Homogeneous fit indices PreviousNext
Mplus Discussion > Categorical Data Modeling >
Message/Author
 Anonymous posted on Wednesday, April 20, 2005 - 6:46 am
Hello

I have made a CFA to compare the fit os different factor structures of a General Health Questionnaire. The proposed models are one factor, two factor or three factor models. The questions to the test are coded in a binary variable so I used the WLSMV estimator.
The 'problem' I got is that the fit indices are almost the same for every model:

Proposed models CFI NNFI RMSEA chi/d.f
Unidimensional 0.99 0.99 0.048 75.43/54
Politi 0.99 0.99 0.049 72.21/52
Andrich 0.99 0.99 0.049 72.40/53
Schmitz 0.99 0.99 0.052 62.42/43
Graetz 0.99 0.99 0.046 69.27/51
Martin 0.99 0.99 0.048 74.48/51
Worsley 0.99 0.99 0.046 70.68/51
Farrell 0.99 0.99 0.048 67.73/51

Is that homogeneity indicating me that all the models fits the data equally well/bad?

Thank you
 BMuthen posted on Wednesday, April 20, 2005 - 9:29 am
Yes, it looks like the models fit the data about the same.
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: