|
Activity Number:
|
335
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
ENAR
|
| Abstract - #310169 |
|
Title:
|
Test Size for Two-Stage and One-Stage Analyses of Cluster Samples of Unequal Size
|
|
Author(s):
|
Jacqueline Johnson*+ and Diane Catellier and Keith E. Muller
|
|
Companies:
|
The University of North Carolina at Chapel Hill and The University of North Carolina at Chapel Hill and University of Florida
|
|
Address:
|
700 Bolinwood Drive Apt 24G, Chapel Hill, NC, 27514,
|
|
Keywords:
|
clustered data ; group randomized trials ; gaussian linear models
|
|
Abstract:
|
Important public health research often requires the use of community based studies due to logistical, ethical and cost constraints. Such designs require special methods of analysis. Gaussian clustered data are often analyzed with either a two-stage analysis of cluster means or a mixed effects linear model on individual level data. For data with a large number of clusters and large number of observations within each cluster, both techniques provide unbiased hypothesis tests. In small samples with unbalanced data, however, even moderate imbalance in cluster size across treatment groups can lead to test size bias. We present the results of a simulation study to describe bias in test size in the two-stage analysis of cluster means and one-stage mixed model for unbalanced clustered data.
|
