Formative Model with TSSEM

1 reply [Last post]
Ryan's picture
Offline
Joined: 02/08/2014

Hi there,

I am using A/Prof Mike Cheung's TSSEM (See: https://dl.dropboxusercontent.com/u/25182759/Fixed%20and%20Random%20Effe...). The 2nd stage of this method is constructing a path model (a SEM model) based on the A, S, and F matrices of OpenMx.

Mike's examples (as well as the examples of OpenMx) show only 'reflective models' (i.e., the arrows of paths are from a latent variable to its elements.

My question is: can we build formative models (i.e., the arrows of paths are from elements to a latent variable)?

Technically we can do so by specifying the A and S matrices. However, does the formative model make sense in terms of OpenMx (i.e., the reticular-action-model approach, McArdle & McDonald, 1984)?

Best,

R

Mike Cheung's picture
Offline
Joined: 10/08/2009
Hi Ryan, I rarely fit

Hi Ryan,

I rarely fit formative models. I cannot comment whether it makes sense in your research.

Regarding whether the TSSEM approach can be used to fit formative models, the answer is yes. The second stage analysis is merely fitting a correlation or covariance structure with WLS estimation method . The RAM specification is also as flexible as the LISREL model and the Bentler-Weeks model.

Regards,
Mike