|
Activity Number:
|
162
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biometrics Section
|
| Abstract - #304523 |
|
Title:
|
Sample Size Calculation with Weighted Sign Tests for Paired Comparisons of Clustered Data
|
|
Author(s):
|
Fan Hu*+ and Chul W. Ahn and William Schucany
|
|
Companies:
|
Southern Methodist University and The University of Texas Southwestern Medical Center at Dallas and Southern Methodist University
|
|
Address:
|
5555 Amesbury Dr, Department of Statistical Science, Dallas, TX, 75206,
|
|
Keywords:
|
Sample size ; Cluster size distribution ; Power
|
|
Abstract:
|
Clustered data often occurs in health studies such as ophthalmologic or periodontal studies. In this poster we propose a sample size calculation of the sign test for paired comparisons of clustered data. Sample size estimation methods of assigning equal weights to units, assigning equal weights to clusters, and using optimal weights are derived based on the weighted sign test provided by Larocque et al. (2007) which incorporates the intraclass correlation coefficient and the cluster size distribution. We conducted a simulation experiment with variation in cluster size to evaluate the performance of the proposed sample size formulas in terms of empirical type I errors and powers.
|
