Second-order CFA - Identifiability Pr... PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Jonathan Gellar posted on Thursday, February 09, 2017 - 8:02 am
I am trying to fit a CFA model, with the following model structure:

timely by Q7* Q10 Q15;
timely@1;
communicate by Q24* Q25 Q29 Q30;
communicate@1;
helpful by Q49* Q50;
helpful@1;
ownhealth by Q42_r* Q43_r;
ownhealth@1;
coordinate by Q28* Q45 Q32;
coordinate@1;
overall by timely* communicate helpful ownhealth coordinate;
overall@1;

I receive the following error/warning message in the output:

THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE
COMPUTED. THE MODEL MAY NOT BE IDENTIFIED. CHECK YOUR MODEL.
PROBLEM INVOLVING PARAMETER 54.

THE CONDITION NUMBER IS 0.110D-23.

Looking at the TECH1 output, Parameter 54 is the coefficient for COORDINATE for the overall factor (the last equation of the model statement). The factor covariance matrix in the Tech4 output is very odd: there are huge covariances between all factors and COORDINATE.

Any ideas what would cause the non-identifiability? Am I not specifying the second order factor (OVERALL) correctly? The model fits normally without this factor, with nothing standing out related to COORDINATE.

Thanks,
Jon
 Linda K. Muthen posted on Thursday, February 09, 2017 - 12:19 pm
Please send the output and your license number to support@statmodel.com. Include TECH4 in the OUTPUT command.
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: