Conference Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 558 - Semi- or Nonparametric Modeling for Data with Complex Structure
Type: Contributed
Date/Time: Thursday, August 11, 2022 : 10:30 AM to 12:20 PM
Sponsor: Survey Research Methods Section
Abstract #320918
Title: Comparison of Chi-Square Tests Computed by Multiple Statistical Programs for Survey Analyses
Author(s): Li-Yen Rebecca Hu* and Yulei He and Katherine E. Irimata
Companies: National Center for Health Statistics and National Center for Health Statistics and National Center for Health Statistics
Keywords: chi-square test; complex survey design; SAS; SUDAAN; the survey package; the samplics package
Abstract:

Chi-square tests are often employed to examine the association of categorical variables, the homogeneity of proportions between two or more samples, and the goodness-of-fit for a specified distribution. Two commonly used chi-square tests for examining data with a complex survey design are the Rao-Scott chi-square test and the Wald chi-square test. In this study, four often-used statistical packages capable of conducting either the Rao-Scott chi-square test or the Wald chi-square test and their variants accounting for a complex survey design were examined and compared for functionality and reported statistics- SAS®, SUDAAN®, the survey package for R, and the samplics package for Python. A comparison of test results from these four programs examining four variables in the 2019 National Health Interview Survey (NHIS) and the third round of the Research and Development Survey (RANDS 3) reveals that the test statistics produced by these two types of chi-square tests and their variants when examining the homogeneity of proportions in a two-way table are mostly similar with minor differences. Sample programming codes and computed test results will be discussed at the session.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2022 program