Last 3 Days
Structural Equation Modeling
posted on Wednesday, May 31, 2017 - 1:36 am
I want to examine a path analyses where X (cat) predicts Y (cat) moderated by (M) and to see whether there are differences across 2 groups (G).
First, I want to test the moderation without the groups by creating a new variable XM which will than contain 6 groups. Can I use chi-square difference test to see whether there is moderation?
Second, I want to see whether the model from the first step differs across 2 groups by using chi-square differences again and than using wald-test for the specific paths.
Is my analytic approach ok?
Bengt O. Muthen
posted on Wednesday, May 31, 2017 - 5:45 pm
Q1: Just see if XM has a significant effect. There is more on this topic in our new book.
Q2: I would suggest using the Wald test (Model Test)
Back to top
Add Your Message Here
This is a private posting area. Only registered users and moderators may post messages here.
Enable HTML code in message
Automatically activate URLs in message