Hi,

My hypothesis is that M can mediates the relationship between X and Y. I have also hypothesised the following in order to check the assumption for running mediation model: (a) X-Y correlated, (b) M-X correlated, (c) M-Y correlated. The test of X-Y relationship at the beginning, there is no direct effect on them so no mediation model can be run. However, (b) and (c) are significantly correlated respectively, can I interpret it as an indirect effect in this case or there is still no effect between these three variables as X-Y no direct effect?