multilevel mediation with mixed effects model


For my thesis I need to model a multilevel mediation model (1-1-1) with mixed effects. lavaan does not model mixed effects (only random intercepts), this is why I am trying the Bauer et al. (2006)(APA PsycNet) approach, which melts the two variables M and Y, using lmer and mediation function after.

In my case they are not in the same scale (interval vs. continious).
Is there any solution to solve this, without losing information?

Or another approach multilevel mediation model with mixed effects can be implemented in R?

Thanks for your help!

The vignettte for the {robmed} package has a good discussion of the packages available for mediation analysis by way of putting its own approach of robust mediation in context (to account for the effect of outliers and departures from normality). I suggest you use a directed acyclic graph approach, as shown some of the figures, to layout your model clearly. The {dagR} package assists that.

This topic was automatically closed 42 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.