| Message/Author |
|
|
|
|
Last time when i did LCA, we had a lot of variables, and used factor analysis to narrow down the variables to be used in LCA.
Now we have a new data set with 5 variables. Is there a general rule how to choose which ones to use for LCA?
If all of them can separate the classes well, it will be fine. But I am not sure if any of them cannot separate the classes well. |
|
|
|
|
| You should try this question on SEMNET. |
|
|
|
|
The ON statement specifies the multinomial logistic regression of the categorical latent variable c on the continuous latent variable f (c on f)
my question is:
how to specify the multinomial logistic regression of the categorical latent variable c1 on another categorical latent variable c2 ? (the ON statement doesn't run) |
|
|
|
|
c1 on c2;
But in the classes= statement, c2 then has to precede c1. |
|
|
|
|
i am sorry, but i can't understand your comment.
please clarify |
|
|
|
|
ooh, many thanks for your help.
i finally understood what did you mean and the command is run. |
|
|
| Back to top |
