Mediator - multiple IV

[Twan]

I have a model with 4 IV's, one mediator, one DV, and 2 control variables, and I'm wondering if I have measured my mediation effect in the correct way.

I first removed the mediator from my model, and performed a bootstrapping in SmartPLS 2.0.
After that, i added the mediator again, and performed again a bootstrapping test in SmartPLS 2.0

Based on the results, I defined the mediaton type per hypothesis. However, I have some doubts if if this is the right way to test the mediation type (and if there exist mediation).
Who can advise me on this :)

Red = non significant
Green = significant

[jmbecker]

You need to assess the indirect effects for significance. You do not need to delete your mediator.
If only your indirect effect is significant: full mediation.
If both your indirect and direct effect are significant: partial mediation.
If only your direct effect is significant and not the indirect effect: no mediation.

[Twan]

You need to assess the indirect effects for significance. You do not need to delete your mediator.
If only your indirect effect is significant: full mediation.
If both your indirect and direct effect are significant: partial mediation.
If only your direct effect is significant and not the indirect effect: no mediation.

Oke, do you also know how i can do that without using a Sobel test, only with the T and/or P values that PLS provides?

[jmbecker]

Yes, you should not use Sobel, but only the T and P values from the bootstrapping for the indirect and direct effects.
SmartPLS2, however, does not report the results for indirect effects, but only for direct and total effects. Thus, you have to use either SmartPLS3, which gives you all necessary results, or calculate the indirect effects for all samples by hand (total-direct) and then calculate SE and T and P values.

[Twan]

Yes, you should not use Sobel, but only the T and P values from the bootstrapping for the indirect and direct effects.
SmartPLS2, however, does not report the results for indirect effects, but only for direct and total effects. Thus, you have to use either SmartPLS3, which gives you all necessary results, or calculate the indirect effects for all samples by hand (total-direct) and then calculate SE and T and P values.

I downloaded SmartPLS 3.0, and that works indeed much easier.
However, according to the results below, it means that there i've only one significant mediated path?





I also tried the 'partial posterior approach', for that, I used the calculator below, and it did give me 4 significant indirect paths. Is this a valid method to use?

http://www2.psych.ubc.ca/~cffalk/mediat ... calculator