Small Area Estimation for the Tobacco-Use Supplement to the Current Population Survey

• Activity Number: 571
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Benmei Liu*+ and Aaron J. Giliary
• Companies: National Cancer Institute and U.S. Census Bureau
• Keywords: credible intervals ; model-based county estimates ; predictor variable selection ; smoking prevalence ; tobacco related measures

Abstract

The Tobacco Use Supplement to the Current Population Survey (TUS-CPS),conducted by the Census Bureau and sponsored by the National Cancer Institute(NCI), is a key source of national and state-level data on smoking and other tobacco use in the U.S. household population. However, policy makers and cancer researchers often need county-level data to evaluate tobacco control programs, and the TUS-CPS does not have a large enough sample at the county level to support estimates with adequate precision. In such case, estimates derived through small area estimation (SAE) techniques may be preferable. Through collaboration between the Census Bureau and NCI, we propose model-based county-level estimates for several different smoking-related variables for all U.S. counties using a Bayesian framework through a Markov Chain Monte Carlo (MCMC) simulation. We applied extensive model selection and diagnosis techniques to choose the best set of auxiliary variables from a pool and the best fit models from a few candidate models. Our small area models generate a new set of estimates with improved precision over the survey-based estimates. This paper describes the methodology used and also demonstrates the accuracy of the model through data exhibits.