I'm running a CFA Bifactor model to validate a translated questionnaire (using AMOS).
The AMOS output showed 'Iteration limit reached. The results that follow are therefore incorrect'. What does this mean? And are the model fit indices in the AMOS output reliable?
In addition, I cannot view the factor loadings (the 'view the input path diagram' button cannot be clicked).
What am I supposed to do?
The model did not converge (you ran out of iterations). Therefore, the current fit statistics are not trustworthy.
One reason may be that you specified error (residual) terms for the latent variables (factors) F1 through F6. The latent variables in this model are exogeneous (purely independent variables) and should not have error terms attached to them. This may have caused AMOS to try to estimate both factor variances and error variances for the latent variables leading to an over-parametrized (and under-identified) model. I would try deleting the error terms e19 through e25 and re-running the model.
There can be other issues with standard bifactor models as well. See:
Article Anomalous Results in G-Factor Models: Explanations and Alternatives
@christian geiser
I deleted the error terms and re-run the model, it works! Thanks!
- Badges
- Science method