Implementation of Adaptive Design on the Medicare Current Beneficiary Survey

• Activity Number: 76 - Paradata for Adaptive Survey Designs and Other Applications
• Type: Contributed
• Date/Time: Sunday, July 29, 2018 : 4:00 PM to 5:50 PM
• Sponsor: Government Statistics Section
• Author(s): Christopher Ward* and Felicia LeClere and Kari Carris and Stephen Cohen and Dean Resnick and Micah Sjoblom and Jennifer Vanicek and Ying Li
• Companies: NORC at the University of Chicago and NORC at the University of Chicago and NORC at the University of Chicago and NORC at the University of Chicago and NORC and NORC at the University of Chicago and NORC at the University of Chicago and NORC at the University of Chicago
• Keywords: Medicare; r-indicators; bootstrapping; variance estimation; paradata; responsive design

Abstract

In light of the steady decline in response rates across a range of surveys, interest has grown among survey researchers to establish methods of measuring representativeness of the collected sample. On the Medicare Current Beneficiary Survey (MCBS), we have implemented an adaptive survey design to ensure that the survey respondents are representative of the sample population. The MCBS is a continuous, multipurpose survey of a nationally representative sample of the Medicare population, conducted by the Centers for Medicare & Medicaid Services (CMS) through a contract with NORC. In this presentation, we review key measures used in adaptive design and describe our findings based on data collected during Fall 2017. We also compare this period to the same period in 2016, provide displays that help analysts interpret R-indicators, and discuss interventions proposed to respond to these measures. We discuss the use of R-indicators that measure both the representativeness to the sample population as a whole and also R-indicators for specific variables that are important. We propose practical data collection interventions as a means of implementing real-time adaptive design.