In Person (Face-to-Face) Sampling

National samples for face-to-face surveys are typically drawn by an area probability method, which relies on US Census figures and maps of the country. The ANES uses a five-stage process to draw its 2008 national sample. This process is called a multi-stage area probability sample and it was developed by the Survey Research Center at The University of Michigan and the National Opinion Research Center at The University of Chicago many years ago. The 2008 survey was conducted by RTI and used some variations on the methods developed earlier:

• The first stage involved a sample of counties. The eight largest Metropolitan Statistical Areas (MSAs) all were included in this selection; other counties were selected in accordance with their population sizes.
• The second stage involved selection of census tracts within the selected counties. Census tracts were selected in accordance with their population sizes.
• The third stage involved selection of Census Block Groups (CBGs) within selected census tracts. CBGs were selected in accordance with their population sizes.
• The fourth stage involved selection of residential mailing addresses within each selected CBG. This resulted in some 4,598 households or an average of 14.4 per CBG.
• The fifth stage involved selection of respondents within selected households resulting in a overall survey sample of 3,088.

The 2008 ANES included an over-sample of African-Americans and Latinos. This was done in order to be able to make valid inferences to both of these groups. In a typical survey, the low number of African-Americans and Latinos (usually just over 10% of the sample each) generally results in too few respondents to represent the larger group. Over-sampling "beefs up" the validity of the observations by including more people. In order for the entire sample to be representative of the American public, the African-American and Latino sub-samples were weighted to bring them back to their true population parameters.