Labeling correlated errors PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Joseph Dien posted on Thursday, March 07, 2013 - 2:32 pm
Hi, I need to include correlated errors terms in my Model Constraints to implement Raykov's 2009 reliability formula. I have the rest of the MPLUS code from his 2010 book. How does one label the pairwise correlations between error terms? I couldn't find anything about them in the MPLUS manual, just how to label error variances.

Thanks!

Joe
 Linda K. Muthen posted on Thursday, March 07, 2013 - 3:26 pm
u1 WITH u2 (p);

labels a residual covariance if u1 and u2 are dependent variables.
 Joseph Dien posted on Friday, March 08, 2013 - 10:47 am
What about the case of u1-u3 WITH u1-u3, where there are three covariances? It accepts the syntax but I'm not clear how the labels are mapping onto the covariances. It occurs to me that a possible alternative is to use the PWITH command to specify the pairs explicitly so that each can be labeled individually but it'd be easier if I could just use the WITH approach.

Thanks!

Joe
 Linda K. Muthen posted on Friday, March 08, 2013 - 11:19 am
Which covariances do you actually want? You cannot identify all of them.
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: