There is a positive effect of X on M
There is a positive effect of M on Y
M mediates the effect of X on Y, or
X has an indirect effect on Y through M
In this approach, there is only one hypothesis to be tested:
H1: M mediates the effect of X on Y, or
H1: X has an indirect effect on Y through M
Theoretical consideration-whether the underpinning theory suggests testing the relationship between X and Y.
Analytical approach. Example: Baron & Kenny's (1986) approach or Preacher & Hayes (2004,2008), which focuses on indirect effect.
Problem statement. Example: if the researcher argues X affect Y because of M, or if the researcher argues the established relationship between X and Y and he/she is interested to whether M mediates the effect or not.
Rungtusanatham, M., Miller, J. W., & Boyer, K. K. (2014). Theorizing, testing, and concluding for mediation in SCM research: tutorial and procedural recommendations. Journal of Operations Management, 32(3), 99-113. Click here.