298 – SPEED: Innovations in Survey Sampling Designs: Administrative Data, Record Linkage, Non-Probability Samples, and More
When to Use Commercial Data for Improved Efficiency
Ned English
NORC At the University of Chicago
Colm O'Muircheartaigh
University of Chicago Harris School of Public Policy
Holly Hagerty
NORC at the University of Chicago
Carina Hoyer
NORC at the University of Chicago
Address-Based Sampling [ABS] designs based on extracts of the United States Postal Service computerized delivery sequence file allow for the enhancement with information from supplementary lists designed to identify households belonging to specific demographic groups, including those that could be considered rare or hard-to-reach. We know from previous research, however, that such targeted lists can have reduced coverage for certain subgroups. Our paper uses area information from the American Community Survey and other sources to predict when a specific supplementary source is most appropriate for a specific target. We also describe the characteristics of survey respondents who are well-identified vs. poorly-identified from a specific source or combination of sources. Our research will be useful to survey practitioners interested in the tradeoffs of the improved efficiency of targeted lists. Validation information is obtained from the fieldwork of the AmeriSpeak® Panel NORC's, multi-mode ABS panel designed to support NORC's mission to deliver reliable data to guide critical programmatic, business, and policy decisions.