Multi-Objective Sample Allocation in a US Veteran Health Survey

• Activity Number: 285 - New Advances in Sample Design and Adjusting for Survey Nonresponse
• Type: Contributed
• Date/Time: Wednesday, August 11, 2021 : 1:30 PM to 3:20 PM
• Sponsor: Survey Research Methods Section
• Author(s): Michael D Larsen*
• Companies: St. Michael's College
• Keywords: Oversampling; Poststratification; Response rates; Sample Design; Stratification; Two-phase sampling

Abstract

The Department of Veterans Affairs (VA) operates the largest health care system in the United States. The Survey of Veteran Enrollees’ Health and Use of Health Care provides annual in-depth information on enrollees’ attributes, health status, health care options and usage, and perceptions of VA health care. The population of ~8.7 million veterans is stratified by multiple factors. Sample size targets are given for ~95 health care markets and ~54 subgroups defined by health care priority groups and Veterans Integrated Services Networks (VISNs). The two sets of targets are not nested. Veterans with enrollment dates on or before March 31, 1999, and veterans in the sampling frame for the previous survey but not selected are oversampled. Response rates vary by stratum. Stratum sizes vary from a few individuals to thousands. Sample is selected in two integrated waves, with the second wave adapted to response rates part-way through the first wave. This talk will describe sampling design principles and implementations used to select the samples in 2018-2021.