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!