Dear fellow researchers,
I wish to do a moderated serial mediation in PROCESS. Hayes book (version 2013) shows a short overview of different models. However, moderated serial mediation models are not in it.
In de attachement you can view moderated serial mediation model 85. But I wish to do another model instead (see other figure). The difference between these 2 models are the paths that are moderated.
Could anyone help me find which model this is in PROCESS? Thank you!
Hello Sofie,
Is there some compelling reason for having to do your analysis with Hayes' PROCESS macro? Your second model is easily set up and analyzed as a path model in any sem program, or as a set of regression equations in just about any statistical package (albeit without the usual panoply of model-data fit indices).
The only task ahead of time would be to create the various interaction terms involving the proposed moderator variable, W.
Good luck with your work.
Hello Sofie,
if you want to stick with PROCESS (not necessary, as David Morse mentioned, but maybe "nicer") I suggest to load the current version 4.1 AND to use the current 3rd edition of the book, NOT the 1st edition, since some changes haven been made!
Not all possible models are directly available through the GUI or "model" command in the syntax. But you are able to create any models you like in the given restriction of PROCESS (e.g. not more than 2 moderators, etc). To build your own model you need to use the syntax instead of the GUI (if you are not already doing it) and to define different matrices, which tell PROCESS which mediator and/or moderator is associated with which variable (BMATRIX, WMATRIX, ZMATRIX). You can find a detailed desctiption on Hayes book, Appendix B"Constructing and Customizing Models in PROCESS".
Thank you both for your help! Using process is indeed a matter of convenience. It has been a while since I have used SEM analysis using Mplus. Maybe this could be an alternative indeed. First I will try to get a hold of the newest version of the book by Hayes. Thank you a lot! Kind regards.
- Badges
- Science topic
- Similar topics
- Correction