Results for the survey are based on telephone and face-to-face interviews conducted under the direction of Princeton Survey Research Associates International, which manages the fieldwork conducted by local research organizations in each country. All surveys are based on national samples except in China, India, and Pakistan, where the sample was disproportionately or exclusively urban.

The table below shows the margin of sampling error based on all interviews conducted in that country. For results based on the full sample in a given country, one can say with 95% confidence that the error attributable to sampling and other random effects is plus or minus the margin of error. In addition to sampling error, one should bear in mind that question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of opinion polls.

Great Britain Sample design: Probability Mode: Telephone adults 18 plus Languages: English, Urdu, and Arabic Fieldwork dates: April 4-26, 2006 Sample size: 902 including a 412 Muslim oversample; general public sample is weighted to be representative of the overall population Margin of Error: 6% for general public; 6% for Muslim oversample Representative: Telephone households

China Sample design: Probability sample in six cities and surrounding rural areas — Shanghai, Beijing, Guangzhou, Xinxiang, Jinzhong, and Luzhou Mode: Face-to-face adults 18 to 60 Languages: Chinese (dialects: Mandarin, Beijingese, Cantonese, Sichun, Hubei, Dongbei, Shanghaiese) Fieldwork dates: April 7-18, 2006 Sample size: 2180 Margin of Error: 2% Representative: Disproportionately urban

Egypt Sample design: Probability Mode: Face-to-face adults 18 plus Languages: Arabic Fieldwork dates: April 5-27, 2006 Sample size: 1000 Margin of Error: 3% Representative: Adult population

France Sample design: Quota Mode: Telephone adults 18 plus Languages: French Fieldwork dates: April 5-19, 2006 Sample size: 905 including a 400 Muslim oversample; general public sample is weighted to be representative of the overall population Margin of Error: 4% for general public; 5% for Muslim oversample Representative: Telephone households

Germany Sample design: Probability Mode: Telephone adults 18 plus Languages: German and Turkish Fieldwork dates: April 5-27, 2006 Sample size: 902 including a 413 Muslim oversample; general public sample is weighted to be representative of the overall population Margin of Error: 6% for general public; 6% for Muslim oversample Representative: Telephone households

India Sample design: Probability Mode: Face-to-face adults 18-64 Languages: Hindi, Gujarati, Tamil, Kannada, Bengali Fieldwork dates: April 15-May 3, 2006 Sample size: 2029 Margin of Error: 2% Representative: Urban only

Indonesia Sample design: Probability Mode: Face-to-face adults 18 plus Languages: Bahasa Indonesia Fieldwork dates: April 8-30, 2006 Sample size: 1022 Margin of Error: 3% Representative: Eighteen provinces representing 87% of population

Japan Sample design: Probability Mode: Telephone adults 18 plus Languages: Japanese Fieldwork dates: March 31-April 21, 2006 Sample size: 500 Margin of Error: 5% Representative: Telephone households

Jordan Sample design: Probability Mode: Face-to-face adults 18 plus Languages: Arabic Fieldwork dates: April 5-27, 2006 Sample size: 1000 Margin of Error: 3% Representative: Adult population

Nigeria Sample design: Probability Mode: Face-to-face adults 18 plus Languages: Yoruba, Igbo, Hausa, and English Fieldwork dates: April 20-29, 2006 Sample size: 1000 Margin of Error: 3% Representative: Adult population

Pakistan Sample design: Probability Mode: Face-to-face adults 18 plus Languages: Urdu Fieldwork dates: April 7-28, 2006 Sample size: 1277 Margin of Error: 3% Representative: Disproportionately urban

Russia Sample design Probability Mode: Face-to-face adults 18 plus Languages: Russian Fieldwork dates: April 6-16, 2006 Sample size: 1000 Margin of Error: 3% Representative: Adult population

Spain Sample design: Probability Mode: Face-to-face adults 18 plus Languages: Spanish and Arabic Fieldwork dates: April 7-May 4, 2006 Sample size: 979 including a 402 Muslim oversample; general public sample is weighted to be representative of the overall population Margin of Error: 4% for general public; 5% for Muslim oversample Representative: Adult population

Turkey Sample design: Probability Mode: Face-to-face adults 18 plus Languages: Turkish Fieldwork dates: April 1-25, 2006 Sample size: 1013 Margin of Error: 3% Representative: Adult population

United States Sample design: Probability Mode: Telephone adults 18 plus Languages: English Fieldwork dates: May 2-14, 2006 Sample size: 1001 Margin of Error: 3% Representative: Telephone households in continental U.S.