Hello,

Perhaps because statistically non-significant direct effects should not be investigated via mediation?

Regards

Arjun, this is generally expected: If X is correlated with M, and M is correlated with Y, then X should be correlated with Y. However, this is not always the case. In the following article the authors show why this is the case:

Rucker, D. D., Preacher, K. J., Tormala, Z. L., & Petty, R. E. (2011). Mediation analysis in social psychology: Current practices and new recommendations. Social and Personality Psychology Compass, 5/6, 359-371.

http://quantpsy.org/pubs/rucker_preacher_tormala_petty_2011.pdf

There is another article which may be of interest for you:

Zhao, Lynch, and Chen (2010). Reconsidering Baron and Kenny: Myths and Truths about Mediation Analysis. Journal of Consumer Research, 37 (2), 197-206.

http://dukespace.lib.duke.edu/dspace/bitstream/handle/10161/4143/279443600001.pdf?sequence=1

These authors present present a nontechnical summary of the flaws in the Baron and Kenny logic, some of which have not been previously noted.

Regards, Karin

Thanks Karin. I'll go through these articles. 

Regards, 

Arjun 

You can find in detail your answer in recently published paper on mediating analyhsis,

Article Making Sense of Mediating Analysis: A Marketing Perspective

Thanks Noor for sharing the article.