Interaction in path analysis PreviousNext
Mplus Discussion > Categorical Data Modeling >
 Kelsey Antle posted on Monday, August 25, 2014 - 2:15 pm

I would like to check and see that I am writing the code correctly for an interaction relationship for an analysis with all observed variables. I apologize if this is elementary, I have been using stata as my primary software and am new to Mplus.

My model is a bit complicated. I have a dichotomous exogenous variable (x1), which I hypothesize will predict a set of mediators (m1, m2). I then hypothesize these two mediators will predict another set of mediators (n1, n2, n3), which will in turn predict a final outcome (y1). Finally, I hypothesize that six moderating variables (z1 z2 z3 z4 z5 z6) influence the relationship between my exogenous variable (x1) and my first set of mediators (m1 m2). I have created interaction terms for each moderator (xz1 xz2 xz3 xz4 xz5 xz6) in Stata. Is the following analysis correct?

y1 on m1 m2 n1 n2 n3 x1;
n1 n2 n3 on m1 m2 x1;
m1 m2 on x1;
y1 on z1 z2 z3 z4 z5 z6 xz1 xz2 xz3 xz4 xz5 xz6;

Thank you so much for your help.


PS: x1, m1, z4, and z5 are dichotomous; m2, m3, n2, n3, y1, z1, z2, and z3 are continuous (i.e., scales measured in whole numbers); n1 is ratio.
 Linda K. Muthen posted on Monday, August 25, 2014 - 3:37 pm
The best way to see if your MODEL command is correct is to run the analysis and check that you obtain the free parameters that you expect. I assume all endogenous variables are continuous.
 Kelsey Antle posted on Wednesday, August 27, 2014 - 7:57 am
Actually, one of the endogenous variables is ratio, do you think that might be the issue?
 Bengt O. Muthen posted on Wednesday, August 27, 2014 - 4:48 pm
That should not be a problem in itself.
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