Tutorial on main effects vs. interaction in a regression model, 2020-12-18
On December 18, 2020, I held a live Zoom session explaining the difference between a linear regression model with two main effects and the same model with an interaction, using a silly dog-related hypothetical data example. I split the recording of the session into the eight videos listed on this page and in a Youtube playlist.
These videos answer the following questions: What is the difference between simple linear regression and multiple regression? What happens when you add a covariate to a regression model? What does it mean to "control for" or "adjust for" a variable? What is the difference between a model with two main effects and a model with a two-way interaction?
This is one continuous session broken into eight videos, so later videos may not make sense on their own. I have provided some keywords in the titles in case you want to skip to a particular part anyway.
Faces are blurred in some of the videos at the request of students who participated in the workshop.
The R code used in these videos is available here: Regression Main Effects vs. Interaction [text].
List of Videos
The videos linked below are available in a single Youtube playlist.
Part 2 - More questions about random seed
Part 3 - Getting to know the data
Part 4 - Unadjusted main effect (simple linear regression)
Part 5 - Two main effects (adjusting for covariate)
Part 7 - Graphical illustration of all three models