Moderation Analysis through Process by Andrew Hayes and SPSS regression analysis

#1
Hello all,

I need help analyzing some data for a moderation analysis. I have a continuous variable (predictor), Exhaustion, a continuous outcome, "Social Loafing" (coded 1, work alone, 2, work in a pair, or 3, work in a group--it's really an ordinal variable), and a continuous variable "Extraversion" which I believe moderates the relationship between Exhaustion and Social Loafing.

I ran a moderation analysis in SPSS using PROCESS by Andrew F Hayes and it resulted in a significant p-value for the interaction only, p = 0.038. However, when I tried to run a multiple regression analysis with Exhaustion (centered), Extraversion (centered) and ExhaustionCentered * ExtraversionCentered (for the interaction effect), there was no significance for any of the b-values in the SPSS output.

I took a couple Statistics courses but am obviously way out of my league here. I had an Advanced Stats class where the professor taught us how to examine moderation by using this multiple regression analysis; but he also said we could use PROCESS. I'm inclined to believe the PROCESS results because of wishful thinking. Are these two analyses really measuring the same thing, the moderating effect of Extraversion on the relationship between Exhaustion and Social Loafing, or are these two analyses not measuring the same thing at all?

Help would be MUCH appreciated. And please, forgive my ignorance. :eek: