Further Developments in a Hierarchical Bayes Approach to Small-Area Estimation of Health Insurance Coverage: State-Level Estimates for Demographic Groups

• Activity Number: 186
• Type: Contributed
• Date/Time: Monday, August 4, 2008 : 2:00 PM to 3:50 PM
• Sponsor: Section on Survey Research Methods
• Author(s): Steven Riesz*+ and Mark Bauder and Donald M. Luery
• Companies: U.S. Census Bureau and U.S. Census Bureau and U.S. Census Bureau
• Keywords: uninsured ; SCHIP ; CDC ; Fay-Herriot model ; multivariate model ; administrative records

Abstract

Fisher et al. (2006) developed a hierarchical Bayes model to estimate the number of people without health insurance within demographic groups for states. The Centers for Disease Control and Prevention are interested in estimates of women without health insurance by demographic groups in families that earn less than 200% of the poverty line. Our approach jointly models survey estimates from the Annual Social and Economic Supplement to the Current Population Survey and estimates from tax, census, food stamp, and Medicaid data using a multivariate, hierarchical approach. We have made important enhancements to the models by including additional administrative data, further elaborating the expectation and variance models for both the survey estimates and the administrative data, and developing an approach to reflect the variation due to raking to survey estimates in the variance estimates.