I planned to explore a factor structure of a scale with SPSS and then confirm it with Mplus. While I get strong models in SPSS, they are not strong on Confirmation (CFI+.79).
I thought I'd try to look at the Factor analysis with Mplus but am puzzled that I get a slightly different factor structure (with different loadings) to my FA with SPSS. I have checked that I am using the same rotation and using Maximum Likelihood which I read is the default on Mplus, but still the results are different.
Is it just because Mplus is more sensitive? Could anyone tell me why?
You may not have the same sample sizes. If you have the same sample, model, and estimator, the results should be the same. If you can't see the problem, send both the Mplus and SPSS outputs files and your license number to email@example.com.