Categorical ESEM Correlation More Than 1 PreviousNext
Mplus Discussion > Categorical Data Modeling >
Message/Author
 Jeanette Lim posted on Thursday, July 21, 2016 - 11:27 pm
Hi,

Hi, I am running a categorical ESEM with 4 factors and 17 indicators. However, I received the following error message:

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 V10.

I have examined the correlations using TECH4 and the correlation between f3 and V10 is more than 1. Furthermore, V10 loads the highest onto f3 and V10 only has 3 categories.

Are my results valid or do I need to respecify my model? What can I do to correct this?
 Linda K. Muthen posted on Friday, July 22, 2016 - 6:00 am
When two factor correlate greater than one, the model is inadmissible. You need to change the model.
 Jeanette Lim posted on Friday, July 22, 2016 - 6:14 am
Hi Linda,

Thanks for your prompt reply. I understand that model respecification is needed when the factors correlate more than 1. However, in this case, it is a factor-indicator correlation. Will this still apply?
 Linda K. Muthen posted on Friday, July 22, 2016 - 10:09 am
I would need to see the output because this does not make sense for the message. Please send the output and your 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: