|
|
 |
| Binary observed variable in SEM |
 
|
| Message/Author |
|
|
|
|
I am working with a SEM that has 5 continuous latent variables and 3 observed outcome variables that are binary (No=0, Yes=1). Since my 3 outcome variables (DIST, BEHAV, and SUIC)are not latent, is it necessary to specify them as "categorical"?
Partial Syntax is:
DIST ON Rel Fam Fri Teach Sch;
BEHAV ON Rel Fam Fri Teach Sch;
SUIC ON Rel Fam Fri Teach Sch;
If I specify them as categorical I use the knownclass/class mixture modeling which also has not worked. The analysis ran for 4 days. Any assistance would be appreciated. |
|
|
|
|
If you have dependent variables that are categorical and you want them treated as categorical in the analysis, you must put them on the CATEGORICAL list. The default is to treat them as continuous.
Please send your output, data set, and license number to support@statmodel.com. |
|
|
| Back to top |
|
|
