Parallel process model with interations PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Amy Christie posted on Friday, November 02, 2012 - 3:56 pm
Hi there,
I am running a parallel process model, with the intercept of the second process centered on the final time point. I would like to test whether the interaction between an exogenous variable (x) and the slope of the first process (s1) predicts the intercept and slope of the second process. I created an interaction term between s1 and x using the XWITH command. The model runs well, but does not provide fit statistics typical with latent growth models (i.e., without including the interaction). Is it no longer a LGM (given the TYPE = RANDOM)? Are these analyses appropriate? Below is my syntax.
Many thanks.

VARIABLE:
NAMES ARE p1 p2 p3 p4 p5 d1 d2 d3 d4 d5 X;
USEVARIABLES ARE p1 p2 p3 p4
p5 d1 d2 d3 d4 d5 X;
MISSING ARE p1 p2 p3 p4
p5 d1 d2 d3 d4 d5 X (9999);
ANALYSIS: TYPE = RANDOM
MODEL:
i1 s1 | p1@0 p2@1 p3@2 p4@3 p5@4;
i2 s2 | d1@-4 d2@-3 d3@-2 d4@-1 d5@0;
Modxs1 | X XWITH s1;
s2 ON i1 s1 x Modxs1;
i2 ON i1 s1 x Modxs1;
p1 with d1;
p2 with d2;
p3 with d3;
p4 with d4;
p5 with d5;
OUTPUT:
STAND TECH1 TECH3 TECH4;
 Linda K. Muthen posted on Friday, November 02, 2012 - 6:54 pm
When means, variances, and covariances are not sufficient statistics for model estimation, chi-square and related fit statistics are not available. You can look at the fit before you add the interaction. See the FAQ Latent Variable Interactions on the website.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: