What this archive contains
- Self-reported weekly attendance from an international survey conducted mainly in 1995–1997.
- Several entries based on an earlier 1990–1991 survey round.
- Context about differences between reported behavior and independently estimated attendance.
Understanding the historical survey
The World Values Survey interviewed adult samples across many societies. The archived figures below record the percentage of respondents who said they attended religious services at least weekly. Although the source framed much of the discussion as church attendance, the broader subject is participation in religious services.
The study reported that the United States had unusually high stated attendance among countries at a comparable level of economic development. In the same survey period, 53% of respondents in the United States described religion as very important in their lives, compared with 16% in Britain, 14% in France and 13% in Germany.
Researchers also observed a broad decline in attendance across many industrialized societies while interest in spiritual meaning did not necessarily decline at the same pace. In societies facing economic or political uncertainty, religion often retained greater reported importance.
Archive and accuracy note
These figures are more than a quarter-century old and should not be presented as current rates. Survey wording, sampling, social expectations and the meaning of “attendance” vary across cultures. Later research may use different methods and produce different estimates.
Reported weekly attendance
The original list has been retained as historical data and sorted from the highest reported percentage to the lowest. Asterisks mark figures drawn from the 1990–1991 survey round. Historical geographic labels are identified where helpful.
| Country or territory | Reported attendance |
|---|---|
| Nigeria | 89% |
| Ireland | 84%* |
| Philippines | 68% |
| Northern Ireland | 58% |
| South Africa | 56% |
| Poland | 55% |
| Puerto Rico | 52% |
| Portugal | 47%* |
| Slovakia | 47% |
| Mexico | 46% |
| Italy | 45%* |
| Dominican Republic | 44% |
| Belgium | 44% |
| United States | 44% |
| Turkey | 43% |
| Peru | 43% |
| India | 42% |
| Canada | 38%* |
| Brazil | 36% |
| Netherlands | 35% |
| Venezuela | 31% |
| Uruguay | 31% |
| Austria | 30%* |
| Britain | 27%* |
| Chile | 25% |
| Argentina | 25% |
| Spain | 25% |
| Slovenia | 22% |
| Croatia | 22% |
| Hungary | 21%* |
| France | 21%* |
| Romania | 20%* |
| Switzerland | 16% |
| Australia | 16% |
| Lithuania | 16% |
| South Korea | 14% |
| West Germany (historical label) | 14% |
| Czech Republic | 14%* |
| Taiwan | 11% |
| Bulgaria | 10%* |
| Ukraine | 10% |
| Moldova | 10% |
| Georgia | 10% |
| China | 9% |
| Armenia | 8% |
| Serbia | 7% |
| Montenegro | 7% |
| Azerbaijan | 6% |
| Belarus | 6% |
| Latvia | 5% |
| Denmark | 5%* |
| Norway | 5% |
| East Germany (historical region) | 5% |
| Sweden | 4% |
| Iceland | 4% |
| Finland | 4% |
| Estonia | 4% |
| Japan | 3% |
| Russia | 2% |
* Based on 1990–1991 survey data.
Why self-reports need caution
Research comparing survey answers with other attendance measures suggested that people in the United States and Canada sometimes reported more frequent participation than observational estimates indicated. The archived analysis cited approximate weekly rates of 20% in the United States and 10% in Canada under those alternative methods. Those estimates are themselves historical and method-dependent.
The same degree of over-reporting cannot simply be assumed in every country. Cultural expectations, religious calendars, congregation records and the definition of a service all affect comparison. A survey percentage is best understood as a response to a particular question in a particular time and place.
Patterns reported in 1998
A subsequent study covering almost 166,000 respondents found declining attendance in 15 of 19 industrialized democracies. Northern Ireland and Great Britain showed small increases. The study also reported increased attendance in five of eight formerly communist societies. These findings describe the survey period and do not establish present-day trends.
Sources recorded with the original essay
- A University of Michigan news release summarizing worldwide religiosity and attendance research, published in December 1997.
- A January 2000 journal report discussing the study's findings about spiritual questions and institutional attendance.
The original external hyperlinks are no longer included because their continued availability and integrity cannot be guaranteed. Readers using these data in research should locate the relevant World Values Survey documentation and cite the exact survey wave.