Median and quantile tests under complex survey design using SAS and R

Yi Pan; Samuel P Caudill; Ruosha Li; Kathleen L Caldwell

doi:10.1016/j.cmpb.2014.07.007

Median and quantile tests under complex survey design using SAS and R

Comput Methods Programs Biomed. 2014 Nov;117(2):292-7. doi: 10.1016/j.cmpb.2014.07.007. Epub 2014 Jul 28.

Authors

Yi Pan¹, Samuel P Caudill², Ruosha Li³, Kathleen L Caldwell²

Affiliations

¹ Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA, United States. Electronic address: jnu5@cdc.gov.
² Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA, United States.
³ Department of Biostatistics, University of Pittsburgh, Pittsburgh, PA 15261, United States.

Abstract

Techniques for conducting hypothesis testing on the median and other quantiles of two or more subgroups under complex survey design are limited. In this paper, we introduce programs in both SAS and R to perform such a test. A detailed illustration of the computations, macro variable definitions, input and output for the SAS and R programs are also included in the text. Urinary iodine data from National Health and Nutrition Examination Survey (NHANES) are used as examples for comparing medians between females and males as well as comparing the 75th percentiles among three salt consumption groups.

Keywords: Complex survey; Median; Quantile; R; SAS.

Published by Elsevier Ireland Ltd.

MeSH terms

Algorithms*
Computer Simulation
Data Interpretation, Statistical*
Female
Humans
Male
Models, Statistical*
Nutrition Surveys / methods*
Programming Languages*
Software*

Grants and funding

CC999999/Intramural CDC HHS/United States