Multilevel autoregressive mediation PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Hung posted on Saturday, March 04, 2017 - 12:21 pm

I have a couple of questions concerning autoregressive mediation in a multilevel fashion.

So in my model, I have 3 variables:
a) Xt-1: X at previous time point
b) Xt: X at subsequent time point
c) Yt: Y at subsequent time point

Within-level: 30 time points per person
Between-level: 350 persons

I would like to specify a mediation model at within-level only using Xt-1 as a predictor, Xt as a mediator, and Yt as an outcome.

My questions are:
1) Does anyone know any method articles discussed this type of multilevel autoregressive mediation?

2) Since my mediator is also my predictor, but at different time points, would this bias fixed effect coefficicents, especially for the mediator to outcome path?

Thank you very much. Any information would help.

Best regards,

 Bengt O. Muthen posted on Sunday, March 05, 2017 - 2:45 pm
Mplus Version 8 will have special features related to this type of modeling.

1) There was a 2011 special issue of MBR related to this. In addition, there was a 2016 Psych Methods article by Huang & Yuan on Bayesian Dynamic Mediation.

2) This is a research topic.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message