Leon posted on Wednesday, April 17, 2013 - 4:27 pm
I am doing confirmatory factor analysis with categorical data (Likert scale data in Psychology).
Would you say that it is enough to report the model fit of the measurement model, estimates, correlations and r-squares of the factors / items for an article.
The reason I ask is because there is no way to compete models with DIFFTEST as they are not nested?
Or do you think it is acceptable to use ML analysis (without specifying categorical) and then to compare the BIC values which is the better model and then use a categorical option for the final model-- and then to report this process.
Leon posted on Wednesday, April 17, 2013 - 6:38 pm
Yes, but it isn't really practical. Except if I am specifying it incorrectly.
*** FATAL ERROR THERE IS NOT ENOUGH MEMORY SPACE TO RUN Mplus ON THE CURRENT INPUT FILE. THE ANALYSIS REQUIRES 8 DIMENSIONS OF INTEGRATION RESULTING IN A TOTAL OF 0.25629E+10 INTEGRATION POINTS. THIS MAY BE THE CAUSE OF THE MEMORY SHORTAGE. YOU CAN TRY TO REDUCE THE NUMBER OF DIMENSIONS OF INTEGRATION OR THE NUMBER OF INTEGRATION POINTS OR USE INTEGRATION=MONTECARLO WITH FEWER NUMBER OF INTEGRATION POINTS SUCH AS 500 OR 5000.
Leon posted on Wednesday, April 17, 2013 - 6:53 pm
Tried it now with montecarlo integeration (2500).
THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO A CHANGE IN THE LOGLIKELIHOOD DURING THE LAST E STEP.
AN INSUFFICENT NUMBER OF E STEP ITERATIONS MAY HAVE BEEN USED. INCREASE THE NUMBER OF MITERATIONS OR INCREASE THE MCONVERGENCE VALUE. ESTIMATES CANNOT BE TRUSTED. SLOW CONVERGENCE DUE TO PARAMETER 20. THE LOGLIKELIHOOD DERIVATIVE FOR THIS PARAMETER IS -0.97063383D-02.