Abstract Details
Activity Number:
|
692
|
Type:
|
Contributed
|
Date/Time:
|
Thursday, August 8, 2013 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistics in Epidemiology
|
Abstract - #308639 |
Title:
|
Design Effects in Three-Level Studies
|
Author(s):
|
Tina Cunningham*+ and Robert E. Johnson
|
Companies:
|
Eastern Virginia Medical School and Vanderbilt University Department of Biostatistics
|
Keywords:
|
Design Effect ;
ICC ;
Cluster
|
Abstract:
|
Experiments with three-level nested levels are frequently seen in different areas of research, such as health service and Epidemiology. For an example, consider a study in which new health interventions are randomized across health centers, while providers within each center will apply the intervention to individual subjects. Design and analysis of three-level study are more complicated than simple single level or two-level models since the multilevel nesting effect introduces more than one intracluster correlation (ICC). One way to account for this intracluster dependence is to factor in the design effect(variance inflation factor).We propose the formula for the design effect in three-level models using generalized least squares solution. The derived formula is a function of the two ICCs and the sample sizes of each level. We explore the changes that could occur in the design effect in different situations. To narrow the focus of the discussion, we limit our attention to continuous outcomes with identity link function and compound symmetric error structure. Examples will be given to demonstrate practical implication of the formulas.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2013 program
|
2013 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.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.