1
$\begingroup$

I am learning mixed models and while reading different materials I still have a couple of questions that I could not find answers.

  1. If I have 2 or more fixed effects and if the interaction is not significant can I just drop the interaction term and re run the model?

  2. If the interaction is significant in LMMs should I follow the same procedure as Anova for multiple comparisons? Similarly for non-significant interaction? What is the method for this kind of situation? I saw in different webs people say in LMM do not explain p values? I can’t understand that?

$\endgroup$
1
$\begingroup$

The rationale for various alternatives is not really different in LMM than in ANOVA. I would advise against doing anything (in either LMM or ANOVA) based just on p values. But other people disagree with me.

I'm not sure what "don't explain p values" means. They may be referring to the random effects rather than the fixed ones, but without context, it is hard to tell.

$\endgroup$

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.