Predictive Relevance for formative measurement

This forum is closed, and read-only.
Locked
Tess
PLS Junior User
Posts: 3
Joined: Tue Nov 15, 2011 2:43 am
Real name and title:

Predictive Relevance for formative measurement

Post by Tess »

1. Can we use Stone-Geisser's Q square to assess the model's capability to predict with formative endogenous latent construct?
User avatar
Hengkov
PLS Super-Expert
Posts: 1599
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Post by Hengkov »

Hi,

No. some literature PLS not recommend it.

Best Regards,
Hengky
Tess
PLS Junior User
Posts: 3
Joined: Tue Nov 15, 2011 2:43 am
Real name and title:

Is there any other methods to asess the

Post by Tess »

Thanks for the reply.

If no, what other methods can we use to assess the predictive power of the structural model with formative endogenous variable?

Akter & Ambra (2011) mentioned that GOF can also be used to determine the overall prediction power of large complex models and is applied for both reflective and formative measures. But formative measures does not have the AVE values that were needed to calculate the GOF. So how would I go about that?

Thanks.
User avatar
Hengkov
PLS Super-Expert
Posts: 1599
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Post by Hengkov »

Hi,

First, some literature PLS no recommend for used formative endogenous construct.

Second: You can compute GoF and GoF relative for formative construct check Espozito Vinzi et al. (2010, p. 58-59) Handbook PLS.

Best Regards,
Hengky
Locked