One hundred cities and districts were initially selected, with probability proportionate to size. Next 200 urban and rural blocks were randomly selected from the 100.
Excluded from the block selection were remote and low-populated areas, including some high-mountain terrain, and clusters of minority and ethnic groups whose command of Russian was limited. The lack of official statistics makes it difficult to estimate the amount of non-coverage, but it is probably somewhere between 3 and 6 percent.
In the blocks, interviewers were assigned random routes with rules to randomly select a household for the interview. Interviewers then recorded the number of adults in the household and randomly selected an adult for the interview. Interviewers revisited the home if the selected adult was not present.
Interviews were conducted in Russian.
As done routinely in surveys, results were weighted, or adjusted, to ensure that responses accurately reflect the population's makeup. The sample was weighted to take into account the sampling method, as well as for age and sex.
No more than one time in 20 should chance variations in the sample cause the results to vary by more than plus or minus 2.9 percentage points from the answers that would be obtained if all adults in Russia were polled.
There are other sources of potential error in polls, including the wording and order of questions.
The questions and results are available at http://www.ap-gfkpoll.com.