1
$\begingroup$

I would like to make sure whether I’m interpreting the results of the lmer model I generated in the right way. The model is:

lmer(distress ~ cond*group + (1|subject), data = data)

Condition has two levels (S1 and S2, S1 is the reference), group has 2 levels (A and B, B is the reference)

  1. Would the coefficient of the three way interaction condS2:groupA indicate that the difference of S2 vs S1 is larger in in group A vs group B?

  2. Maybe this is a stupid question but, can this also not be tested with a simple regression with a predictor group and the difference scores (S2-S1) as the DV? Would it be inappropriate because of the disregard of the random effects?

  3. If I want to test whether conditions differ, within group A and within group B, do I run two different lmers for two groups seperately?

Thanks in advance!

Improve this question
$\endgroup$
2
  • 1
    $\begingroup$ check my answer to a similar question here $\endgroup$
    – utobi
    Commented Mar 20, 2023 at 20:16
  • $\begingroup$ thank you for your very fast feedback, utobi. I was wondering, if I wanted to test whether the difference of S1 vs S2 is larger/smaller in Group A vs B, should I then create a different model here or would your interprations under the other post mean essentially the same thing? $\endgroup$
    – gfndngo
    Commented Mar 20, 2023 at 21:18

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.