I did a mediation analysis with one mediator and used PROCESS. My mediation analysis revealed:
Path c = significant (total effect)
path a = significant
path c' = not significant (direct effect)
path b = not significant.
Based on the 5000 bootstrapped interval there is no indirect effect
I know that it means that there is no mediation. But what does it say about the relation between X and Y. Can I still argue in my discussion that I confirm recent studies that found an effect of X on Y? (Although the direct effect was insignificant).
Thanks in advance!
Best Answer
Remember that having p>.05 doesn't mean there is no effect. It means you don't have enough evidence to make a conclusion about the presence or absence of an effect. You do not have enough evidence to claim there is an indirect effect nor a direct effect conditional on the mediator. There may or may not be mediation, and there may or may not be remaining direct effect. The lack of significance simply tells you you don't have enough information to make a conclusion on those specific effects. The results are consistent both with there being no effect and there being an effect that is not detectable with the sample size you have.
You do have enough evidence to claim there is a total effect, however, and that seems to be a claim you wish to evaluate. You can say that you have replicated other studies that have found a similar total effect.