The typical World Poll survey in a country consists of 1,000 completed questionnaires. However, in some countries, oversamples may be collected in major cities. For example, we collected an additional 500 interviews in Moscow.
In countries where face-to-face surveys are conducted, census listings of Primary Sampling Units (PSU), consisting of clusters of households, are the main way of selecting the sample. Typically, the PSU are stratified this way:
-
Cities with population = 1,000,000 or more
-
Cities with population = 500,000 to 999,999
-
Cities with population = 100,000 to 499,999
-
Cities with population = 50,000 to 99,999
-
Towns with population = 10,000 to 49,999
-
Towns/Rural villages with populations less than 10,000
PSU are proportionally allocated to the population in each stratum, and typically 125 PSU are sampled with an average of eight interviews, one interview per sampled household, per PSU. If maps of the PSU are available, then they are used; otherwise, the selected PSU must be mapped. Random route procedures are used to select sampled households. Interviewers must make at least three attempts to survey the sampled household, unless an outright refusal occurs. If an interview cannot be obtained at the initial sampled household, the household to the immediate right of the initial household is selected. If the first attempt at this household is unsuccessful, then the house immediately to the left of the initial household is selected. Attempts to obtain an interview can be made at up to nine households.
In the RDD survey, at least five call attempts are made to reach a person, aged 15 and older, in each household. Typically the design is not stratified, but otherwise the other processes and procedures follow those used in the face-to-face design.
Contact Us
Acquire the power to understand and act on what the world is thinking. For more information, please e-mail worldpollpartners@gallup.com or call 202.715.3030 (Washington, D.C.) or +44 (0) 20 7950 4400 (London).
