Bifactor Model - No Convergence PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Karen Fong posted on Monday, April 30, 2018 - 11:50 am
Hello Drs. Muthen, I am trying to run a Bifactor Model on my 18-item, 3 factors (6 items in each factor) scale. N=229. I ran a 3-factor model and a Higher Order Model and they ran fine. However, for Bifactor Model, I got the message "No Convergence. Number of Iterations Exceeded." I am just learning SEM right now, and I ran a Bifactor model fine for a homework assignment. The model I have is (note the different acronyms for factors and items. The names of the factors were changed after administering the survey):

ig by ai1-ai6;
pc by cr1-cr6;
el by il1-il6;

pil by ai1* ai2 ai3 ai4 ai5 cr1-cr6 il1-il6;

ig with pc@0;
ig iwth el@0;
pc with el@0;

pil with ig@0;
pil with el@0;
pil with pc@0;

I searched through this forum and I'm pretty sure the model statements are correct, but it is not converging. I tried other remedies that I found on this forum and elsehwere, increasing iterations, adding NCONVERGENCE, using the Starting Values listed in the output, and to no avail.

Any help is appreciated. Thank you so much.
 Linda K. Muthen posted on Monday, April 30, 2018 - 12:21 pm
When you free all factor loadings as you do for pil, you must fix the factor variance to one.
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: