Coverage Implications of Targeted Lists for Rare Populations

• Activity Number: 349
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Ned English*+ and Ipek Bilgen
• Companies: NORC and NORC
• Keywords: Address-based samples ; targeted lists ; hard-to-reach populations ; minority populations ; frame construction ; modeling coverage

Abstract

One challenge to using a multi-mode address-based sampling (ABS) design to target minority populations can be low eligibility. At issue is the relative cost efficiency of ABS as an alternative sampling design is inversely related to household eligibility. One potential way to improve operational efficiency would be to enrich the original address frame by using race/ethnicity-targeted lists. Such lists can be used to stratify a general population address list by indicating households likely to contain members of a targeted racial/ethnic group. This paper provides an initial investigation into the impact the use of targeted lists may have on household coverage and resulting survey data in a health survey. Using a binary logistic model, we find that the coverage of race/ethnicity targeted lists declines in dense, urban areas with large populations of renters and low priority group density. List coverage is less likely to be adequate for African American households compared to Asian or Hispanic households due to the use of surnames.