GESIS Training Courses

Scientific Coordination

André Ernst
Tel: +49 221 4703736

Administrative Coordination

Janina Götsche

Causal Mediation Analysis

Online via Zoom
General Topics:
Course Level:
Software used:
5 days
Students: 200 €
Academics: 300 €
Commercial: 600 €
Additional links
Lecturer(s): Felix Thoemmes

About the lecturer - Felix Thoemmes

Course description

Mediation analysis has been used by social scientists for the last 50 years to explain intermediate mechanisms between an assumed cause and effect. During these years many advances in statistical mediation analyses were made, including the use of multiple mediators, models for limited dependent variables, latent variable modeling, improved standard errors, and the combination of mediation and moderation analysis. However, only very recently were the causal foundations and underlying assumptions of mediation analysis clarified. These more recent advances used potential outcomes notation and graphical causal models to illuminate the types of causal effects that can be estimated - and more importantly, which assumptions are needed to recover an unbiased causal effect. This course will briefly review the traditional approaches to mediation analysis, then review fundamental topics for causal inference, and then discuss the novel methods that fall under the rubric of “causal mediation analysis.” The causal mediation methods put the assumptions of the analysis front and center, and because the causal assumptions are often untestable, tools like sensitivity analysis become important.
The course is mostly lecture-based, but will also provide numerous opportunities to practice the studied concepts using applied data examples in R.

Target group

The proposed target group of this course are researchers in the social sciences at all career stages who want to learn more about causal mediation analysis.

Learning objectives

By the end of the course participants will:
  • understand the definition of the causal mediation effects that have been proposed in the literature,
  • should be able to identify the underlying assumptions needed for each analysis,
  • should be able to perform these analyses in R,
  • and finally should be able to confidently interpret the results of said analyses.


  • Solid working knowledge of the R programming language, with some familiarity of the tidyverse programming approach
  • Solid understanding of regression analysis
  • Some understanding of causal effects definitions using potential outcomes and DAGs
    Organizational structure of the course
    The course will be delivered via zoom, but all instructions will be live (no pre-recordings). The majority of the course is lecture-based, but we will have a numerous opportunities to practice what we learned on applied data examples. During the practical applications, participants will be given (simulated or real) data to work with in R. During these exercises the instructor will be available to assist.
    Software requirements
    We will be exclusively using the R programming language. It is advisable that participants have RStudio (or VSCode with R add-ins) already installed. Additionally, the tidyverse suite of packages, and the “mediation” package are required, but can be installed during the course.
    Monday, 11.12.
    14:30 - 14:45Introduction to the workshop / logistics
    14:45 - 15:45Introduction and review of classic statistical mediation analysis
    15:45 - 16:00Short break
    16:00 - 16:45Introduction to potential outcomes notation
    16:45 - 17:00Discussion and questions
    Tuesday, 12.12..
    14:30 - 15:45Introduction to causal mediation analysis
    15:45 - 16:00Short break
    16:00 - 16:45Practice exercise in R
    16:45 - 17:00Discussion and questions
    Wednesday, 13.12.
    14:30 - 15:15Assumptions underlying causal mediation analysis
    15:15 - 15:45Practice exercise in R
    15:45 - 16:00Short break
    16:00 - 16:45Continuation of assumptions underlying causal mediation analysis
    16:45 - 17:00Discussion and questions
    Thursday, 14.12.
    14:30 - 14:45Sensitivity analysis for causal mediation effects
    14:45 - 15:45Practice exercise in R
    15:45 - 16:00Short break
    16:00 - 16:30Overview of some alternative estimators
    16:30 - 17:00Discussion and questions
    Friday, 15.12.
    14:30 - 14:45Design-based inference
    14:45 - 15:45Bring your own data and problems
    15:45 - 16:00Short break
    16:00 - 16:30Future directions and unexplored issues
    16:30 - 17:00Discussion and questions

    Recommended readings