Geographic Oversampling for Race/Ethnic Minorities Based on Data from the 2010 U.S. Population Census

• Activity Number: 304
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Graham Kalton and Sixia Chen*+
• Companies: Westat and Westat
• Keywords: Rare populations ; Design effect ; Disproportionate stratification ; Sampling minority

Abstract

Based mainly on 1990 Census data, Waksberg et al. (1997) examined the effectiveness of disproportionate stratification to oversample persons in areas with greater concentrations of a minority population in order to yield a specified effective sample size for that minority in a national survey. The areas were defined using block and block group Census information. The effectiveness depends on both the degree of the minority's geographical concentration and on the relative costs of the full survey data collection to the screening costs. This paper updates the Waksberg et al. findings using 2010 Census data and compares the findings across the two censuses. In addition, we extend the application to surveys of individual census regions and Core Based Statistical Areas (CBSAs). Oversampling more than one minority population in a survey is also discussed.