forum.smartpls.com

Dear experts,

My path model consisting of 13 reflective constructs as shown in the figure below. 7 exogenous constructs, 1 mediator, 4 moderators, and 1 endogenous construct form my path model.
The calculation method for moderating effects is two stage.

should I add the interaction terms of the 4 moderators in my path model before I check for assessment of the reflective measurement items and also assessing the indirect and direct effects between the constructs?

Thank you,
Best wishes,
Kamal Karkonasasi

The assessment of the measurement model should not be affected by moderators if you use the two-stage approach, because it is fixed in the second-stage and estimated in the first stage (without the moderator) anyway.

The assessment of direct and indirect paths strongly depends on the moderator. Therefore, you should assess that after including them.

Dear Dr. Jan-Michael Becker,

Thank you for your reply.
Would it be possible to elaborate further?

Warm regards,
Kamal Karkonasasi

What do you not understand?
Measurement model --> does not matter with two-stage.
Direct & indirect effects --> assess them in the moderator model.

jmbecker wrote: ↑Thu Nov 08, 2018 1:02 pm What do you not understand?
Measurement model --> does not matter with two-stage.
Direct & indirect effects --> assess them in the moderator model.

Thank you for your fast reply dear, Dr. Jan-Michael Becker.

Therefore, because I used the two-stage approach for moderating effects, I should have first added moderating effects (using the 2-stage approach) to my path model and then I could assess my constructs' measurement items and the direct and indirect effects.

What is your opinion about the above statement, dear Dr. Jan-Michael Becker?

Thank you.

Is ok