Step 1: | The independent variable (a) must have a significant total effect on the dependent variable (Y).a |
Step 2: | The independent variable must have a significant direct effect on the mediator (m).b |
Step 3: | Use the regression coefficient from step 1 and add m as a predictor to test the effect of m on Y. Then m must have a significant direct effect on Y.a |
Step 4: | Check whether the effect of a on Y disappears/diminishes after adding m.a |