any goodness index

This forum is closed, and read-only.
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 Lyudmil,

No, for communality average all construct in model (exogenous + endogenous), only R-squares for endogenous (because exogenous R-squares = 0), OK.

Communality / AVE must higher than 0.5, your result above some construct < 0.5, so not acceptable (check outer loading indicators and compute GoF again)

Best Regards,
Hengky
lyudmi
PLS Junior User
Posts: 8
Joined: Sat Mar 24, 2012 8:22 am
Real name and title:

Post by lyudmi »

Thank you so much, Hengky!
lyudmi
PLS Junior User
Posts: 8
Joined: Sat Mar 24, 2012 8:22 am
Real name and title:

Post by lyudmi »

Hi, Hengky.

You say that:
Hengkov wrote: Your communality construct very low, please check outer loading must higher than 0.6 or 0.7.
In which book, can I find that?

What if the values are less than 0.6 or 0.7 - should I remove the items with loadings lower than these values and test again what GoF will be?

Thanks,

Lyudmil
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,

Yes right.

Best Regards,
Hengky
Eins
PLS Expert User
Posts: 44
Joined: Thu Jun 14, 2012 5:36 am
Real name and title:

Post by Eins »

Some of your constructs have AVE of less than .5. The AVE will be improved if the indicator reliability (or outer loading) is .7 or higher. Just look at the outer loading for each item- it should be .7 or higher.
Locked