What statistical test do I need to use for a moderator variable?

#1
For my study, I'm trying to test whether a variable acts as a moderator between the relationship of two other variables. I read that I should use moderational multiple regression with an interaction effect. However, I heard you can only use it if the moderator variable has 2 categories/ is dichotomous. The way my moderator variable is measured it has 3 different categories. What can I do when my moderator has 3 categories? Is there a test for this? I can't find one (and I'm only a Bachelor student, so not extremely complicated stats would be good). Thanks!

(My possible moderator variable is coping. The scale measures 3 styles/categories of coping. I could maybe make 2 categories, "adaptive" and "maladaptive", out of it but it would be artificial and info could get lost)
 

Karabiner

TS Contributor
#2
You can use a moderator with 3 levels .

In linear regression, it would be represented by 2 dummy variables (plus the respective interaction terms with the predictor).

Alternatively, you could perform an analysis of covariance, including the predictor, the moderator, and their interaction.

With kind regards

Karabiner