The world's religious map is changing. Why does Europe tell a different story?

SVET ZA 5 MINÚT · 11 October 2026 · approximately 9 minutes to read

Fewer people in Europe identify as Christian. Does that mean the whole world is heading in the same direction? And when more people say they have no religious affiliation, are they automatically becoming atheists?

The answers are more interesting than a simple division between a “religious world” and a “non-believing Europe.” Behind the percentages lie generations of families, shifting borders, wars, personal choices and political pressure. The same religious label can also mean very different things to two different people.

First, what are we comparing?

The historical comparison in our reel comes from World Christianity 2025: Regional Perspectives, a study drawing on the World Christian Database. Its figures for 1900 are historical reconstructions; those for 2025 are projections published around the turn of 2024–2025. These are not two perfectly accurate worldwide censuses. [1]

For the country rankings below, we use a separate Pew Research Center dataset, published in 2025 and describing the population in 2020. That distinction matters: publication year is not measurement year. International comparisons require researchers to assemble and reconcile many national sources. [2, 3]

We therefore do not join these sources into a single trend line. A difference between their findings need not mean that hundreds of millions of people changed their beliefs within five years. It may reflect different definitions, evidence and estimation methods. Religious affiliation is a more complicated variable than someone's age or height.

The research institution's background also deserves transparency. The historical study's authors are affiliated with the Center for the Study of Global Christianity at Gordon-Conwell Theological Seminary. That affiliation alone establishes neither unreliability nor accuracy. What matters is understanding where the figures come from, what they measure and where they differ from other research. [1]

Europe is changing. But the world is not an enlarged Europe

In the historical series used in the reel, Christians fall from almost 95% of Europe's population in 1900 to 74.5% in 2025. Globally, the change is much smaller: 34.5% to 32.3%. These are population shares, not numbers of people. [1]

Imagine a town of 100 residents, 30 of whom are Christian. Several decades later, the town has 200 residents and 50 Christians. The number of Christians has risen, but their share has fallen from 30% to 25%. Without an explanation, saying that “Christianity is declining” could describe two different things.

The global average works in a similar way. A region with slow-growing or stable population does not necessarily outweigh a region where the population is expanding rapidly. Europe's experience is therefore not a universal forecast for the future of every other society.

In its separate comparison of 2010 and 2020, Pew found that the number of Christians worldwide increased even as their global share fell. Sub-Saharan Africa also overtook Europe in its Christian population. Leaving Christian affiliation is an important component of the decline in many Western countries. [4]

Part of the answer to why Christianity has not simply followed Europe's religious trajectory lies in the changing demographic weight of different regions. Christianity is not a European institution with a few overseas branches. It is a global community encompassing different languages, traditions and social experiences.

Belief, membership and having no religion are different things

Asked “What is your religion?”, someone might answer “Christianity” despite rarely attending church. Another person might say “none” while believing in a spiritual world, praying or observing family rituals.

We therefore need to distinguish at least three things: affiliation, personal belief and practice. A single survey question cannot capture them all. Nor does a high historical Christian share tell us how many people prayed regularly or agreed with their church's teachings.

“Religiously unaffiliated” includes atheists, agnostics and people who do not identify with any organised tradition. An atheist does not believe in a god or gods; agnosticism concerns uncertainty or the limits of knowledge. In everyday responses, these categories can overlap.

China offers an important example. Pew notes that people who do not report a religious affiliation may still participate in spiritual or traditional rituals. The Western idea of membership in one particular religion does not describe every way people relate to tradition. [5]

This is why asking whether religion is “disappearing” requires another question: which part of religion? Formal membership could decline while some customs continue. A public identity might persist even when private belief changes. A useful comparison must tell readers which of these it measures.

Which countries are the most Christian, Muslim or non-religious?

The following ranking compares shares of residents, not numbers of believers. It uses Pew's estimates for 2020 and covers the countries and territories included in that study: 201 places with at least 100,000 residents in 2010 or 2020. It is not a ranking of every microstate, including Vatican City. [2, 3]

Source: Pew, Appendix B. Ties reflect the published rounding. These percentages estimate affiliation, not strength of belief or religious freedom. Israel follows the source's geographic definition; the figure does not combine Israel with the Palestinian territories. [2]

This table cannot reliably identify the “most atheist” country. China has the highest unaffiliated share, but claiming that 89.6% of Chinese people are convinced atheists would be wrong. Such a ranking would require a comparable question about personal belief and equally strong evidence across countries.

Islam also illustrates the difference between shares and numbers. Indonesia has the largest Muslim population in absolute terms, but not the highest Muslim percentage. A large country with a lower share can have more adherents than a small country with almost universal affiliation. [6]

Nor does a near-universal religious label tell us that a country is internally uniform. The category alone cannot reveal differences in observance, interpretation, family traditions or individual convictions. A league table is a starting point for questions, not a complete portrait of a society.

Why are Czechia and Slovakia so different?

In the same Pew dataset, the religiously unaffiliated account for 72.8% of Czechia's population and 25.3% of Slovakia's. Christians account for 26.4% and 73.7%, respectively. Again, these are estimates for 2020, not a reproduction of national census results. [2]

A shared state and a similar experience of communist government did not produce identical religious societies. We cannot attribute everything to one political period. Research on Central and Eastern Europe identifies different paths after communism; in the Czech context, it also highlights older relationships between religion, national identity and church authority. [7]

None of these percentages demonstrates that Czechs are “more rational” or Slovaks “more moral.” Such judgments would put something into the numbers that they do not measure. More useful questions concern how convictions are passed down within families and what place churches occupy in everyday life.

Neighbours can share a language environment, economic connections and a political past without sharing the same relationship with religious institutions. That is a reason to examine local history carefully, rather than treating a continent as one homogeneous religious community.

Generations, migration and war: different mechanisms of change

Religious groups change through births, deaths and changes of affiliation, and at country level through migration as well. Age structure matters: a younger population has more people approaching the years when families are formed. Pew uses these mechanisms to explain important changes between 2010 and 2020. [8]

Imagine two groups of equal size. One contains many young adults; the other is dominated by older generations. Their future trajectories can differ without anyone persuading anyone else to adopt another faith. Demography is not a referendum on which religious teaching people find most convincing.

Fertility also needs careful interpretation. A group's average cannot tell us what an individual woman or family will decide. Explanations need to examine education, economic circumstances, healthcare and access to family planning as well. A religious label by itself does not separate these influences.

Migration changes the local map, but moving across a border does not by itself change the worldwide number of adherents. A Muslim or Christian family moving to Europe changes the religious composition of two countries. At global level, it remains the same family. People with no religion migrate too.

This distinction helps separate two questions often merged in public debate: “Why has this country's population changed?” and “Why has this religion's global share changed?” The answers need not be the same. A local explanation should not automatically become a worldwide one.

Not all changes were peaceful. European Jewish history cannot be understood without the Holocaust: roughly two-thirds of Europe's pre-war Jewish population were murdered. Subsequent emigration also shaped the map. This decline cannot be explained simply as a gradual weakening of belief. [9, 1]

What is hidden inside “other religions”?

In the historical chart, this combined category includes Chinese folk traditions, ethnic religions, Sikhism and new religious movements, among others. It adds together different groups; it is not one religion. [1]

Particular care is needed when comparing databases here. Someone who does not formally identify with a religion but observes traditional rituals may be classified differently under different methods. A shrinking bar cannot automatically be interpreted as an equivalent disappearance of spiritual practices.

Imagine a drawer labelled “other.” If part of its contents is moved to a more precisely named drawer, the first drawer becomes smaller even though nothing has disappeared. Long-term comparisons need to distinguish actual social change from changes in how we describe it.

That does not make all estimates useless or all comparisons impossible. It means that a chart should keep its categories and sources consistent and explain their limits. The alternative is to turn a difference in classification into a dramatic story that the evidence does not support.

What does this mean for everyday life?

The following connections are questions for society, not evidence that a particular percentage of believers automatically causes a particular outcome.

Culture and public space. How should a church building be used when its local congregation no longer needs it on the same scale? How can its heritage be protected while respecting its significance to believers? How should communities accommodate changing needs for burial grounds and ceremonial spaces?

Schools, families and living together. More diverse societies need to decide how to teach about religions, protect freedom of religion and freedom not to believe, and prevent discrimination. The central question is not only which group is largest, but also how minorities and individuals are treated.

Politics and public discussion. Religious identity can be part of political language, but it is not a reliable guide to an individual's opinions. Two members of the same church may disagree profoundly about migration, social policy or personal rights. Treating them as a single political voice hides that diversity.

These questions also show why changes in affiliation matter beyond places of worship. They concern shared spaces, public institutions and the rules people use to resolve disagreement. The percentages help describe the setting; they do not decide what those rules should be.

A religious map shows how people are classified or identify themselves. To understand how they live, we also need to understand their experiences, institutions and freedom to choose. That is why the history behind the percentages is more revealing than a ranking alone.

Sources and methodological note

The historical percentages are rounded calculations from one series. Pew's figures are used separately for country comparisons and more recent trends. Differences between the sources are not treated as changes over time. The examples and questions in the final section are explanatory analysis, not measured findings from the studies.

1. Zurlo, G. A., Johnson, T. M., Crossing, P. F.: World Christianity 2025: Regional Perspectives. International Bulletin of Mission Research, 49(1), 2025. Historical series and group definitions.
2. Pew Research Center: Religious composition by country — Appendix B, 2025. Country percentages and rankings for 2020.
3. Pew: Global religious change — Methodology. Coverage, definitions and estimation sources.
4. Pew: Christian population change, 2025. Regional patterns, shares and counts.
5. Pew: Non-religion in China, 2023. Affiliation versus practice.
6. Pew: Muslim population change, 2025. Distribution of the Muslim population.
7. Pew: Religious Belief and National Belonging in Central and Eastern Europe, 2017. Historical and social differences.
8. Pew: Factors driving religious change, 2010–2020, 2025. Demographic mechanisms.
9. United States Holocaust Memorial Museum: Jewish Population of Europe in 1933. Pre-war population and the Holocaust.