Saving lower level mediation effects PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Jack Brown posted on Sunday, December 08, 2013 - 6:18 am
Dear team,

I am running a multilevel model on daily diary data, and anticipate mediation effects at level 1 (a 1-1-1 model, using Preacher et al.'s terminology).

Daily diary estimates of mood, stressors, and coping approach used (Level 1) are nested within individuals, and trait level variables for these individuals are also captured in the data set (Level 2), such as mood regulation difficulties.

I have used Preacher, Zhang, and Zyphur's (2010) Mplus syntax to derive estimates of the mediation effects, but would like to take this a step further and use Level 2 variables to now predict size of the mediation effects.

Is it possible to save these a*b terms in a data file (such that each individual would have a potentially different mediation effect size?) and then regress the a*b term onto trait (level 2) variables?

Kind regards,
 Bengt O. Muthen posted on Sunday, December 08, 2013 - 4:32 pm
Why not make a and b both random and predict each in a 2-level model that has level-2 predictors of them. Gives more detail than looking at the product.
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