Abstract Details
Activity Number:
|
10
|
Type:
|
Invited
|
Date/Time:
|
Sunday, August 4, 2013 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Committee on Women in Statistics
|
Abstract - #307175 |
Title:
|
Competing Versions of Ignorability Assumptions in Causal Mediation Analysis
|
Author(s):
|
Booil Jo*+ and Elizabeth Stuart
|
Companies:
|
Stanford University and Johns Hopkins Bloomberg School of Public Health
|
Keywords:
|
causal inference ;
mediation ;
sequential ignorability ;
principal ignorability
|
Abstract:
|
Based on the assumption of ignorability, propensity score and related methods often provide elegant solutions to highly complex causal inference problems. The drawback is that it is often hard to come up with practical ways of evaluating the validity of the results. In this presentation, we discuss the possibility of considering two alternative versions of ignorability in the context of causal mediation analysis. Sequential ignorability (Imai et al., 2010) has been developed to support causal interpretation directly focusing on the potential mediator (M). Principal ignorability (Jo & Stuart, 2009) instead focuses on the potential mediator type (C) characterized by the relationship between the treatment assignment and potential mediator response (M). Although both assumptions commonly involve M, the two may lead to different conclusions about the relationship between treatment, the mediator, and the outcome. We show that their joint use may provide critical insights into inferences. A job search intervention study, where self efficacy was expected to mediate the intervention effect on mental health and employment outcomes, will be used as a key example.
|
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.