Dear Forum participants,
I would like to ask you if the procedure described by these authors is fine. They test just 2 models: one with a mediator and one without. I think it is more than enough...:
"Here we further test for mediation following the approach outlined by Baron and Kenny (1986). First, we tested a simple model with a direct path between perceived social presence and attitude (eliminating perceived usefulness, trust and enjoyment) in PLS. The path coefficient between perceived social presence and attitude was significant (b=0.459**). When the mediating variables of perceived usefulness, trust and enjoyment were added to the simple model, the path coefficient between perceived social presence and attitude became insignificant (b=0.086)." (Hassaneina and Head "Manipulating perceived social presence through the web interface and its impact on attitude towards online shopping", International Journal of Human-Computer Studies, 65:8 2007).
However, I read the Baron and Kenny and they describe that i should in fact evaluate 3 models....:
"First, the independent variable
must affect the mediator in the first equation; second, the independent
variable must be shown to affect the dependent variable
in the second equation; and third, the mediator must affect the
dependent variable in the third equation. If these conditions all
hold in the predicted direction, then the effect of the independent
variable on the dependent variable must be less in the third
equation than in the second. Perfect mediation holds if the independent
variable has no effect when the mediator is controlled."
Thank you very much!!!!
With warm regards,
Hanna Krasnova