Bifactor model, AMOS and Mplus PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
Message/Author
 Michal Berkowitz posted on Tuesday, June 03, 2014 - 2:25 pm
Hello,
I ran a "bifactor" model with 3 latent factors and 10 indicators: 2 specific factors with 5 indicators each; and one general factor with loadings from all 10 indicators. N = 158.

At first the model did not converge. Then I increased the number of iterations from 1000 to 10000 and the following message appeared: THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE
COMPUTED. THE MODEL MAY NOT BE IDENTIFIED. CHECK YOUR MODEL.
PROBLEM INVOLVING PARAMETER 13.


When I ran the same model in AMOS there were no warnings of any kind and it generated std.errors and fit statistics. The model itself was not good. Fit was actually ok, but the loadings on one of the specific factors did not make sense (some were positive, some negative).

My questions are:

1. why the differences between AMOS and Mplus?

2. What can I do in Mplus in order to test such a model?

This is the syntax I used:

analysis: type = general;
iterations=10000;

model: f1 BY y1 y2 y3 y4 y5;
f2 BY y6 y7 y8 y9 y10;
f3 BY y1 y2 y3 y4 y5 y6 y7 y8 y9 y10;

f1 with f2@0;
f1 with f3@0;
f2 with f3@0;

I should mention I am new both to Mplus and CFA/SEM.

Thank you very much!
michal
 Bengt O. Muthen posted on Tuesday, June 03, 2014 - 5:44 pm
We need to see the full output (tech1 tech5 sampstat); please send to Support together with your license number. Also send the Amos output. And preferably the data.
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: