High chi-sq value, n=206, other fit i... PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Carolyn Dayton posted on Tuesday, September 16, 2008 - 8:13 am
Dear Dr. Muthen,

I am running an SEM model with 7 latent variables and 29 indicators. N=206. All residual variances are positive. My chi-square value is quite high = 525, df=410, p=.0001. However, CFI and TLI both = .96 and RMSEA is.037. I’m confused about why my chi-square value is so high, since my sample size is not that big. Overall, this model seems to fit the data well except for the high chi-square. Should I be concerned about the high chi-square value?

Also, one of my latent variable R-square values is “undefined -.067862+00”. My guess is that this R-square becomes zero due to the other variables carrying so much of the variance. Could this be the case?

Thank you,

Carolyn Dayton
 Linda K. Muthen posted on Tuesday, September 16, 2008 - 8:38 am
Which estimator are you using?
 Carolyn Dayton posted on Tuesday, September 16, 2008 - 8:55 am
FIML
 Linda K. Muthen posted on Tuesday, September 16, 2008 - 9:00 am
I think we need to see your full output and license number at support@statmodel.com. Having an undefined R-square does not square with no negative residudal variances.
 Carolyn Dayton posted on Tuesday, September 16, 2008 - 9:15 am
I've just now emailed it. Thanks so much for your help with this!
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: