Fit Indices Problem PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 P P posted on Monday, June 15, 2015 - 3:09 pm
Hi everyone, I'm running into a problem with my path analysis in Mplus working off summary data. I'm looking at two different outcome variables, and I'm testing models where I just swap out one outcome for the other while keeping the rest of the model the same. My problem is that I get perfect fit indices with out outcome but not for the other outcome (identification doesn't seem to be an issue because the # of variables is the same).

Basically, I'm looking at 6 variables (A through F) in two different models in which I keep 4 of the variables and swap out one of the last 2, keeping the models the same except for the exchanged variables. Here is the structure:

A B C all interrelated
Paths from A B C to D
Paths from A B D to E/F (E and F are different variables, models are tested with either E or F, but not both).

For outcome E, I get perfect fit indices while I don't for outcome F. Why is this happening, or, what are some things that can contribute to this happening?
 Bengt O. Muthen posted on Monday, June 15, 2015 - 3:50 pm
We need to see both outputs.
 P P posted on Monday, June 15, 2015 - 5:00 pm
Ugh, the output is like 20KB when the max is 1.5KB...how do I resolve this?
 P P posted on Monday, June 15, 2015 - 5:11 pm
Found a way to put both things online: http://textuploader.com/nzhh
 Linda K. Muthen posted on Monday, June 15, 2015 - 7:45 pm
They should not be that large. Please send them both along with your license number to support@statmodel.com.
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: