Odd CFA results using MLR PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Chris Coleman posted on Sunday, April 14, 2013 - 8:40 am
Hi,

Consider these competing models and results (same dataset, MLR estimator used, no error/warning messages):

M1:
f1 by o1* o2-o8;
f1@1;
X2=108, df=20, scf=1.33
(H0 scf=1.35, H1 scf=1.34)

M2:
f1 by o1* o2-o8;
f2 by o2* o3 o5 o7;
f1@1 f2@1;
f1 with f2@0;
X2=143, df=16, scf=0.81
(H0 scf=1.64, H1 scf=1.34)

The bifactor model should always yield a lower X2 than the simple one-factor model, shouldn't it?

Thanks,
 Linda K. Muthen posted on Sunday, April 14, 2013 - 4:12 pm
I think this holds for ML not MLR. Please end the outputs and your license number to support@statmodel.com. if you want further information.
 Chris Coleman posted on Monday, April 15, 2013 - 12:23 pm
Hi Linda,

Thanks for your reply. You're absolutely right; the pattern of X2s was as expected with the ML estimator.
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: