Modeling moderating effect with "Zwei-Phasen-Ansatz&quo

[christianf]

Hello PLS-Experts,

we modeled an moderating effect. Our exogenous variable is formative, our moderator is reflective and the endogenous variable is also reflective specified. In consequence we used the "Zwei-Phasen-Ansatz" for the modeling of the interaction term. Therefore we calculated the main effect model an read out the standardized latent variable scores for the involved variables. Then we modeled the interaction term which has one indicator resulting out of the multiplication of the casewise LV Scores of the exogenous variable and the moderator.

Now our question:
In the main effect model the pathes from the exogenous variable and also from the moderator to the endogenous variable are positive (0,392 and 0,117).
But if we calculate the interaction term then, the path from the interaction term to the endogenous variable is negative (-0,058).

Now we want to know if this is possible and what could be the reason?

Thanks in advance for your kind support.

Greetings Christopher and Christian

[Diogenes]

Hi Christian,
this is a possible result indeed.
1) exogenous has a positive effect on the endogenous
2) moderator has a positive effect too
3.1) the negative interaction means that when the moderator has a bigger value, the effect of the exogenous on the endogenous decrease (even been positive).
3.2) the result (-0,058) is so close to zero, run the bootstrap to see if it is significant.
Best regards.
Bido

[iris_afandiphd]

Is it possible to use on the Unified Theory of Acceptance and Use of Technology(UTAUT) model with moderators such as gender, age etc.