Hi there, I am trying to look at the typologies of food outlets (within 1km from home) and eating behaviours. To do so, I have used the gsem command to fit a latent class model (I have ran the AIC BIC test and a three-class model deemed to be best fit):

gsem (supermarket butcher conveniencestore bakery fastfood cafe<-), logit lclass(c 3)
estat lcprob
estat lcmean

The next step is to examine the association between the 3 typologies and eating behaviour (continous outcome).
I got stuck on this step. I've tried different ways of modelling but kept getting error messages

e.g.
. gsem ( fruit_consumption -> lsupermarket butcher conveniencestore bakery fastfood cafe, logit), lclass(c 3)
option lclass() not allowed;
option lclass() is not allowed with models specified with continuous latent variables


Can anyone help assist with the second step? Is there a way to save each typology and run a regression separately?

Thanks in advance.