Activity Number:
|
80
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 12, 2002 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Biometrics Section*
|
Abstract - #301328 |
Title:
|
Testing for Simultaneous Pairwise Marginal Independence
|
Author(s):
|
Christopher Bilder*+ and Thomas Loughin
|
Affiliation(s):
|
Oklahoma State University and Kansas State University
|
Address:
|
301 MSCS Bldg, Stillwater, Oklahoma, 74078, USA
|
Keywords:
|
categorical data ; bootstrap ; multiple response ; correlated binary data ; survey ; chi-square test
|
Abstract:
|
Questions that ask respondents to "choose all that apply" from a set of items occur frequently in surveys. Categorical variables that summarize this type of survey data are called both pick any/c variables and multiple-response categorical variables. It is often of interest to test for independence between two categorical variables. When both categorical variables can have multiple responses, traditional Pearson chi-square tests for independence should not be used because of the within-subject dependence among responses. A modified version of the Pearson statistic is proposed to perform the test, and bootstrap procedures are used to approximate its sampling distribution. First- and second-order corrections to the proposed statistic are given in order to use a chi-square distribution approximation. A Bonferroni adjustment approach is proposed to perform the test when the joint set of responses for individual subjects are unavailable.
|