This repository contains materials for my data analysis seminar focuses on linear regression analysis to explore questions about mediated and moderated effects. This material was first prepared for the Advanced Data Analysis class of the Master in Communication Science of the University of Vienna and can be accessed at this link: Mediation, Moderation, and Conditional Process Analysis with R.
Computer applications will focus on R statistical language, the Rstudio environment (https://www.rstudio.com),,/) and the PROCESS software by Andrew F. Hayes (http://processmacro.org)../)
The course is subdivided into the following learning units:
-
The introductory part of the course is dedicated to a brief review of the basic principles of linear regression and the setup of R and PROCESS for statistical analysis. We discuss multiple regression analysis and its principles to understand how to fit, visualize, interpret, and evaluate multiple regression models in R. We also get a first overview of mediation, moderation, and conditional process models.
-
We then focus on mediation analysis, seeing how to fit, visualize, interpret, and evaluate mediation models using PROCESS in the R environment.
-
The following learning unit is dedicated to moderation analysis and explaining how to fit, visualize, interpret, and evaluate moderation models using PROCESS in the R environment.
-
The final learning unit is an overview of conditional process analysis and aims at explaining how to fit, visualize, interpret, and evaluate conditional process models using PROCESS in the R environment.
The last part of the course is dedicated to the project work.
By the conclusion of this course, students should be proficient in:
-
Executing and comprehending the outcomes of linear regression, moderation, mediation, and conditional process models.
-
Statistically testing competing theories of mechanisms through the evaluation of indirect effects in models that encompass multiple mediators.
-
Representing and investigating interactions in regression models to accurately interpret interaction effects.
-
Calculating and inferring about conditional indirect effects to estimate the contingencies of mechanisms.
-
Utilizing the R language and PROCESS to conduct, depict, and comprehend linear regression, moderation, mediation, and conditional process models.
The handbook of the course and the primary source of most of the material here presented is:
Andrew F. Hayes. Introduction to Mediation, Moderation, and Conditional Process Analysis. A Regression-Based Approach. 2018. SECOND EDITION. THE GUILFORD PRESS, New York, London.