Controlling for Method Bias: A Critique and Reconceptualization of the Marker Variable Technique

The marker variable technique is an easy-to-use technique for estimating the magnitude of method bias within a study. However, its validity has not yet been established. This paper addresses three issues assessing the validity of the technique and finds that it is subject to significant validity threats. A redefinition of the marker variable correlation is proposed, which partly addresses the theoretical critiques of the technique. The findings confirm Podsakoff et al.’s (2003) critique that the marker variable technique does not capture key sources of method bias. Implications of the findings for estimating and controlling for method bias within individual studies are addressed. Disciplines Physical Sciences and Mathematics Publication Details Yetton, P., Sharma, R. & Crawford, J. (2011). Controlling for method bias: a critique and reconceptualization of the marker variable technique. 17th Americas Conference on Information Systems (pp. 1-9). United States of America: AMCIS. This conference paper is available at Research Online: http://ro.uow.edu.au/infopapers/3852 Yetton et al. Critique and Re-conceptualization of the Marker Variable Technique Proceedings of the Seventeenth Americas Conference on Information Systems, Detroit, Michigan August 4-7 2011 1 Controlling for Method Bias: A Critique and Reconceptualization of the Marker Variable Technique Philip Yetton Australian School of Business University of New South Wales Sydney NSW, 2052, Australia [email protected] Rajeev Sharma University of Wollongong Wollongong NSW 2522, Australia [email protected] Jeff Crawford Collins College of Business The University of Tulsa Tulsa, OK 74104, USA [email protected] ABSTRACT The marker variable technique is an easy-to-use technique for estimating the magnitude of method bias within a study. However, its validity has not yet been established. This paper addresses three issues assessing the validity of the technique and finds that it is subject to significant validity threats. A redefinition of the marker variable correlation is proposed, which partly addresses the theoretical critiques of the technique. The findings confirm Podsakoff et al.’s (2003) critique that the marker variable technique does not capture key sources of method bias. Implications of the findings for estimating and controlling for method bias within individual studies are addressed.The marker variable technique is an easy-to-use technique for estimating the magnitude of method bias within a study. However, its validity has not yet been established. This paper addresses three issues assessing the validity of the technique and finds that it is subject to significant validity threats. A redefinition of the marker variable correlation is proposed, which partly addresses the theoretical critiques of the technique. The findings confirm Podsakoff et al.’s (2003) critique that the marker variable technique does not capture key sources of method bias. Implications of the findings for estimating and controlling for method bias within individual studies are addressed.