Application of unstandardized total effects in IPMA

Report and discuss bugs of SmartPLS 2. To report bugs of the new SmartPLS 3 software, create a ticket in our helpdesk.
Post Reply
Nordue
PLS Junior User
Posts: 1
Joined: Wed Nov 04, 2015 9:14 am
Real name and title: Anne Duedahl, Cand.Merc.Business Intelligence

Application of unstandardized total effects in IPMA

Post by Nordue » Wed Nov 04, 2015 9:25 am

Hello,

I have run the IPMA analysis in Smart PLS 3, but when I check my performance-importance maps for both constructs and indicators, I notice that it is the unstandardized total effect, which is used to indicate importance.

I reviewed chapter 7 in A primer on partial least squares structural equation modeling (PLS-SEM) by Hair, Hult, Ringle and Sarstedt (2014), and I found that the authors used the standardized total effects/indicated by the path coefficients when conducting the IPMA analysis.

Thus, I am puzzled as I would expect the importance/performance map for both constructs and indicators to use the standardized total effects for measuring importance.

Does anyone have an explanation for this?

Best regards,

Anne

User avatar
cringle
SmartPLS Developer
Posts: 796
Joined: Tue Sep 20, 2005 9:13 am
Real name and title: Prof. Dr. Christian M. Ringle
Location: Hamburg (Germany)
Contact:

Re: Application of unstandardized total effects in IPMA

Post by cringle » Thu Nov 05, 2015 2:03 pm

Please use the unstandardized coefficients as they better represent the cetris paribus considerations.

You can cite the following source which will appear soon:

Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2016). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM) (2 ed.). Thousand Oaks, CA: Sage.

Best regards
Christian Ringle

JSILVA
PLS Junior User
Posts: 2
Joined: Tue Sep 08, 2015 3:08 pm
Real name and title: Javier Silva - Marketing Professor - IAE Business School - Buenos Aires

Re: Application of unstandardized total effects in IPMA

Post by JSILVA » Tue Jan 12, 2016 2:40 pm

Dears,

I have a construct X using a standardized total effect that results as the most important. However, by using unstandardized total effect changes its relative position and becomes the third in matters of importance. Could you help me to understand this?

Thank you in advance
Javier

martal27
PLS Junior User
Posts: 1
Joined: Mon Feb 22, 2016 10:13 am
Real name and title: Marta López Ron Graduada en Matemáticas

Re: Application of unstandardized total effects in IPMA

Post by martal27 » Mon Feb 22, 2016 10:32 am

Hi everyone,
I`ve been reading some of the posts and I have the same questions. I can see that it´s better to use the unstandardized coefficients for the importance-performance matrix, but if I want to express the importance relative of my latent variables with the standardized total effects, how can I change the importance-performance matrix?


Thanks a lot!

Post Reply