|
Activity Number:
|
178
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, July 30, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #307785 |
|
Title:
|
Estimation and Testing for Association with Multiple-Response Categorical Variables from Complex Surveys
|
|
Author(s):
|
Christopher R. Bilder*+ and Thomas M. Loughin
|
|
Companies:
|
University of Nebraska-Lincoln and Simon Fraser University
|
|
Address:
|
Department of Statistics, Lincoln, NE, 68583-0963,
|
|
Keywords:
|
choose all that apply ; correlated binary data ; generalized loglinear model ; marginal model ; pick any ; simultaneous pairwise marginal independence
|
|
Abstract:
|
Many survey questions invite respondents to "choose all that apply" from a series of items. Variables that summarize the resulting data are often referred to as multiple-response categorical variables (MRCVs). Analyzing associations among MRCVs is a challenge because each one is a vector of correlated binary responses, and associations between these vectors can take on many possible forms. This talk will show how one can model these MRCVs in situations where the data has arisen from a complex survey sampling design. Generalized loglinear models are proposed for analyzing the associations between items of different MRCVs. Model comparison test statistics along with their derived asymptotic distributions are presented in order to choose a good fitting model. Model-based estimates of the odds ratios and their corresponding standard errors are provided.
|
