GESIS Training Courses
user_jsdisabled
Search

Scientific Coordination

Julia Leesch
Tel: +49 221 47694-169

Administrative Coordination

Claudia O'Donovan-Bellante
Tel: +49 621 1246-221

Causal Mediation Analysis

Lecturer(s):
Dr. Michael Kühhirt

Date: 23.09 - 24.09.2019 ics-file

Location: Mannheim B2, 8 / Course language: Englisch

About the lecturer - Dr. Michael Kühhirt

Course description

Investigating causal relations often leads to questions regarding the processes and mechanisms underlying a specific effect. Is an effect mediated by one or more other variables? In practice, this question is frequently assessed by analyzing changes in regression coefficients after adding the putative mediators to the model. The modern literature on causal inference demonstrates, however, that this approach yields valid conclusions regarding mediation only under specific assumptions that are rarely made explicit in applied research.
 
This course uses graphical causal models and counterfactual definitions of direct and indirect effects to make transparent the conditions under which mediation analysis yields valid conclusions. In addition to the discussion of classic approaches to mediation the course also introduces modern regression-based methods of causal mediation analysis as well as formal sensitivity analysis. Each section of the course is accompanied by practical exercises, including the estimation of direct and indirect effects in Stata and R. As the course concludes, interested participants will have the opportunity to present their own mediation analysis and discuss the implications of causal mediation analysis for their own research. Alternatively, further topics like multiple mediators, time-varying mediation, or alternative estimation approaches will be discussed.


Target group

The course is targeted at quantitative social researchers with interest in the analysis of causal effects and causal mechanisms.


Learning objectives

Participants will be able to
  •        formally define direct and direct effects and know their substantive interpretation
  •        use different regression-based methods to estimate direct and indirect effects
  •        understand and scrutinize the substantive assumptions necessary to endow the resulting estimates with a causal interpretation.    


Prerequisites

Participants should have a good working knowledge of descriptive and inductive statistics as well as linear and nonlinear regression (such as logistic regression). Experience with Stata or R will be useful.


Schedule

Recommended readings

More Information