Complex models PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Amrit Nandan posted on Wednesday, September 12, 2007 - 11:22 am
Hi I am working on a very complex model. The details are as follows:

# LV's = 15 (each having on an average of about 6 variables of the 5 point rating scale type)

Final Destination variable = 1 (another 5 point rating question type)

The underlying structure between the LV's leading to the final destination variable can be complex and has quite a lot amount of flexibility.

I tried few iterations which leads to results like :

df = 2835
CMIN = 10477.5
p = 0.000

RMSEA
my model = 0.116
independence model = 0.19

The model is in pretty bad shape. Is there any methodical way of coming to a good model which gives some good underlying business hypothesis to these variables?

Or is it that I would have to continue shooting in the dark?
 Linda K. Muthen posted on Tuesday, September 18, 2007 - 4:27 am
I would start out with an EFA of the observed variables for your latent variables to see if the structure you are specifying is even close to fitting the 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: