Hi,
I am working on my master thesis and want to do a specific analysis. I want to see if in a diary study variable 'A' correlates with variable 'C' through variable 'B' (mediation). Furthermore I want to see if variable 'D' influences how variable 'A' and 'B' correlate (moderation on mediation).
Variable A, B, C, and D are measured daily and are thus clustered.
I saw that I can do this analysis with MlMed for SPSS, but I'm struggling with my next variable:
I want to see if the effect of 'D' is different for people who score differently on variabele 'E' (so a between person variable).
Does anyone know a program I can use (multilevel, preferably for SPSS, but if thats not possible I'm willing to learn something new)?
I added my model as a picture to clarify things:
Love the model, SPSS won't work here as it can't deal with this kind of full "moderated moderated" mediation (it won't do the cross-level 2-way moderation involving the E/between person variable)
A mixed model would do the job, coding this in R using the lme package is super easy (literally one line). Are these latent variables (multiple indicators per construct)? If so, maybe consider SEM as well
- Badges
- Science topic
- Similar topics
- Design
- Architecture