Wednesday 29 October 2014

Common Method Variance

It is quite a days that i did not up date my blog even there are few interesting questions have been asked by my colleagues... today i would like to share one of the most rare and interesting question... the explanation is quite long since a took a longer time to convince my friends about that...

It is about a common method variance.. my friend was sent his paper to the journal.. which is well known journal that i cant mention here.. one of the reviewer argued that.. since he used a single source data.. a common method variance (CMV) has not been addressed in his paper... how should this CMV should be address and write in the paper????


Before i answer the question.. to all the researchers that would like to send their writing to the good journal, this issue must be addressed if you are using a single source data... the CMV firstly been addressed by podsakoff & Morgan (1986), (2003) and in the 2011, one of the editor also in a good journal mentioned that, he will not look at all for the paper who are failed to address the CMV if they are using the single source data..... wahhhh.. that serius... ermm.. it shows how importance this issue must be overcome before, and during the writing process...

According to the Podsakoff, (1986) (2003), CMV is due to the way that the data was collected.

1) Using  a single source data for the IV and DV
2) Using the same scale for the IV and the DV
3) Provide the questions in survey according to the title of the section...

So, how to overcome????
Podsakoff also mentioned in that paper how to overcome this problem..
1) dual sources... which is quite impossible if you run your study in Malaysia..
2) Apply diffrent scale for the iv and dv..such as 5 scale for IV and 7 for the DV
3) Jumble up the questions.. so the respondents dont know what are the measuring ....

the last one... how to write... this is the reason why this blog is develop for...

Since the study use a single source data, there is a potential for common method variance. However, the Harman single factor test was conducted to determine the extent of this bias. Podsakoff and Organ (1986) mentioned that common method bias is problematic if a single latent factor would carry the majority of the explained variance. The un-rotated factor analysis indicated that the first factor explained 42.2% of the total variance explained, and thus the common method bias is not a serious issue in this study.

Ngah et al, 2014 Barriers and enablers in adopting Halal transportation services:
A study of Malaysian Halal Manufacturers.  International Journal of Business and Management Vol. II (2), 2014 61..

http://www.iises.net/wp-content/uploads/pp-49-70_ijobmV2N2.pdf

No comments:

Post a Comment