Constraints PreviousNext
Mplus Discussion > Confirmatory Factor Analysis >
 Eric Teman posted on Saturday, May 26, 2012 - 3:04 pm
To estimate reliability in a CFA, I used the constraint feature in Mplus. I am wondering if there is a theoretical difference in the following, because empirical results suggest there is.

You can estimate reliability by summing the squared unstandardized factor loadings, multiplying by the latent variance, then dividing all by total variability.

What if, though, you first fixed latent variances to 1, then following the same procedure. Why do the empirical reliability estimates differ?
 Eric Teman posted on Saturday, May 26, 2012 - 3:12 pm
Nevermind, I forgot to account for the 1 in the unstandardized factor loadings. The methods are equivalent.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message