Sampling, data collection, and weighting procedures for address-based sample surveys

Abstract

Address-based sampling(ABS) involves drawing probability samples from address frames; in the United States, ABS frames are usually based, in part, on U.S. Postal Service mail delivery databases. Its use has grown rapidly over the past 10 years to the point where it is considered a viable alternative to random-digit-dial dual frame telephone sample designs. Although telephone surveys are well established in the U.S., they have seen declining response rates and the need to sample from both the landline telephone number sampling frame and the cell phone sampling frame. ABS typically involves sampling from a single address frame and opens up the possibility of using various contact and data collection modes including mail, Internet, telephone, and in-person. Mixed modes are often employed in ABS with anywhere from one to multiple phases of contact and/or data collection. The American Association for Public Opinion Research (AAPOR) Council, operating under the auspices of the AAPOR Standards Committee, convened a task force to prepare a comprehensive report on Address-based Sampling. This paper is drawn from the AAPOR report and focuses on sampling, approaches to data collection, and weighting procedures. We also address an important and somewhat challenging aspect of ABS designs for population surveys, the random selection of an adult from the sample household