
Message/Author 

Anonymous posted on Tuesday, November 30, 2004  4:15 pm



Dear Dr. Muthen, I am running a single group CFA with categorical observed indicators. My ultimate goal is to do a multiple group comparison. But first, I am trying to fit my model for each group. To do so, in mplus I just wrote Y ON X1X5 and ran the model for each group seperately. The results produced factor loadings, thresholds, factor variance, and residual variance. My questions : 1 Do I need to constrain my thresholds in a single group CFA? 2 I did not get scale factors in my single group CFA. Do I need to specify scale factros and/or maybe constrain them? 3 In the output, I also did not see the factor mean. Shoud I specify it in my model? In general, could you please let me know what should I specify in my model for a single group CFA with categorical outcomes? 


You should use the BY statement to specify a CFA model. See Example 5.2 in the Mplus User's Guide. 1. No. 2. No. They are fixed at one as the default. 3. Factor means are zero. 

Back to top 

