Abstract
The procedure setfor mediafion analysis in the Baron and Ken- ny's (1986) classical research article did not go without criti- cism (MacKinnon and Fairchild, 2009; Hayes, 2009; Bullock, Green,
& Ha, 2010; Zhao, Lynch and Chen, 201o), and thanks to those crifics, that Kenny (2012) had to bring a number of modificafions and improvements through his today's contem- porary mediafion analyfic procedure. This paper presents a comparison of what Baron and Kenny (1986) had originally proposed, and what Kenny (2012J has now suggested afier incorporafing crifics’ concerns. For this purpose, the two ap- proaches of mediafion analysis have been pracfically applied on an organizafional jusfice - trust in supervisor - employees’ job safisfacfion case, wherein variable ‘trust in supervisor’ was used to mediate between various facets of organizafion jusfice and
employees’job satisfacfion. Whereas the classic approach required the esfimafion of the four paths (c, a, b & c’), through four steps and running three regression equa- fions and tesfing for their
statisfical significance; the contem- porary mediafion analysis has shortened the paths from four to three (a, b & c’), requiring taking only two steps (Steps 2 and 3), and running two regression equafions, declaring the first equafion as unnecessary. The classic approach required that path b needs to become stafisfically significant and path c’ insignificant in step 3 & 4 for a complete mediafion; con- temporary approach asks for adding c’ with ab for determin- ing total effect c, and
then decomposing c in to direct effect (c’/c) and indirect or mediafion effect (ab/c). Whereas classic approach aimed at solving for full or parfial mediafion in ab- stract form, the contemporary approach has the edge over the classic, in quanfifying the mediafion effect (ab). In case of our solved example, classic approach could only help to indi- cate that ‘trust-in-supervisor variable is parfially mediafing’, while in case of contemporary approach mediafion effect of this variable for INJ facet of organizafional jusfice was not only quanfified (17.36%) and tested for its non-zero effect, but zero-effect of other two facets, DJ and IJ, were also dif- ferenfiated.JEL Classification:C01, C12, C13, C18
Ahmadullah Shah,, Shujahat Haider Hashmi , Anwar F. Chishti. (2013) Much has changed since Baron and Kenny's (1986) classic paper: let us learn what Kenny's (2012) con- temporary mediation analysis prescribes, Jinnah Business Review, Volume 1, Issue 1.
-
Views
448 -
Downloads
38
Previous Article
