Abstract
The structural equation modelling (SEM) technique has been touted as a useful tool for tightening links between theoretical and empirical operations management (OM) research. Despite SEM's increasing prominence in the field, leading scholars continue to call for a deeper infusion of theory into empirical OM research. To strengthen ties between theory and analysis in OM research, this study evaluates previous OM applications of SEM and identifies specific ways we can use SEM to advance operations management theory. Through judicious use of SEM techniques, we believe that OM researchers have the opportunity to confirm and extend existing theoretical frameworks. Further, we offer guidance on how to operationalise measurement models such that researchers accurately depict the causality of a construct. To demonstrate how to advance theory, we use an illustrative example of SEM in an OM context based upon data gathered from a survey of over 200 respondents.
Notes
Notes
1. Appendix 1 provides the details of our literature review methodology.
2. While it is important to appropriately conceptualise constructs as either formative or reflective, we would like to note that formative representations are fraught with problems like interpretational confounding and external consistency (Howell et al. Citation2007). Discussion of these issues is beyond the scope of this paper, but interested readers are referred to Howell et al. on the care needed for good formative representations.
3. Indicators of formative constructs may be referred to as cause, causal, composite, or formative indicators. We use the term formative indicators simply for consistency. For a detailed review of issues related to formative issues please see the Journal of Business Research volume 61, issue 12, special issue on formative indicators (2008).
4. PLS can be acquired from a variety of sources. VisualPLS is available at http://fs.mis.kuas.edu.tw/~fred/vpls/. SmartPLS is available at http://www.smartpls.de/forum/. PLSGraph can be purchased from Wynne Chin–[email protected].