I have a model in which there is a negative relationship between IV and mediator and mediator to the outcome. But the indirect effect in case of mediation analysis comes out as positive. how it can be possible? how can I interpret it
@Huma Indirect effect is nothing but the product of path a (X-->M) and path b (M-->Y) which mathematically, in your case, makes it positive (product of two negatives makes it positive). However, when interpreting the results, you need to be cautious and adhere to the theory and ensure whether the statistical finding makes sense or not.
@imran thanks for answer but do we explain it as positive indirect effect ? As this is totally opposite to what we have theorize in our case.. or can v give the reason that product of two negative make it positive?