AMOS: 0 variance above latent variable

#1
Hi all,

I just ran my model, and it fits really well and all the parameter coefficients are strong and at least moderately high. I was confused though because the Variance reported above the latent variable is zero. The model's structure is four measured variables loading into a single latent variables, which feeds into a single measured variable (see attached). I'm not sure what this means, and can't seem to find an answer anywhere.

Any thoughts? Should I be concerned?

THANK YOU!!

PS......I should note that the little zero appears above the latent variable when I draw it using the Indicator icon. It seems to be setting the intercept value for the latent variable to zero automatically. I am trying to show how the variables on far left make up the latent variable in the middle, and measure how much that contributes to the final outcome of addiction.
 
Last edited:
#2
wait...I think I got it. I think I was thinking about it backwards (that the observed variables made up the latent variable). So the latent variable is actually making up the variance of the observed variable (i.e.: it is the construct that underlies the indicator). So it makes sense that there is a zero variance in the model because the latent variable is making up all the other variables in the model.

Does this sound correct?