|
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 - #307819 |
|
Title:
|
Simple Practitioner-Oriented Hypothesis Tests for Various Types of Multiple Response Survey Data
|
|
Author(s):
|
D. Roland Thomas*+ and Yves J. Decady
|
|
Companies:
|
Carleton University and Statistics Canada
|
|
Address:
|
Sprott School of Business, Ottawa, PA, K1S 2T9, Canada
|
|
Keywords:
|
Multiple response ; pick any ; Rao-Scott tests
|
|
Abstract:
|
The focus of the paper will be tests of association featuring various types of multiple response data, including "pick any out of r" data in which respondents are invited to select all items that apply to them from a list of r items, "pick k out of r" data in which respondents are limited to selecting k items, along with variants for which respondents are asked to "order r" and "order k out of r" items. Also discussed will be tests where the original multiple response data have been aggregated into a smaller number of categories. Rao-Scott adjusted chi-squared tests will be used, with a view to developing simple tests that can be implemented by practitioners using standard software.
|
