Model fit PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
 MBH posted on Monday, March 31, 2014 - 11:12 pm
Dear Linda and Bengt,

I'm running a CFA prior to running SEM. I've constructed a CFA of 3 variables and here are the model fit information I'm getting:

RMSEA (Root Mean Square Error Of Approximation)

Estimate 0.000
90 Percent C.I. 0.000 0.000
Probability RMSEA <= .05 0.000


CFI 1.000
TLI 1.000

Chi-Square Test of Model Fit for the Baseline Model

Value 15690.622
Degrees of Freedom 3
P-Value 0.0000

WRMR (Weighted Root Mean Square Residual)

Value 0.006

Does the statistics above mean that the model is a perfect fit to be included subsequently in the SEM?

I've been reading that CFI should be below 1.00 and be between 0.90-1.00 and not be exactly 1.00.

Thanks for your help.
 Linda K. Muthen posted on Monday, March 31, 2014 - 11:53 pm
A factor with three indicators is just identified. Fit cannot be assessed. The model has no degrees of freedom.
 MBH posted on Tuesday, April 01, 2014 - 12:13 am
Thanks for the quick reply.

So this means that the factor cannot be included in further analysis for SEM?
 Bengt O. Muthen posted on Tuesday, April 01, 2014 - 12:31 am
You can include it. It will affect the fit of the total model.
 MBH posted on Tuesday, April 01, 2014 - 12:42 am
Dear Bengt,

That's what I thought. But in situations like this, it is arbitrary to either include it or not to include it?
 Linda K. Muthen posted on Tuesday, April 01, 2014 - 11:56 am
It is your decision whether you want it in the model.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message