CFA PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 amna khalid posted on Wednesday, March 13, 2013 - 1:03 pm
Hi,

I am trying to run CFI on a scale in MPlus in which an item is used to compute two different sub-scales using reverse scoring and forward scoring. When I run the analysis in Mplus it gives me this warning
WARNING: THE SAMPLE COVARIANCE IS SINGULAR. THE SAMPLE CORRELATION OF ARSQ5 AND ARSQ5R IS -1.000. NO CONVERGENCE. NUMBER OF ITERATIONS EXCEEDED.

I am using the following input command.

F1 by ARSQ3 ARSQ7R ARSQ8 ARSQ10 ARSQ17R;
F2 BY ARSQ1 ARSQ4 ARSQ9 ARSQ14;
F3 BY ARSQ5R ARSQ6 ARSQ11 ARSQ15;
F4 BY ARSQ2 ARSQ5 ARSQ12 ARSQ13 ARSQ16;
F5 by F1-F4;


Kindly advice.
 Linda K. Muthen posted on Wednesday, March 13, 2013 - 1:38 pm
You cannot use two variables in an analysis that have a correlation of one or minus one.
 amna khalid posted on Thursday, March 14, 2013 - 8:50 am
Thank you for your reply. Is there a way to fix this correlation. I have tried incorporating this command ARSQ5 with ARSQ5R @0 but still getting the same warning.
 Linda K. Muthen posted on Thursday, March 14, 2013 - 9:27 am
You need to use just one of the variables. Fixing the correlation to zero is not the solution.
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: