International Survey Research
Pew Research Center regularly conducts public opinion surveys in countries outside the United States as part of its ongoing exploration of attitudes, values and behaviors around the globe. To date, the center has conducted more than 480,000 interviews in 91 countries, mainly in conjunction with the longstanding Global Attitudes and Religion & Public Life projects. Cross-national studies constitute the bulk of Pew Research Center’s international survey research. Such studies pose special challenges when it comes to ensuring the comparability of data across multiple languages, cultures and contexts.
Pew Research Center staff are responsible for the overall design and execution of each cross-national survey project, including topical focus, questionnaire development, countries to be surveyed and sample design. The center’s staff frequently contract with a primary vendor to identify local, reputable research organizations, which are hired to translate questionnaires, administer surveys in the field and process data. Both primary vendors and local research organizations are consulted on matters of sampling, fieldwork logistics and translation. In addition, Pew Research often seeks the advice of subject matter experts and experienced survey researchers regarding the design and content of its cross-national studies.
Field periods for cross-national studies vary by country, but typically extend from three to six weeks. To the degree possible, Pew Research Center attempts to synchronize field work across countries in order to minimize the chance that major events or developments might interfere with the comparability of results. In some instances, Pew Research modifies its research strategy in response to key developments. For example, in spring 2010, the center fielded an additional 500 interviews in Mexico in order to assess the impact of the newly passed Arizona immigration law on Mexican views of the United States. Similarly, following the U.S. military raid that killed Osama bin Laden in May 2011, Pew Research fielded a new survey in Pakistan – within weeks of the preceding poll – to gauge the degree to which Pakistani views of al Qaeda and the U.S. were affected by bin Laden’s death.
Pew Research Center’s cross-national studies target the permanent, non-institutional adult population (ages 18 and older) in a given country. The center strives for samples that cover as much of the adult population as possible, given logistical, security and other constraints. Coverage limitations are noted in the detailed methods for each country. Examples of populations not covered in Pew Research surveys include: remote communities in Russia’s Far North; areas of Pakistan that are subject to insecurity; and cellphone-only users in Japan (where Pew Research Center relies on landline-only telephone surveys). As a general rule, Pew Research describes as “national” samples those that include 80% or more of a country’s adult population.
Sample sizes for Pew Research’s cross-national studies typically range from 700 to 1,500 interviews per country, with 1,000 interviews the most common sample size. Reported margins of error take into account not only the nominal sample size, but design effects due to clustering and weighting, where applicable. All margins of error are reported at the 95% confidence level, which means that 19 out of 20 times, the survey results would be expected to fall within the stated range either above or below the reported value.
In the case of both telephone and face-to-face surveys, weighting procedures standardly involve two stages. The first corrects for unequal selection probabilities, and the second adjusts key socio-demographic distributions – such as gender, age, education – to align as closely as possible with reliable, official population statistics.
Learn more about our international survey methods for our spring 2015 Global Attitudes survey here.
Pew Research Center is a charter member of the American Association of Public Opinion Research (AAPOR) Transparency Initiative.
In the next section, we provide additional information about questionnaire development and translation, plus face-to-face and telephone sample designs.