A Procedure for Evaluating and Comparing Small-Area Variability of Binary Outcomes

• Activity Number: 37
• Type: Contributed
• Date/Time: Sunday, August 4, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Pavlina Rumcheva*+ and Donald Malec and Peter Meyer and Nathaniel Schenker
• Companies: National Center for Health Statistics and National Center for Health Statistics and National Center for Health Statistics and National Center for Health Statistics
• Keywords: National Health Interview Survey ; binary outcomes ; small area ; variability ; Bayesian analysis

Abstract

A screening procedure for evaluating and ranking the small-area variability of binary outcomes from the National Health Interview Survey is presented. It is created to facilitate the process of efficiently allocating resources for the development and estimation of small-area statistical models. The procedure is developed within a Bayesian framework and is examined under different sampling and prior distribution assumptions. Posterior means and standard deviations for the small-area variability of several health outcomes are numerically computed under different model assumptions, and posterior probabilities compare the outcome variability rankings. The performance of the screening procedure is evaluated with computer simulations.