IV to MV & MV to DV significant, but no mediating effect

Questions about the implementation and application of the PLS-SEM method, that are not related to the usage of the SmartPLS software.
Post Reply
aznur
PLS Junior User
Posts: 3
Joined: Thu Sep 19, 2019 10:18 pm
Real name and title: Aznur

IV to MV & MV to DV significant, but no mediating effect

Post by aznur »

I have a question, i need to know in sem pls analysis, if the direct effect of an independent variable (iv) to a mediating variable (mv) is significant and mv to dependent variable (dv) is also significant. (V -> MV (sig), MV --> DV (sig)

But after bootstrapping test , the result shows the mediating variable is no longer significant to mediate IV and DV.

It is a full mediation if we follow barron n kenny method.
I know a lot of limitations if using Barron and Kenny's method.
But how to justify the roles of mediation here. Is the mv really a mediator or not? any explanation for this result? and how should i interpret the results?

Any reading can I refer to?

Thank you
Post Reply