I'm running a mediation analysis using Hayes' PROCESS macro and would like to confirm my interpretation of the results.

In my study, I manipulated nostalgic advertising (X), measured the level of nostalgia it elicited (M), and used purchase intention as the dependent variable (Y).

As expected, the manipulation (X) and the mediator (M) are strongly correlated, since the ads were explicitly designed to evoke nostalgic feelings. The analysis shows a significant and positive indirect effect through M. However, the direct effect of X on Y (controlling for M), often called the “net effect”, and total effect is no longer significant.

My questions are:

Thanks in advance!