Pre post data with pre post mediator PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
 Letoynia J Coombs posted on Friday, November 09, 2012 - 1:42 pm
I have patients nested in clinics and repeated measures (only 2) from patients on a outcome and on a mediator. The patients (not the clinics) are also assigned to a control group and an intervention group.

I am considering mplus because of the mediator which can vary within a patient and between patients. Maybe within and between clinics too, I'm not sure about that part.

Can I approach this problem with mplus given that there are only 2 time periods?
If so, should I be considering the long or the wide data format? Thank you.
 Linda K. Muthen posted on Friday, November 09, 2012 - 8:21 pm
You could do an intercept only growth model with two timepoints. I would put the data in the wide format. You could do a two-level analysis with clinics the highest level if you have at least 20 and preferable a minimum of 30-50 clinics.
 ljc posted on Tuesday, August 13, 2013 - 12:58 am

Would I do something like

iy| y1@-1 y2@1

to force the intercept be the difference?

 Linda K. Muthen posted on Tuesday, August 13, 2013 - 3:10 pm
The growth model language would not work the way you want. You should instead say:

f BY y1@-1 y2@1;
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