Title
|
Statistical Methods for the Analysis of Ordered Categorical Data
|
Date / Time / Room
|
Sponsor
|
Type
|
08/12/2002
8:00 AM -
12:00 PM
Room: S-Conference Room D
|
ASA, Biometrics Section*
|
Other
|
Organizer:
|
n/a
|
Chair:
|
n/a
|
CE Presenter
|
Charles S. Davis - Elan Pharmaceuticals, Inc.
Charles S. Davis - Elan Pharmaceuticals, Inc.
|
Description
Ordered categorical response variable occur frequently in many types of research. The purpose of this course is to introduce, discuss, and provide examples of several approaches for analyzing data in which the response variable is categorical with more than two ordered categories. One approach is based on the use of randomization model test statistics for two-way and stratified two-way contingency tables. In addition, regression methods for modeling a categorical response as a function of one or more categorical and/or continuous covariates are useful in many areas of applications. Thus, extensions of the binary logistic regression model to ordered categorical responses will be also be described. Multiple logit, cumulative logit, mean score, and proportional odds regression models will be discussed. The course will also introduce and discuss extensions of these types of models to the repeated measurements setting (multiple measurements per subjects) will also be described. Examples using SAS software will be presented. The course is directed to statisticians who wish to learn (and/or to review) several approaches to the analysis of ordered categorical outcome variables; participants should have some knowledge of basic methods for the analysis of categorical data.
|
|