comparison of path coefficients

Frequently asked questions about PLS path modeling.
Post Reply
sandra
PLS User
Posts: 14
Joined: Sat Mar 11, 2006 7:08 pm
Real name and title:

comparison of path coefficients

Post by sandra »

Hi,

I would like to test if one path coefficient (e.g. b1) is bigger than another (b2). Usually I would use (after Regression):

t=b1-b2/(varb1+varb2-2covb1b2)^0,5

Is this test possible with PLS coefficients, too? Unfortunately the covariance is not provided - or does PLS assume the Cov to be zero?

Thanks a lot

Sandra
stefanbehrens
PLS Expert User
Posts: 54
Joined: Wed Oct 19, 2005 5:53 pm
Real name and title:

Post by stefanbehrens »

Hi Sandra,

given that - at a structural level - PLS is just a multiple regression between latent variables, your test should work.

While the covariance of coefficients is not reported by SmartPLS, you could simply export the path coefficient data from the bootstrap report to Excel and then calculate the covariance there.

Cheers,
Stefan
User avatar
Diogenes
PLS Super-Expert
Posts: 905
Joined: Sat Oct 15, 2005 5:13 pm
Real name and title:
Location: São Paulo - BRAZIL
Contact:

Post by Diogenes »

Hi all,

a reference in viewtopic.php?t=172

to create an I.C. with bootstrapping, since PLS doesn´t have normality assumption.

Best regards
Prof. Dr. Diogenes de Souza Bido
Post Reply