Fit indices and stand residus with no... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Carol Barry posted on Sunday, September 13, 2009 - 11:07 am
Dr. Muthen,

I am a graduate student using Mplus for growth modeling. My data are nonnormal, so I am using MLM estimation in order to obtain the S-B adjusted chi-square and standard errors.

First, are the fit indices (in particular, the CFI and the RMSEA) reported in the output the S-B adjusted fit indices?

Second, I want to examine the standardized residuals for both the means and covariances to assess local fit/misfit. I realize I can only obtain standardized residuals using ML, MLR, or MLF estimation. To get around this, I re-ran my models using MLR (noting that the fit indices and standard errors changed only very slightly from the MLM estimation) and examined these standardized residuals. Is there any way to obtain standardized residuals using MLM estimation and, if not, is the approach I took appropriate?

Thank you in advance.
 Bengt O. Muthen posted on Sunday, September 13, 2009 - 12:00 pm
Yes on your first question.

MLR is an appropriate approach here - it also handles missing data.
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: