Proportion of Variance in Latent Inte... PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Joseph posted on Friday, February 07, 2014 - 12:19 am
Hello Dr.Muthen,

For a simple latent interaction model,

A BY A1 A2;
P BY P1 P2;
AxP | A XWITH P;
Y ON A P AxP;

I am able to calculate the proportion of variance contributed by the interaction term (AxP) by the formula (19) mentioned at :

https://www.statmodel.com/download/LV%20Interaction.pdf

However , I am wondering how I can calculate the proportion of variance contributed by 'A'

Can I calculate it as (squared unstandardized loading of A * Variance of A )/ Variance of Y.

And if the variance of A is set as 1 :
A@1;

Then : proportion of variance contributed by 'A'

= Squared unstandardized loading of A / Variance of Y

Am I correct in my understanding
 Bengt O. Muthen posted on Friday, February 07, 2014 - 7:55 pm
A is correlated with P so the variance decomposition isn't clear. I wouldn't use that proportion.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: