This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 518
Type: Contributed
Date/Time: Wednesday, August 4, 2010 : 10:30 AM to 12:20 PM
Sponsor: ENAR
Abstract - #308701
Title: How Robust Is the Proportional Odds Model?
Author(s): Scott William Miller*+ and Sharon D. Yeatts and Yuko Y. Palesch
Companies: CDRH/FDA and Medical University of South Carolina and Medical University of South Carolina
Address: 10903 New Hampshire Ave, Silver Spring, MD, 20993-0002,
Keywords: proportional odds model ; robust ; ordinal response
Abstract:

The proportional odds model (POM), used to analyze ordinal data, can increase power relative to dichotomizing, assuming proportionality of odds ratios holds. The literature describes the POM as robust, but we have not found papers illustrating this property. We used simulation studies to explore the robustness of the POM to non-proportionality; we examined the score test, as well as the bias and power for the common odds ratio. The results suggest that the parameter estimates from the POM may be biased for non-proportional data. As previously known, the score test was sensitive to the sample size, but neither the sample size nor collapsing adjacent categories appear to correct the bias. We conclude that the PO assumption may be sensitive to the particular dataset, and recommend that clinical trialists utilizing this method take steps to investigate the fit of the data to the POM.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2010 program




2010 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.