Gaussian option not allowed for latent class analysis

[Carmelia Carew]

Hi,

I'm trying to use latent class analysis for continuous variables and following the stata help file my input is in the format given in their example (using my own variables):

--------
We are not limited to logit models for our items. If the behavior items are instead continuous, we can type
. gsem (alcohol truant weapon theft vandalism <-, gaussian), lclass(C 3) --------

Anyone else come across this and found a way around it?
Thanks

[Marcos Almeida]

Welcome to the Stata Forum / Statalist.

The variables are supposed to be continuous so as "fit" in a "gaussian" model. You may wish to try with another example, this time, instead of binary variables, truly continuous ones.

Assuming you do have continuous variables, please try to emulate the example 52 of the manual.

In short, please add the constant in the "right' part of the gsem command. The "gaussian" option, by the way, being the default for latent class profile, may be omitted.