Model underidentification Amos

[Caroline Ng Tseung-Wong]

Hi,

  Would be great if anyone could explain to me why Amos is saying my
model is probably underidentified in the notes for model:

i'm running a multigroup latent means analysis with three age groups.  
My unconstrained model consist of three latent factors (all correlated)
and each with two indicators loading only on the respective factor.  
There are no error co-variance. I have manage to do a multigroup (COV)
structure showing group invariance of my three-factor model.  However,
when running the multigroup for latent means using amos graphics and the
automated approach, fixing one of the group to mean zero and
constraining the three factors for the other group (as shown in Byrne,
2010), I do not get a computed chi square value in notes for model
although the df is positive it says i should impose three additional
constraints? I'm new to Amos and CFA for that matter, can someone help
me make sense of that! Thanks!

[SR Millis-3]

FYI:

 

http://www2.gsu.edu/~mkteer/identifi.html

 

~~~~~~~~~~~
Scott R Millis, PhD, ABPP, CStat, CSci
Professor & Director of Research
Dept of Physical Medicine & Rehabilitation
Dept of Emergency Medicine
Wayne State University School of Medicine
261 Mack Blvd
Detroit, MI 48201
Email: [hidden email]
Email: [hidden email]
Tel: [phone]
Fax: [phone]

--- On Mon, 7/19/10, cal <[hidden email]> wrote:

From: cal <[hidden email]>
Subject: Model underidentification Amos
To: [hidden email]
Date: Monday, July 19, 2010, 2:21 AM

Hi,

  Would be great if anyone could explain to me why Amos is saying my
model is probably underidentified in the notes for model:

i'm running a multigroup latent means analysis with three age groups.  
My unconstrained model consist of three latent factors (all correlated)
and each with two indicators loading only on the respective factor.  
There are no error co-variance. I have manage to do a multigroup (COV)
structure showing group invariance of my three-factor model.  However,
when running the multigroup for latent means using amos graphics and the
automated approach, fixing one of the group to mean zero and
constraining the three factors for the other group (as shown in Byrne,
2010), I do not get a computed chi square value in notes for model
although the df is positive it says i should impose three additional
constraints? I'm new to Amos and CFA for that matter, can someone help
me make sense of that! Thanks!

---

View this message in context: Model underidentification Amos
Sent from the SPSSX Discussion mailing list archive at Nabble.com.