Data Analytics for Organizational Development – From Regression to Structural Equation Modeling

Abstract

Abstract Structural equation modeling has been widely used by academic researchers and was rarely seen in other organizations. However, since data analytics has got a foothold in a rapidly increasing number of public and private organizations all over the world, advanced analytics techniques such as structural equation modeling (SEM) for making critical decisions are needed. Although we see SEM being introduced into practice here and there, traditional approaches like regression are still in the majority, even when SEM could help giving more educated advice to business leaders. Part of the reason is that SEM is not easily available and is more complex as compared to using regression methodology. This study aims to show the benefits of using SEM to help practitioners in using data for decision‐making and for organizational development. We compare the application of SEM and multiple linear regression and explain the condition for using either approach when analyzing organizational behavioral research data. Our study is based on a set of such data collected at public and private organizations in Southeast Asia. The discussion focuses on the practical application of multiple linear regression and structural equation modeling as a valuable input for organizational development.