How to interpret/report?

#1
Hi all,

I'm having trouble interpreting my results. I have 1 outcome variable, 1 mediator, 3 predictors and 3 control variables. I am doing three seperate mediation analyses, for every predictor variable once. The control variables are taken as covariate in PROCESS. But I am in doubt if I have to take the other two predictors as covariate as well. The theory i am using for my thesis postulates there are only indirect effects of the predictors (via the mediation). PROCESS can't control for this indirect effect, only for the total effect.

When i do control for the other two predictors. All the variance goes to only one predictor because of the high correlation between them (no official multicollinearity). When I don't control for the other two predictors all mediation analyses are significant. I really don't know what to do with these results. When I do control for the other predictors, i think i am really underestimating the effect of the other two. But when I don't i guess i am overestimating the results?

Kind Regards,

Rik
 
Last edited: