Outer weight sign

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
pipposca
PLS Junior User
Posts: 5
Joined: Tue Nov 07, 2006 2:31 pm
Real name and title:

Outer weight sign

Post by pipposca »

Hi,
I’m interested in calculating the impact of manifest variables in the ultimate (dependent) latent variable. As I read in forum, I’ll use the outer weights but, after estimation procedure (Mode B), some of these are with negative sign while others have positive sign. Is this possible?
Can I use PLS path modelling with standardized objective data?
Thanks for your helps.
Giuseppe
jjsailors
PLS Expert User
Posts: 62
Joined: Fri Oct 14, 2005 1:43 am
Real name and title:

Re: Outer weight sign

Post by jjsailors »

pipposca wrote:Hi,
I’m interested in calculating the impact of manifest variables in the ultimate (dependent) latent variable. As I read in forum, I’ll use the outer weights but, after estimation procedure (Mode B), some of these are with negative sign while others have positive sign. Is this possible?
Can I use PLS path modelling with standardized objective data?
Thanks for your helps.
Giuseppe
Specify the model with reflective indicators; this only impacts the method by which PLS weights are estimated. Once the weights are estimated their signs will match their correlations with the dependent variable(s). Your problem is likely multicollinearity among the indicators.

John
John J. Sailors, PhD
Associate Professor of Marketing
The University of St. Thomas
Opus College of Business
Minneapolis, MN
Post Reply