Calibration of RANDS During COVID-19: A Web-Based Panel Survey

• Activity Number: 352 - Small Area Estimation, Analysis of Complex Sample Survey Data, and New Advances for Health Surveys
• Type: Contributed
• Date/Time: Thursday, August 12, 2021 : 10:00 AM to 11:50 AM
• Sponsor: Survey Research Methods Section
• Author(s): Katherine Irimata* and Yulei He and Van Parsons and Hee-Choon Shin and Guangyu Zhang
• Companies: NCHS and US Centers for Disease Control and Prevention and CDC and National Center for Health Statistics and CDC
• Keywords: calibration; National Health Interview Survey; Research and Development Survey; web surveys

Abstract

The National Center for Health Statistics (NCHS) monitors the health of the U.S. through various data systems including household surveys, establishment surveys, physical health assessments, and vital records. NCHS also conducts the Research and Development Survey (RANDS), a series of commercial panel surveys, which collects data primarily through web administration for methodological research purposes. In early 2020, the COVID-19 pandemic disrupted NCHS’ usual data systems. However, due to the flexibility of RANDS, NCHS quickly adapted the RANDS program to collect health information during the pandemic and to publicly release experimental estimates on outcomes including loss of work, telemedicine access and use, and reduced access to health care. Due to potential bias in panel surveys compared to traditional household surveys, calibration was used to adjust the RANDS panel weights prior to the data release. Raking was implemented to adjust the panel weights to marginal totals from NCHS’ 2018 National Health Interview Survey on selected demographic and health features. This talk introduces RANDS during COVID-19 and discusses considerations for calibrating panel survey weights.