Kent State University
Question
Asked 18 November 2019
Should mediation analysis be conducted if direct relationship is insignificant?
Hi,
In my research, the direct relationship between IV and DV is insignificant. However, when I add a mediator, between the IV and DV, the indirect relationship becomes significant. According to Baron and Kenny (1986 ), four steps, the direct relationship between the significant relationship between IV and DV should exist. However, in my case, the results of IV and DV do not exhibit the significant relationship. Do you think there is a justification for mediation if a direct relationship does not exist? Note that the indirect relationship proves to be highly significant. If it can be done, how can it be interpreted and justified? Any source which can help will be highly appreciated.
Most recent answer
In addition, the facts from Experimental Designs seem to support what what Chaudry Bilal Ahmad Khan and Archit Vinod Tapar and @Cristian Ramos-Vera report since in a factorial model interactions can exist without corresponding main effects, See the link:
The material on factorial designs. Best, David Booth
All Answers (4)
Indian Institute of Management, Rohtak
You may still conduct the mediation analysis if the direct relationship is insignificant. You may refer to the following paper pg. 199 of Zhao, Lynch & Chen 2010 which states "There need not be a significant zero-order effect of X on Y, rXY, to establish mediation."
For further interpretation and deeper understanding you may refer:
Zhao, X., Lynch Jr, J. G., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truths about mediation analysis. Journal of consumer research, 37(2), 197-206.
2 Recommendations
Institute of Space Technology
Thank you, everyone, for your suggestions. I have also found an article that suggests that mediation analysis can be conducted even if there is no significant direct relationship between IV and DV. However, instead of mediation, the indirect effect has been used as there is no significant direct relationship between IV and DV. For your reference and anyone who reads this question, please read the article below.
Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication monographs, 76(4), 408-420.
1 Recommendation
Kent State University
In addition, the facts from Experimental Designs seem to support what what Chaudry Bilal Ahmad Khan and Archit Vinod Tapar and @Cristian Ramos-Vera report since in a factorial model interactions can exist without corresponding main effects, See the link:
The material on factorial designs. Best, David Booth
Similar questions and discussions
Related Publications
This study compares various methods of mediations analysis. Firstly, it compares the two methods of calculating indirect effect which are product of coefficient and difference of coefficients respectively. Secondly, the study compares the three methods of testing the significance of indirect effect vis avis Sobel's test, Aroian test and Goodman's t...