Model Fit question. PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Gavin Elder posted on Thursday, May 12, 2011 - 12:34 pm
Hello Drs. Muthen,

We received a reviewer comment in regards to our model fit statistics. All pre-reviewers with MPlus experience agreed that the model fit was above adequate, and mentioned no concerns. We are wondering (1) if the model fit should concern us, and (2) if it should, what we can do about this concern (any citations would be helpful).

Thank you very much,
Gavin Elder


-Chi-square (df=5)= 2.85, p = 0.72;
-comparative fit index = 1.00;
-Tucker-Lewis fit index = 1.00;
-RMSEA = 0.00 (CI: 0.00, 0.07)

Reviewer Comment:"A TL of 1, and an RMSEA of 0 is unusual, to say the least. It may point to an anomaly or to an identification problem (even with df=5, a model might be locally underidentified). I suggest to discuss this with an MPLUS expert."
 Bengt O. Muthen posted on Thursday, May 12, 2011 - 2:15 pm
Mplus would give you a message if the model is not identified and you would not see SEs. Overly good fit results such as these tend to occur with small samples or low correlations between the variables.
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: