A Comparison of Median Testing Methods for Complex Survey Data Using Various Statistical Packages

• Activity Number: 124
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Section for Statistical Programmers and Analysts
• Author(s): Christine Davies and Lauren Klein Warren*+ and Joshua Warren
• Companies: RTI International and RTI International and The University of North Carolina at Chapel Hill
• Keywords: survey data ; nonparametric ; SUDAAN

Abstract

When dealing with highly skewed data, medians are often more informative for comparing the centers of the distributions of two populations than the means. The SUDAAN® software package is capable of handling the sophisticated covariance structures characteristic of a complex survey sample design when estimating a median and its standard error, but does not have an option for comparing population medians from dependent samples. This paper will explore several methods for comparing population medians while accounting for possible correlation within a complex sample. A test inversion method introduced by Binder and Patak (1994) will be compared with alternative methods. Simulation study results as well as a real data example using the NSDUH Public Use Files will be discussed. Capabilities and limitations of statistical software packages, including SUDAAN, will be outlined and a recommendation for the optimum method for comparing medians will be included.