Mitigating Common Method Bias in Construction Engineering and Management Research

Abstract

The importance of reliability in conducting empirical research has been emphasized by construction engineering and management (CEM) scholars. Survey data commonly are used in empirical CEM studies. However, the use of such data in empirical analysis is susceptible to common method bias (CMB) issues, which can generate false correlations between dependent and explanatory variables, leading to spurious and unreliable results. To facilitate recent discussion about the rigorous application of research methodologies in CEM research, we provide a comprehensive overview of major CMB sources and examine articles published in the last 5 years to identify potential CMB issues in the Journal of Construction Engineering and Management. We found that approximately 90% of articles that used survey data in the empirical analysis were likely to be affected by CMB. The major source is the common rater effect, followed by the item characteristics effect and the measurement context effect. Because only a handful of studies tested or controlled for CMB, we propose effective and targeted ex ante procedural controls and ex post statistical controls to mitigate its negative impact. This study sheds light on an important and pervasive methodological issue in empirical CEM studies, and provides guidelines for improving the reliability of future empirical research in the CEM field.