On the Use, Usefulness, and Ease of Use of Structural Equation Modeling in MIS Research: A Note of Caution

There is a need for caution in the application of structural equation modeling in the absence of well-developed substantive knowledge and the need for substantive knowledge to drive modeling, exploration, and interpretation of results.

Abstract

Structural equation modeling (SEM) using LISREL, EOS, PLS, or other second genration data analysis techniques is increasingly being applied in MIS research. These techniques are important because they provide powerful ways to address key IS research problems such as understanding IT usage. However, they may lead to inappropriate conclusions if statistical criteria are permitted to drive analysis and override substantive understanding of a problem. The purpose of this note is to suggest the need for caution in the application of structural equation modeling and, in particular, to emphasize the need for substantive knowledge to drive modeling, exploration, and interpretation of results. The application of SEM in the absence of well-developed substantive knowledge can lead to equivocal results and may distract researchers from promising research paths.