2

I've tried searching but came up short as this is probably a very generalized question, but it goes:

What is the difference between an interaction term and simply including (adjusting) for that same variable in a regression model?

Example:

We have dependent variable y, independent variable x and independent variable x1.

What is the difference between a regression that goes something like this (depending on your software):

regress y x x1

and using interaction:

regress y x#x1
Paze
  • 2,291

1 Answers1

2

These are completely different models.

The first:

regress y x x1

fits fixed effects for x and x1

The second:

regress y x#x1

fits a fixed effect for the interaction between the variables only (that is, the main effects in the first model are omitted).

It very rarely makes sense to fit the 2nd model. This has been discussed here several times before. If you want to fit an interaction you should also include the main effects.

regress y x##x1
Robert Long
  • 60,630
  • @Paze does this answer your question ? If so, please consider marking it as the accepted answer, or if not please let us know why. – Robert Long Oct 17 '20 at 13:27
  • 1
    Sorry Robert I've been neck-deep in the last semester of school plus clinical work and with the drafting of my PhD. Thank you for your answer, it definitely clarified it. – Paze Oct 24 '20 at 19:03
  • No worries, that sounds like a big workload. Feel free to connect with me on LinkedIn if you use it ! – Robert Long Oct 24 '20 at 19:04