What is difference(Or relation) between "Latent Class Analysis"and Structural Equation models" and "Latent Growth Curve Models"?


Latent growth curve models are a kind of structural equation model (but they can also be thought of as a multilevel model).

Latent class models and (most) structural equation models are a kind of latent variable model. In a latent class model, the latent variable(s) is(are) considered to be categorical. In a structural equation model (and a latent growth model) the latent variable(s) is(are) considered to be continuous.

It's possible (but a little unusual) to have structural equation models without latent variables.

This image:

enter image description here

Which I have taken from http://statmodel.com/features.shtml shows the relationship. y and u are continuous and categorical outcomes. f and c are continuous and categorical latent variables. x predictors. The ellipse of A is conventional structural equation modeling, with continuous latent variables (and one or more of these latent variables can represent a growth term), the ellipse B represents latent class analysis.

(There's also a between layer and within layer - that's the multilevel part).

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.