Mediation Growth Curve Modeling PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Chelsey Wilkes posted on Tuesday, January 31, 2012 - 6:47 pm
Hello Drs. Muthen & Muthen,
I am trying to conduct a mediation analysis in Mplus using a longitudinal growth curve analysis. The DV (HARS) has 4 time points and both of my IVs (T1ASIm and T1DRS) are static. I wanted to have the mediator as a growth process but am unable to within warnings. I keep recieving "WARNING: THE LATENT VARIABLE COVARIANCE MATRIX (PSI) IS NOT POSITIVE DEFINITE."Also, under the Indirect Command Output, all it has are 0's and 999.000 under Est./S.E. Here is the syntax:
ANALYSIS:ESTIMATOR = MLR;iteration=10000;
MODEL:iHARS sHARS | T1HARS@-2 T2HARS@-1 T3HARS@0 T4HARS@1;
T1HARS with T3HARS;T2HARS with T4HARS;iHARS with sHARS;
iMFQ sMFQ | T1MFQfrq@0 T2MFQfrq@-1 T3MFQfrq@-2 T4MFQfrq@-3;T1MFQfrq with T3MFQfrq;T2MFqfrq with T4MFQfrq;iMFQ with sMFQ;T1HARS with T1MFQfrq;T2HARS with T2MFQfrq;T3HARS with 3MFQfrq;T4HARS with T4MFQfrq;T1HARS with T1ASIm;T1HARS with T1DRS;T1ASIm with T1DRS;T1ASIm with T1MFQfrq;T1DRS with T1MFQfrq;
iHARS on T1ASIm T1DRS;sHARS on T1ASIm T1DRS;
MODEL INDIRECT:
iHARS ind T1ASIm;sHARS ind T1ASIm;
OUTPUT: STANDARDIZED SAMPSTAT TECH1 TECH4 MODINDICES CINTERVAL
Thank you! Chelsey
 Linda K. Muthen posted on Wednesday, February 01, 2012 - 11:48 am
Please send your output and license number 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: