

Factorial Invariance & missing data 

Message/Author 

Bit na, LEE posted on Sunday, August 25, 2013  8:49 am



Dear Linda, I hope that all things are doing well. I am working on factorial Invariance between a full version test (B test) and a short version test (A test)  A test has been developed as a short version of B test). Therefore, since A test is a short version, there are some missing data in A test data set. However, Mplus could not read the missing data when conducting factorial invariance analysis � which the syntax said as follows: *** WARNING in ANALYSIS command MODEL=NOMEANSTRUCTURE is not allowed in conjuction with TYPE=MISSING. Request for MODEL=NOMEANSTRUCTURE will be ignored. Would you please give me any advice to solve this problem? If you want, I could send the original data set and code. Thank you so much Sincerely, Bitna 


The model must contain means when TYPE=MISSING is used. This is not a problem. A model with unstructured means is the same as a model with no means. 


Hi, Linda I have one more question. So, when missing is defined as "DEFINE: IF (Y3 EQ 0) THEN Y3=_MISSING", "MODEL=NOMEANSTRUCTURE "is also not allowed? 


Tht is correct. A model with unstructured means yields the same results as a model with no means. 

Back to top 

