculates worldwide changes as the 2020 share minus the 2010 share, in percentage points. Moving from 20% to 25% is a five-point difference between population estimates. It does not show that five percent of the same people changed religion. Rounding, demographics and definitions affect how the result can be interpreted.

A percentage is not a percentage point.

If a group changes from 20% to 25% of the population, the difference is 5 percentage points. It is not evidence that 5% of the same people changed religion. These are population estimates at two dates.

“Unaffiliated” is a category, not a verdict.

It describes people with no religious affiliation. They may differ in beliefs, rituals, spirituality, family relationships, or reasons for answering that way. It does not mean they all left religion or experienced harm.

Small numbers stay uncertain.

The original table rounds values. “Less than 0.1%” stays an inequality, not zero. When either input uses that notation, we do not report a change. Rounding can hide small differences; we do not rank tiny changes as meaningful trends.

A national average is not a safety plan.

These figures do not measure local acceptance, family reactions, legal protections, therapy outcomes, or distress. Births, deaths, migration, measurement, and religious switching can all influence population composition. Review present circumstances separately.

Who was raised with a religion and no longer identifies with one?

Rage 2 Rebuild’s separate switching table covers 22 countries in Pew’s March 26, 2025 report. The denominator in every column is all adults surveyed in the country. Participants recalled their childhood affiliation and reported current identity in the 2023–2024 surveys. Those measures describe reported transitions, not distress or the effects of receiving support.

Religious switching: percentages of all adult respondents in each country
CountryRaised unaffiliatedLeft unaffiliated groupBecame unaffiliatedCurrently unaffiliated
Spain10%2%35%44%
Sweden22%2%32%52%
South Korea27%9%31%49%
Germany14%2%31%43%
Canada14%3%30%41%
Netherlands26%2%30%53%
United Kingdom19%2%29%46%
Australia27%4%28%51%
Japan31%3%27%55%
France22%3%25%44%
Chile10%2%25%33%
Italy3%1%21%23%
Colombia5%3%20%23%
United States13%3%20%29%
Argentina10%4%17%24%
Mexico4%1%17%20%
Singapore11%4%13%20%
Brazil5%2%12%15%
Greece3%1%12%14%
Peru3%1%10%12%
South Africa5%2%8%11%
Hungary21%4%2%19%

“Became unaffiliated” describes adults raised with a religion who now report no religious affiliation. It is not the percentage of religiously raised people who left: that would use a different denominator. These figures do not count all changes within a broad religion, and they do not measure religious harm. Rounding means the columns may not add up exactly. Read the original table, PDF page 33 (printed page 32).

This survey table and the 2010–2020 population estimates below have different populations, dates, and methods. They should not be joined into a single trend line or compared as if they were the same study.

Keep the source attached to the number

How should I cite a religious-identity calculation?

Cite Rage 2 Rebuild’s calculation and the original Pew table together, naming the place, category, all-age population, 2010–2020 period and percentage-point measure. Each of the 877 worldwide records includes both inputs and source pages. Where a country guide exists, its exact table-row link lets another reader examine the same evidence.

Rage 2 Rebuild. Religious-composition change calculations, 2010–2020. Version September 5, 2026. Secondary analysis of Pew Research Center's rounded Appendix B estimates, published June 9, 2025. Measure: change in all-age population share, in percentage points. Add the place, category and original PDF page for the particular finding.

Download all 877 calculation records as JSON or download the CSV. Each JSON record has an identifier, both input shares, the calculated difference and the original source pages. Records with a matching country guide also link to the exact visible table row.

A composition change is not a count of departures

The subtraction compares two population estimates. It does not follow the same people for ten years, distinguish private doubt from public identity, or count how many people left a particular congregation. The separate switching table compares recalled childhood identity with adults' current responses. Neither dataset can tell a reader how much guilt they should feel or how quickly life will improve.

When either published share is an inequality, its exact value is unknown. Omitting that difference is a limitation of the calculation, not evidence of no change. Small rounded differences also should not be turned into league tables of where people are more accepting or where leaving is easier.

Use and attribution

The original surveys, estimates and tables remain the work of their named publisher. These downloads explain our arithmetic and retain its provenance; they do not confer a new license to the underlying source material. Country-report excerpts elsewhere use the separately identified Freedom of Thought license. Read how we distinguish original guidance, calculations and quoted evidence.

How can I compare religious identity by place?

Rage 2 Rebuild’s explorer lets you choose a religious category and search the worldwide place records. The results show the 2010 share, 2020 share and calculated difference where both inputs are numeric. Select a linked country name to open its guide and exact evidence table. An omitted difference is not evidence of no change.

208 places shown. All ages; national or regional estimates.

Scroll the table sideways on a narrow screen to see the original source links.

Religiously unaffiliated: estimated population share and change, 2010–2020
Place20102020Change (pp)Original table
World23.3%24.2%+0.9 ppPage 1
Asia-Pacific33.0%32.8%-0.2 ppPage 1
Europe18.7%25.3%+6.6 ppPage 1
Latin America-Caribbean7.8%11.9%+4.1 ppPage 1
Middle East-North Africa0.4%0.4%0.0 ppPage 1
North America17.2%30.2%+13.0 ppPage 1
Sub-Saharan Africa3.1%2.6%-0.5 ppPage 1
AfghanistanLess than 0.1%Less than 0.1%Not calculatedPage 1
Albania9.5%7.7%-1.8 ppPage 1
Algeria1.1%1.3%+0.2 ppPage 2
Angola4.5%6.1%+1.6 ppPage 2
Argentina9.7%9.2%-0.5 ppPage 2
Armenia1.2%1.2%0.0 ppPage 2
Aruba5.6%5.9%+0.3 ppPage 2
Australia24.9%42.3%+17.4 ppPage 2
Austria13.5%22.4%+8.9 ppPage 2
Azerbaijan1.1%4.8%+3.7 ppPage 2
Bahamas1.9%1.9%0.0 ppPage 2
Bahrain0.4%0.3%-0.1 ppPage 2
BangladeshLess than 0.1%Less than 0.1%Not calculatedPage 3
Barbados20.8%20.1%-0.7 ppPage 3
Belarus5.0%13.8%+8.8 ppPage 3
Belgium32.0%39.0%+7.0 ppPage 3
Belize16.0%15.6%-0.4 ppPage 3
Benin4.4%4.9%+0.5 ppPage 3
BhutanLess than 0.1%Less than 0.1%Not calculatedPage 3
Bolivia4.0%9.5%+5.5 ppPage 3
Bosnia-Herzegovina1.1%1.0%-0.1 ppPage 3
Botswana14.7%14.7%0.0 ppPage 3
Brazil8.1%13.5%+5.4 ppPage 4
Brunei0.2%0.2%0.0 ppPage 4
Bulgaria8.4%10.0%+1.6 ppPage 4
Burkina Faso0.6%0.6%0.0 ppPage 4
Burundi1.9%1.3%-0.6 ppPage 4
CambodiaLess than 0.1%Less than 0.1%Not calculatedPage 4
Cameroon3.1%2.1%-1.0 ppPage 4
Canada23.9%34.6%+10.7 ppPage 4
Cape Verde13.6%20.6%+7.0 ppPage 4
Central African Republic1.0%0.5%-0.5 ppPage 4
Chad2.7%3.1%+0.4 ppPage 5
Channel Islands14.1%14.2%+0.1 ppPage 5
Chile13.6%30.3%+16.7 ppPage 5
China87.4%89.6%+2.2 ppPage 5
Colombia7.2%9.9%+2.7 ppPage 5
Comoros0.1%0.1%0.0 ppPage 5
Costa Rica7.5%10.1%+2.6 ppPage 5
Croatia4.7%6.7%+2.0 ppPage 5
Cuba23.5%21.6%-1.9 ppPage 5
Curacao5.9%5.9%0.0 ppPage 5
Cyprus6.5%6.6%+0.1 ppPage 6
Czech Republic68.5%72.8%+4.3 ppPage 6
D.R. Congo1.4%1.2%-0.2 ppPage 6
Denmark14.5%16.6%+2.1 ppPage 6
Djibouti1.1%1.1%0.0 ppPage 6
Dominican Republic14.9%19.6%+4.7 ppPage 6
Ecuador4.3%8.4%+4.1 ppPage 6
EgyptLess than 0.1%Less than 0.1%Not calculatedPage 6
El Salvador11.6%11.9%+0.3 ppPage 6
Equatorial Guinea5.0%5.0%0.0 ppPage 6
Eritrea1.0%1.1%+0.1 ppPage 7
Estonia31.5%43.6%+12.1 ppPage 7
Eswatini4.0%6.2%+2.2 ppPage 7
EthiopiaLess than 0.1%Less than 0.1%Not calculatedPage 7
Federated States of Micronesia0.7%0.7%0.0 ppPage 7
Fiji0.3%0.3%0.0 ppPage 7
Finland16.6%25.0%+8.4 ppPage 7
France34.5%42.6%+8.1 ppPage 7
French Guiana3.3%3.4%+0.1 ppPage 7
French Polynesia4.9%5.0%+0.1 ppPage 7
Gabon7.8%5.8%-2.0 ppPage 8
GambiaLess than 0.1%Less than 0.1%Not calculatedPage 8
Georgia0.4%1.0%+0.6 ppPage 8
Germany31.8%36.1%+4.3 ppPage 8
Ghana5.3%1.1%-4.2 ppPage 8
Greece3.7%4.7%+1.0 ppPage 8
Grenada5.8%5.8%0.0 ppPage 8
Guadeloupe2.5%2.5%0.0 ppPage 8
Guam1.6%1.8%+0.2 ppPage 8
Guatemala6.1%6.0%-0.1 ppPage 8
Guinea3.4%1.3%-2.1 ppPage 9
Guinea-Bissau2.4%11.6%+9.2 ppPage 9
Guyana3.1%2.9%-0.2 ppPage 9
Haiti7.6%9.4%+1.8 ppPage 9
Honduras9.6%9.7%+0.1 ppPage 9
Hong Kong68.2%71.4%+3.2 ppPage 9
Hungary25.0%27.0%+2.0 ppPage 9
Iceland15.9%19.8%+3.9 ppPage 9
IndiaLess than 0.1%Less than 0.1%Not calculatedPage 9
IndonesiaLess than 0.1%Less than 0.1%Not calculatedPage 9
IranLess than 0.1%Less than 0.1%Not calculatedPage 10
IraqLess than 0.1%Less than 0.1%Not calculatedPage 10
Ireland6.1%15.6%+9.5 ppPage 10
Israel3.6%4.4%+0.8 ppPage 10
Italy6.7%13.3%+6.6 ppPage 10
Ivory Coast10.0%5.6%-4.4 ppPage 10
Jamaica21.8%23.7%+1.9 ppPage 10
Japan52.2%57.5%+5.3 ppPage 10
JordanLess than 0.1%Less than 0.1%Not calculatedPage 10
Kazakhstan2.9%2.5%-0.4 ppPage 10
Kenya2.4%1.6%-0.8 ppPage 11
KiribatiLess than 0.1%0.1%Not calculatedPage 11
Kosovo0.1%0.1%0.0 ppPage 11
Kuwait0.2%0.2%0.0 ppPage 11
Kyrgyzstan1.6%1.2%-0.4 ppPage 11
LaosLess than 0.1%Less than 0.1%Not calculatedPage 11
Latvia11.0%17.0%+6.0 ppPage 11
LebanonLess than 0.1%Less than 0.1%Not calculatedPage 11
Lesotho1.5%1.7%+0.2 ppPage 11
Liberia1.5%1.1%-0.4 ppPage 11
LibyaLess than 0.1%Less than 0.1%Not calculatedPage 12
Lithuania6.8%7.1%+0.3 ppPage 12
Luxembourg20.2%25.3%+5.1 ppPage 12
Macao67.7%67.6%-0.1 ppPage 12
Madagascar22.5%22.3%-0.2 ppPage 12
Malawi1.0%0.9%-0.1 ppPage 12
Malaysia1.7%0.8%-0.9 ppPage 12
MaldivesLess than 0.1%Less than 0.1%Not calculatedPage 12
Mali1.4%2.5%+1.1 ppPage 12
Malta2.5%5.2%+2.7 ppPage 12
Martinique2.4%2.7%+0.3 ppPage 13
Mauritania0.1%0.1%0.0 ppPage 13
Mauritius0.7%0.8%+0.1 ppPage 13
Mayotte0.2%0.2%0.0 ppPage 13
Mexico4.8%10.6%+5.8 ppPage 13
Moldova1.3%0.2%-1.1 ppPage 13
Mongolia37.4%40.8%+3.4 ppPage 13
Montenegro1.3%1.1%-0.2 ppPage 13
Morocco0.1%0.1%0.0 ppPage 13
Mozambique18.9%14.3%-4.6 ppPage 13
MyanmarLess than 0.1%Less than 0.1%Not calculatedPage 14
Namibia0.2%5.6%+5.4 ppPage 14
NepalLess than 0.1%Less than 0.1%Not calculatedPage 14
Netherlands45.5%54.1%+8.6 ppPage 14
New Caledonia10.0%10.5%+0.5 ppPage 14
New Zealand44.2%51.4%+7.2 ppPage 14
Nicaragua12.5%13.3%+0.8 ppPage 14
Niger0.2%1.2%+1.0 ppPage 14
Nigeria0.4%0.4%0.0 ppPage 14
North Korea72.9%72.9%0.0 ppPage 14
North Macedonia1.4%0.5%-0.9 ppPage 15
Norway14.7%22.5%+7.8 ppPage 15
OmanLess than 0.1%0.1%Not calculatedPage 15
PakistanLess than 0.1%Less than 0.1%Not calculatedPage 15
Palestinian territoriesLess than 0.1%Less than 0.1%Not calculatedPage 15
Panama6.0%6.2%+0.2 ppPage 15
Papua New Guinea0.4%0.3%-0.1 ppPage 15
Paraguay0.5%4.2%+3.7 ppPage 15
Peru2.8%5.1%+2.3 ppPage 15
PhilippinesLess than 0.1%Less than 0.1%Not calculatedPage 15
Poland2.7%8.6%+5.9 ppPage 16
Portugal7.2%13.8%+6.6 ppPage 16
Puerto Rico10.3%10.4%+0.1 ppPage 16
Qatar0.3%0.3%0.0 ppPage 16
Republic of the Congo5.6%5.2%-0.4 ppPage 16
Reunion2.1%2.1%0.0 ppPage 16
Romania0.2%0.9%+0.7 ppPage 16
Russia13.0%20.2%+7.2 ppPage 16
Rwanda0.9%1.0%+0.1 ppPage 16
Samoa0.1%Less than 0.1%Not calculatedPage 16
São Tomé and Príncipe12.5%9.7%-2.8 ppPage 17
Saudi ArabiaLess than 0.1%0.1%Not calculatedPage 17
SenegalLess than 0.1%Less than 0.1%Not calculatedPage 17
Serbia1.2%4.0%+2.8 ppPage 17
Seychelles2.2%2.4%+0.2 ppPage 17
Sierra LeoneLess than 0.1%Less than 0.1%Not calculatedPage 17
Singapore16.6%19.9%+3.3 ppPage 17
Slovakia24.7%25.3%+0.6 ppPage 17
Slovenia26.4%32.3%+5.9 ppPage 17
Solomon Islands0.3%0.3%0.0 ppPage 17
SomaliaLess than 0.1%Less than 0.1%Not calculatedPage 18
South Africa5.8%3.1%-2.7 ppPage 18
South Korea41.3%48.3%+7.0 ppPage 18
South Sudan0.4%0.4%0.0 ppPage 18
Spain19.0%26.4%+7.4 ppPage 18
Sri LankaLess than 0.1%Less than 0.1%Not calculatedPage 18
St. Lucia6.0%5.9%-0.1 ppPage 18
St. Vincent and the Grenadines7.9%7.9%0.0 ppPage 18
Sudan0.1%0.6%+0.5 ppPage 18
Suriname7.4%8.1%+0.7 ppPage 18
Sweden21.5%28.9%+7.4 ppPage 19
Switzerland20.1%30.8%+10.7 ppPage 19
Syria1.9%2.0%+0.1 ppPage 19
Taiwan22.8%23.1%+0.3 ppPage 19
TajikistanLess than 0.1%Less than 0.1%Not calculatedPage 19
Tanzania2.2%3.5%+1.3 ppPage 19
ThailandLess than 0.1%Less than 0.1%Not calculatedPage 19
Timor-Leste0.2%Less than 0.1%Not calculatedPage 19
Togo7.6%8.0%+0.4 ppPage 19
Tonga0.3%0.6%+0.3 ppPage 19
Trinidad and Tobago2.5%2.4%-0.1 ppPage 20
Tunisia0.4%0.4%0.0 ppPage 20
Turkey0.2%2.5%+2.3 ppPage 20
Turkmenistan0.5%Less than 0.1%Not calculatedPage 20
U.S. Virgin Islands3.7%4.0%+0.3 ppPage 20
UgandaLess than 0.1%Less than 0.1%Not calculatedPage 20
Ukraine13.2%15.1%+1.9 ppPage 20
United Arab Emirates0.3%0.3%0.0 ppPage 20
United Kingdom28.8%40.2%+11.4 ppPage 20
United States16.5%29.7%+13.2 ppPage 20
Uruguay36.3%52.4%+16.1 ppPage 21
Uzbekistan0.6%0.6%0.0 ppPage 21
Vanuatu1.1%1.4%+0.3 ppPage 21
Venezuela8.4%9.7%+1.3 ppPage 21
Vietnam62.1%67.7%+5.6 ppPage 21
Western Sahara0.3%0.3%0.0 ppPage 21
YemenLess than 0.1%Less than 0.1%Not calculatedPage 21
ZambiaLess than 0.1%Less than 0.1%Not calculatedPage 21
Zimbabwe12.5%10.5%-2.0 ppPage 21

Does religious switching measure a different kind of change?

Yes. Rage 2 Rebuild distinguishes worldwide population composition from Pew’s separate adult switching study. The latter compares recalled childhood affiliation with current reported identity, rather than comparing all-age populations ten years apart. It covers 36 surveyed countries, with smaller eligible sets for some analyses. Neither approach measures outcomes from using this website.

Pew’s separate 2025 report on religious switching in 36 countries compares what adults say they were raised in with their current identity. That is closer to the question of leaving, but its broad categories do not capture every move between denominations or every private change of belief.

Neither study measures the success of this website. We do not publish invented guilt, shame, or recovery percentages. If we conduct original research in future, it will need separate voluntary research consent, a published method, a defined sample, and protection against identifying people in small groups. A contact message is never automatically a research contribution.

Where can I check a source or report an error?

Rage 2 Rebuild links the worldwide calculations to Pew’s Appendix B, published June 9, 2025, and its methodology. Check the place, category and source page before reporting a discrepancy. A correction inquiry only needs the claim and evidence; a personal religious history is unnecessary. Source publication dates remain separate from this website’s editorial date.

If you spot an error, send the place, category, source page, and proposed correction through the contact form. No personal religious history is needed. We will review the source before changing a figure.

Read it, see it, or listen

Read a number without giving it the wrong meaning.

Take one part at a time. The illustrations below summarize choices to consider; they do not depict a person’s experience or promise an outcome.

Identify the measure: Population composition; Adult switching survey; Country assessment
1 / 4

Identify the measure

Start with what a source actually measures. Population composition, childhood-to-current adult identity and institutional country assessments answer different questions. They are not interchangeable measures of distress, individual departures or the effect of this website.

Keep the denominator: All ages or adults?; Which country?; Which years?
2 / 4

Keep the denominator

Read the denominator, geography and dates with the number. A share of all adults is different from a share of adults raised in a religion. A national estimate cannot be treated as a measurement of a city, family or support group.

Check the calculation: Percentage points; Original rounding; No inequality subtraction
3 / 4

Check the calculation

The composition change column subtracts the rounded 2010 share from the rounded 2020 share, in percentage points. It is not relative percentage growth. When an input is an inequality, the site leaves the change uncalculated.

Trace the evidence: Source and page; Method and limits; A correction with context
4 / 4

Trace the evidence

Use the linked original source and methodology to check a claim. A correction is most useful when it identifies the place, category, source page and proposed change. No personal religious history is needed to report a data concern.

Original editorial graphics and synthetic narration. This is not a recording of Elder X, a client, a therapist or a documented event.

Listen in about a minute.

The video reads the four illustrated sections above. You can pause, turn on captions or use the written version at your own pace. Playback is optional and does not submit a response.

Read the complete video transcript

Start with what a source actually measures. Population composition, childhood-to-current adult identity and institutional country assessments answer different questions. They are not interchangeable measures of distress, individual departures or the effect of this website.

Read the denominator, geography and dates with the number. A share of all adults is different from a share of adults raised in a religion. A national estimate cannot be treated as a measurement of a city, family or support group.

The composition change column subtracts the rounded 2010 share from the rounded 2020 share, in percentage points. It is not relative percentage growth. When an input is an inequality, the site leaves the change uncalculated.

Use the linked original source and methodology to check a claim. A correction is most useful when it identifies the place, category, source page and proposed change. No personal religious history is needed to report a data concern.

Bring your own question to Elder X
Religion & Leaving Religion: Evidence Library | Rage 2 Rebuild