Fundamental mediation analysis question

Michael_Jasper

So, I have 3 variables: A, B, C. Which I assign to be X, M, Y:

X = A

M = B

Y = C

Mediation analysis (bootstrapping) suggests an indirect effect: i.e. A (X) determines C (Y) through B (M).

After the above positive result, does then testing it the other way around add any further supporting data to 

this positive result? i.e.: 

X = C

M = B

Y = A

IF in this opposite direction there is NO indirect effect observed. 

So there appears to be some directionality. But I am unsure of the relevance of this. Does it add to the 

case for the mediation? 

EJ

I'm not sure, but I think the first question is whether you'd expect the same result without the mediator, and I don't think you do. See for instance https://stats.stackexchange.com/questions/22718/what-is-the-difference-between-linear-regression-on-y-with-x-and-x-with-y

Cheers,

E.J.