0
$\begingroup$

I have a question regarding testing interaction terms in regression. When we have significant interaction term, do we still interpret main effects?

I have made two models, one with interaction term included, and the other variable changed values. Should I still interpret them or just the interaction term?

Thank you in advance,

$\endgroup$
0
$\begingroup$

This must depend strongly on the nature of the interaction. Suppose that we have two factors A and B with levels 1 and 2. Now suppose the means are A1B1 0, A1 B2 1, A2B1 1, A2B2 0. How can you make any meaningful statement about the effect of either A or B in isolation here? It is only when you know the interaction that the picture becomes clear.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.