Differences in Factor Structure when ... PreviousNext
Mplus Discussion > Categorical Data Modeling >
 Robert Brooks posted on Sunday, March 16, 2014 - 5:17 pm
I have been completing EFA on dichotomous and ordinal scales using the Bayes estimator
I have also been scoring the factors using imputation and the Bayes Estimator
To complete the imputation I in effect use a CFA based on the EFA
I have been getting substantial differences between the two sets of analyses, including:
Substantial loadings in the EFA becoming non-significant;
The 2 factors which in the EFA had a small correlation (0.2) now correlated so highly (0.9) that a one factor solution was a better model?
The two factor model was conceptually a better model than the one.
 Linda K. Muthen posted on Monday, March 17, 2014 - 3:55 pm
Please send outputs that illustrate the problem and your license number to support@statmodel.com.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message