How to interpret good path coefficients and total effect but low f square?

Frequently asked questions about PLS path modeling.
Post Reply
rebeccachan1111
PLS Junior User
Posts: 4
Joined: Sat Oct 26, 2019 12:50 am
Real name and title: Rebecca (research student)

How to interpret good path coefficients and total effect but low f square?

Post by rebeccachan1111 » Tue Nov 05, 2019 5:45 am

Hi all
I have a construct which is shown to have low (0.015) f square to a DV. However the path coefficients from pointing that construct to the DV, directly and indirectly, are significant and the total effect of that construct to the DV is actually high. As f square essentially measures the impact to a DV with and without a particular IV, how can I interpret this seemingly contradicting result? If the total effect is high, taking out that construct should have significant impact (higher f2) to that DV?

jmbecker
SmartPLS Developer
Posts: 1059
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

Re: How to interpret good path coefficients and total effect but low f square?

Post by jmbecker » Wed Nov 13, 2019 12:18 pm

The relative impact may be large, but if you omit the construct other construct will take its role and explain the same variance in the dependent construct. This is likely due to some (high) collinearity between the variables.
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

rebeccachan1111
PLS Junior User
Posts: 4
Joined: Sat Oct 26, 2019 12:50 am
Real name and title: Rebecca (research student)

Re: How to interpret good path coefficients and total effect but low f square?

Post by rebeccachan1111 » Fri Nov 15, 2019 11:47 pm

Thanks Dr Becker, that explains it and makes sense.

RandellD
PLS Junior User
Posts: 1
Joined: Sat Nov 16, 2019 10:02 am
Real name and title: Randell

Re: How to interpret good path coefficients and total effect but low f square?

Post by RandellD » Mon Nov 18, 2019 9:26 pm

How high collinearity are we talking here exactly, Dr Becker?

Post Reply