Model with latent variables
Posted on
LearnSEM8
Joined: 08/14/2012
Forums
Hello!
I have a very general question regarding Covariance based SEMs. I have a full SEM model (with latent variables) and my client is wondering how a change in one attribute for one of the factors could affect the end dependent variable. Are standard SEM methodologies capable of answering this question? If so, how should I approach it?
A quick revert will be really helpful! Thanks!
Like regression?
If you have a model (or a part of a model) that's as simple as regression (e.g. http://openmx.psyc.virginia.edu/docs/OpenMx/0.1-1/Examples_Path.html ) then the same kind of thinking applies.
If the model is basically
y ~ 1.2 + .8*x + e
then a one unit increase in x corresponds to a .8 unit predicted increase in y. This thinking generalizes to models in which x and/or y are latent variables.
You can turn a path diagram into a system of equations. This is where the name "structural equation modeling" comes from. Once you have the system of equations, you can figure out the effect that one variable has on another.
Log in or register to post comments