Indirect effect (for a non-significant relationship)

Frequently asked questions about PLS path modeling.
Post Reply
thomasmakz@gmail.com
PLS Junior User
Posts: 4
Joined: Wed Dec 02, 2015 12:23 pm
Real name and title: Thomas Mak

Indirect effect (for a non-significant relationship)

Post by thomasmakz@gmail.com » Wed Feb 24, 2016 2:41 am

My situation is:

Y1-->Y3, path coefficient = 0.173; p= 0.054
Y1-->Y2, path coefficient = 0.413; p= 0.000
Y2-->Y3, path coefficient = 0.500; p= 0.000

my question is
Q1. can i state the result like this: although the relationship Y1-->Y3 is not significant, Y1 has indirect effect on Y3 through Y2 (0.413 X 0.5),
Q2 if yes, is there any significance test for the indirect effect?
Q3 if yes of Q1, shall i sum up the indirect effect and direct effect (not significant) to total effect?
Thank you very much

jmbecker
SmartPLS Developer
Posts: 1094
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: Indirect effect (for a non-significant relationship)

Post by jmbecker » Fri Mar 04, 2016 3:17 pm

Q1: You might conclude this if the indirect effect is significant. The situation you describe is called a mediation model (in your case with an insignificant direct effect a full mediation).
Q2: Yes, you also get the significance information for indirect effects from the bootstrap results. (SmartPLS 3)
Q3: It totally depends on what you want to study. The total effect might be interesting for you (you also get significance information on total effects from the bootstrap results) or it might not be. You should familiarize yourself with the concept of mediation and assess your model again based on that concept.
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
GoogleScholar: http://scholar.google.de/citations?user ... AAAJ&hl=de

User avatar
Hengkov
PLS Super-Expert
Posts: 1618
Joined: Sun Apr 24, 2011 10:13 am
Real name and title: Hengky Latan
Location: AMQ, Indonesia
Contact:

Re: Indirect effect (for a non-significant relationship)

Post by Hengkov » Thu May 19, 2016 6:50 pm

Q1: Yes, if you examine the direct effect first.
Q2: Many approaches are available today, but in the PLS you can use the bootstrap CI.
Q3: I think the total effect is not in accordance with your studies.

Post Reply