In a model that I am working on, we have a significant mediation and significant moderation by two moderators, both located between the IV and mediator. However, the overall index of moderated mediation is insignificant. In the model, despite significant moderation and mediation, there is no change in the value of coefficient between the IV and DV. Why is this happening?
@Swati You should provide a screenshot of your results then it would be easier for others to provide specific advices. Due to lack of information, it's not possible to get to know what type of variable your moderator is. Whether you tested the moderation interaction at +1 SD -1SD or Percentile method. Moreover, it's also not clear whether you tested moderation at two levels i.e., on both paths X>M and M>Y, on X>M path or on M>Y path and what software has been used.
Imran Anwar I am using two moderators, Machiavellianism and political will, both on a continuous scale. I used PROCESS model 9 wherein both the moderators W and Z are in the X-->M path. Tested the moderation interaction at +/- 1 SD. I first tested the mediation which is significant (partial mediation) and then introduced both the moderators (fyi, I ran the model with both the moderators independently too but the moderated mediation is still insignificant).
I recommend that you use the JAMOVI program, it is free and easy to perform the respective analysis, because it allows you to analyze models of moderated mediation considering specifically the each category of the moderating variable instead of the standard deviation, since it reports that the moderated mediation is not significant, you can consider the differences between the moderated paths categorically.
@Cristian Ramos-Vera Thank you for the suggestion. I will try it out. I was not aware a software like this exists.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials
- Carbon Materials
- Graphite
- Graphene