Sample and Interviewing
The sampling design and interviewing for this study was carried out by Response Analysis Corporation of Princeton, New Jersey.The "Response Analysis National Probability Sample" relies entirely on prior specification of locations, households, and specific individuals to be interviewed.None of the selection steps are left to the discretion of the interviewer. The sequence of steps used included:
Selection of a national sample of 103 primary areas (counties or groups of counties) stratified by geographic region, type of community, and other population characteristics.
Selection of 300 interviewing locations, or secondary areas (census enumeration districts or block groups) from the national sample for use in this study.
Field counts by trained interviewers to divide interviewing locations into sample segments of 10 to 25 housing units.
Selection of specific sample segments in each interviewing location for field administration of the survey.
Random selection of the eligible person to be interviewed using a specific scheme assigned for each sample household.
More detailed information on each of these steps is available in the methodological report prepared by Response Analysis Corporation (Weisbrod, 1976). This will be supplied at cost (write to Response Analysis Corporation, Research Park, Route 206, Princeton, N.J. 08540).
Eligible families consisted of a couple who identified them