Latent indicator coefficient =1

[E. Bernardo]

Dear SEM Experts,

In CFA/SEM, the regression weight of the path from a latent variable to one of its indicators/items is constraint to 1.0 (a user value) so that the other parameters can be estimated.  Can we remove this 1.0, and put it instead as variance of the latent? Any available reference?

Thank you in advance for your comments.

[Bayram YILMAZ]

Dear Bernardo,

You may want to check MPLUS website and Muthens' explanations on the discussion list.

Bayram

On Mon, Dec 2, 2013 at 11:40 AM, E. Bernardo <[hidden email]> wrote:

Dear SEM Experts,

In CFA/SEM, the regression weight of the path from a latent variable to one of its indicators/items is constraint to 1.0 (a user value) so that the other parameters can be estimated.  Can we remove this 1.0, and put it instead as variance of the latent? Any available reference?

Thank you in advance for your comments.

--

"If we knew what we were doing, it wouldn't be called research, would it?" Albert Einstein

[Ryan]

Yes. Most SEM textbooks cover this topic. 

Ryan

On Mon, Dec 2, 2013 at 4:40 AM, E. Bernardo <[hidden email]> wrote:

Dear SEM Experts,

In CFA/SEM, the regression weight of the path from a latent variable to one of its indicators/items is constraint to 1.0 (a user value) so that the other parameters can be estimated.  Can we remove this 1.0, and put it instead as variance of the latent? Any available reference?

Thank you in advance for your comments.