Configural invariance PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Leslie Vaughan posted on Thursday, August 30, 2012 - 6:11 am
Looking at configural invariance in Cog. at T1 and time 2.
Error Message is:
WARNING: THE LATENT VARIABLE COVARIANCE MATRIX (PSI) IS NOT POSITIVE DEFINITE. THIS COULD INDICATE A NEGATIVE VARIANCE/RESIDUAL VARIANCE FOR A
LATENT VARIABLE, A CORRELATION GREATER OR EQUAL TO ONE BETWEEN TWO LATENT
VARIABLES, OR A LINEAR DEPENDENCY AMONG MORE THAN TWO LATENT VARIABLES.
CHECK THE TECH4 OUTPUT FOR MORE INFORMATION. PROBLEM INVOLVING VARIABLE COG2.
LV's are correlated 1.56 (how to interpret?)Should the Cog2 means and variances/covariances generally be free?
Are there any particular start values I should use?

USEVAR = vfa
trailb digitb erecall vfa2
trailb2 digitb2 erecall2;
MISSING = .;
ANALYSIS:
TYPE = MEANSTRUCTURE;
ESTIMATOR = ML;
ITERATIONS = 20000;

Cog1 BY vfa*1 trailb digitb erecall;
Cog2 BY vfa2*1
trailb2 digitb2 erecall2;
[Cog1@0];
[Cog2@0];
Cog1@1;
Cog2@1;
Cog2 WITH Cog1;
[ vfa trailb digitb erecall];
[ vfa2 trailb2 digitb2 erecall2];
vfa trailb digitb erecall;
vfa2 trailb2 digitb2 erecall2;
vfa WITH erecall;
vfa2 WITH erecall2;

Thanks very much for your suggestions.
 Linda K. Muthen posted on Thursday, August 30, 2012 - 10:26 am
A correlation greater than one means the model is misspecified. You would need to change the model. You could ask for MODINDICES (ALL) in the OUTPUT command to see if the modification indices offer any guidance.
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: