Analyzing the Need for an Oversample of Low Income Beneficiaries in the Medicare Current Beneficiary Survey

• Activity Number: 76 - Sample Design
• Type: Contributed
• Date/Time: Sunday, July 30, 2017 : 4:00 PM to 5:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Ying Li* and Whitney Murphy and Kirk Wolter and Debra Reed-Gillette and Cheryl L. Sharpless
• Companies: NORC at the University of Chicago and NORC at the University of Chicago and NORC at the University of Chicago and Centers for Medicare & Medicaid Services and CMS
• Keywords: Medicare ; survey ; sampling frame ; low-income ; oversample design

Abstract

The Medicare Current Beneficiary Survey (MCBS) is a continuous, multipurpose survey of a nationally representative sample of the Medicare population, conducted by the Centers for Medicare & Medicaid Services through a contract with NORC at the University of Chicago. The MCBS evaluated the potential benefit of oversampling low-income beneficiaries to allow for improved estimation of parameters of health disparities experienced by this population. This paper demonstrates the methods used to conduct the evaluation, which included examining alternative approaches for oversampling low-income beneficiaries, such as using U.S. census data on income at the tract level, data from consumer databases on income at the household level, and geographic variation indicators of low-income status at the beneficiary level, and examining the effects of oversampling low-income beneficiaries on the standard errors of key statistics. Findings from the evaluation revealed that there were already sufficient numbers of low income beneficiaries in the MCBS for analytic purposes.