Degrees of Freedom PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
 Michael Cheng posted on Tuesday, August 15, 2006 - 11:34 am
I recently ran two CFAs in Mplus. One model allows full covariances between all the factors while the second model restricts one of the covariance paths to zero. When I examine the output, however, both models have the same number of degrees of freedom and I don't understand why. Could you illuminate this issue?

Model 1:
DEM BY demss1c@1 demss2c demss3c;
SOM BY somss1c@1 somss2c somss3c;
LPE BY lpess1c@1 lpess2c lpess3c;
CYN BY cynss1c@1 cynss2c cynss3c;
ASB BY asbss1c@1 asbss2c asbss3c;
PER BY perss1c@1 perss2c perss3c;
DNE BY dness1c@1 dness2c dness3c;
ABX BY abxss1c@1 abxss2c abxss3c;
HPM BY hpmss1c@1 hpmss2c hpmss3c;

Model 2 adds:
LPE with HPM@0;

Thanks for your assistance.
 Michael Cheng posted on Tuesday, August 15, 2006 - 11:37 am
Oh, by the way, I'm using MLMV estimation.
 Linda K. Muthen posted on Tuesday, August 15, 2006 - 1:37 pm
MLMV does not compute degrees of freedom in the regular way. See formula 110 of Technical Appendix 4.
 Markus Martini posted on Tuesday, August 27, 2013 - 2:56 am
Hello! I ran a model with a CFA and one dependent variable using the mlm estimator. Model fit is good, however my degrees of freedom are higher than the sample size.
Is this normal?
Thank your very much indeed.
 Linda K. Muthen posted on Tuesday, August 27, 2013 - 2:21 pm
The degrees of freedom are based on the model not the sample size. This can happen.
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