1
$\begingroup$

Previously, I conducted a model like the following. I have repeated measures (Time factor: pre-post) of depressive mood in two different groups (Group factor: neutral and experimental). Each participant in only one group. I checked the Coping as a moderator.

lmer(DepressiveMood ~ Coping*Time*Group + (1|ID), data = data)

Currently, I created 3 Levels among the participants based on their pre-depressive mood scores as a low, average, and high depressive mood. So, the difference can be more sizeable. However, I could not create the model. Is this following model correct? My question is individuals in three different levels of depressive mood (Level factor) are different?

lmer(DepressiveMood ~ Coping*Time*Group + (1|Level|ID), data = data)
$\endgroup$

1 Answer 1

1
$\begingroup$

If you are using lme4 (do you?) and you wanted to have a random effect for the interactions of Level and ID, you would use:

lmer(DepressiveMood ~ Coping*Time*Group + (1|Level:ID), data = data)

(note the colon : instead of the bar |).

If you want an extra random effect both for Level and ID, separately, use:

lmer(DepressiveMood ~ Coping*Time*Group + (1|Level) + (1|ID), data = data)

$\endgroup$

Your Answer

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

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