Abstract Details
Activity Number:
|
447
|
Type:
|
Topic Contributed
|
Date/Time:
|
Wednesday, August 6, 2014 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Biometrics Section
|
Abstract #312298
|
View Presentation
|
Title:
|
Mediation in Cluster Randomized Trials: Insights from Modern Causal Inference vs. Traditional Approaches
|
Author(s):
|
Elizabeth Turner*+ and Matthew Jukes and Elizabeth Adelman and Magaret Dubeck
|
Companies:
|
Duke University and Room To Read and Harvard Graduate School of Education and University of Virginia
|
Keywords:
|
Cluster randomized trial ;
Mediation ;
Education ;
Complex intervention ;
Causal inference
|
Abstract:
|
Cluster randomized trials are commonly used to evaluate the effectiveness of interventions with multiple components, which are often referred to as 'complex' interventions. When effectiveness of such a complex intervention is demonstrated, it is important to determine which components contribute most by assessing changes in cluster- and individual-level mediators on the causal pathway from the intervention to the outcome. Yet a number of challenges arise with such mediation analyses, including likely confounding of the mediators and outcome, the challenges of analyzing multiple mediator variables and complexity in the data structure due to the clustered nature of the study design. Using data from an educational intervention trial in Kenya, we present insights in to these challenges and solutions from modern causal inference methods, which we contrast with traditional mediation approaches (including the Baron and Kenny approach). These methods will be elucidated in light of recent analytic advances in the broader field of analysis of cluster randomized trials.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development 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.