Chi-square LISREL vs. Mplus PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Eric Teman posted on Friday, April 25, 2014 - 5:10 pm
I am doing an observed path analysis in Mplus and LISREL. I am trying to reproduce the independence null (baseline) chi-square that LISREL reports. Please tell me if this is the right way to have Mplus do it for the following model:
AM ON SES@0;
GPA ON AM@0 IQ@0;
SES;
IQ;
SES WITH IQ@0;

This allows estimation of all variances of DVs and IV's and fixes the covariance of IV's to 0.

However, I am unable to reproduce this in LISREL. I am using the following model commands: MO NY=2 NX=2 PH=DI,FR PS=DI,FR GA=FU,FI BE=FU,FI

Shouldn't the chi-squares match up?
 Linda K. Muthen posted on Saturday, April 26, 2014 - 10:50 am
Make sure that you have have the same number of parameters in each analysis. Mplus may correlate the residuals of am and gpa as the default.
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: