Interaction in Growth Modeling PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 JMC posted on Sunday, June 30, 2013 - 10:16 pm
Dear Drs. Muthen,

Thank you in advance for your help. I am running my first growth model and I think I found everything I need in the codebook, but I want to have demographic interactions. What is the code for that? I found covariates, but not interactions.

Thank you,
Jenna
 Linda K. Muthen posted on Monday, July 01, 2013 - 5:28 am
Interactions for observed variables are created using the DEFINE command, for example,

DEFINE:

int = x1*x2;

Put int at the end of the USEVARIABLES list and use it on the right-hand side of ON.
 JMC posted on Monday, July 01, 2013 - 8:35 am
Thank you!

I am writing with two questions:

1.First when doing a quadratic growth model with parallel constructs, what should the correlations between I s q look like in. In a parallel linear function it looks like this:
s1 ON i2;
s2 ON i1;
What does this look like when q is added in?

2. I keep getting this error message:
THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE COMPUTED. THE MODEL MAY NOT BE IDENTIFIED. CHECK YOUR MODEL. PROBLEM INVOLVING PARAMETER 25.
Can you guide me on how to solve this so I can get interpretable results?

Thank you,
Jenna
 Bengt O. Muthen posted on Tuesday, July 02, 2013 - 6:02 pm
1. You don't have to use ON but can use WITH also in models without q. But if you want to use ON and have q, I would say

s1 q1 on i2;
s2 q2 on i1;


2. For us to diagnose that you would have to send the output to Support@statmodel.com.
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: