Hayes vs. Baron & Kenny.

Some info in mediation analysis according to Prof Hayes. Watch here http://www.youtube.com/watch?v=P0YdlBh1pqs

Thanks Prof Ramayah for sharing the insight in FB. Here's the link:  https://www.facebook.com/update_security_info.php?wizard=1#!/ramayah.thurasamy?hc_location=stream

My article Awang et al. (2007). Entrepreneurial orientation and effectiveness of small and medium agro-based enterprises (SMAEs) in Malaysia – Mediating effect of knowledge and network strategies  Social and Management Research Journal, 4(2), 1-21,  that utilized Preacher and Hayes (2004) in detecting mediation. The excerpt as follows: "... Mediating effect analysis was observed in a three-step regression analysis proposed in Baron and Kenny (1986). Mediator type was ascertain as suggested by Preacher and Hayes (2004). Estimation criteria suggested in equations as follows:

Y = i₁ + cX                             (1)

M = i₂ + aX                            (2)

Y = i₃ + c’X + bM                  (3)

According to Baron and Kenny (1986), four conditions to be observed in determining the mediation effect. First condition as in equation (1), the effect of X on Y denotes the total effect c. Second condition as in equation (2) the effect of X on M denotes the total effect a. Third condition as in equation (3) the effect of M on Y denotes the total effect of b. And, fourth condition as in equation (3) the indirect effect of X on Y denotes the total effect of c’. When the effect of X on Y decreases to zero with the inclusion of M, full mediation is said to have occurred (James & Brett, 1984). When the effect of X on Y decreases by a nontrivial amount, but not to zero, partial mediation is said to have occurred (p. 6-7).

Let's enter the dynamic nature of research world ...