Invariance testing on a Bifactor model PreviousNext
Mplus Discussion > Categorical Data Modeling >
 Roberto Guerra posted on Tuesday, December 10, 2013 - 1:35 pm
Hello, we are attempting to test for multigroup (gender) invariance using MLR for a bi-factor model and could use some guidance. I have included in a reply to this message the input we are attempting to use for our configural model (shortened to fit on the message board). Can you confirm whether we are correctly freeing the intercepts for a bifactor model, and whether you detect any other issues with the model? One question we have regards how to specify loadings, intercepts, and residual variances from an item (e.g., icu7) to the bifactor vs. that item to its subscale factor. (I have shortened the model below, and only added in certain excerpts, but it demonstrates the general issue at hand) Thanks in advance!
 Roberto Guerra posted on Tuesday, December 10, 2013 - 1:36 pm
icu1@1 (L5)
icu2* (L6)
icu3* (L7)
icu4* (L8);
icu1@1 (L1)
icu2* (L2);
icu3@1 (L3)
icu4* (L4);
[icu1*](I1); [icu2*](I2); [icu3*](I3); [icu4*](I4); [icu1*](I5); [icu2*](I6); [icu3*](I7); [icu4*](I8);
icu1*(E1); icu2*(E2); icu3*(E3); icu4*(E4); icu1*(E5); icu2*(E6); icu3*(E7); icu4*(E8);
F1*; F2*; F3*; [F1@0]; [F2@0]; [F3@0];

Model male:
F3 by icu1@1, icu2-icu4*; F1 by icu1@1, icu2*; F2 by icu3@1, icu4*; [icu1-icu4*]; icu1-icu4*; F1*; F2*; F3*; [F1@0]; [F2@0]; [F3@0];
 Linda K. Muthen posted on Tuesday, December 10, 2013 - 1:46 pm
Please send the full output and your license number to

Please limit posts to one window.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message