Robustness checks in PLS-SEM: A review of recent practices and recommendations for future applications in business research

Abstract

For the last decade, scholars have employed partial least squares structural equation modeling (PLS-SEM) extensively in business research. However, when applying PLS-SEM, researchers need to perform various robustness checks before and after model estimation. This study showcases the findings of a review of PLS‐SEM use in business research, by examining papers published between 2016 and 2021 in business journals. The study explores the extent to which researchers have performed robustness checks regarding nonnormality, endogeneity, unobserved heterogeneity, nonlinearity, and heteroskedasticity. The findings highlight that statistical rigor remains a serious problem in business-related studies employing PLS-SEM. Despite some encouraging improvements in the last few years, the vast majority of recent business-related studies using PLS-SEM have systematically overlooked robustness checks. This study calls for continued emphasis on the importance of robustness checks and the correct application of appropriate techniques, providing recommendations and guidelines for future PLS-SEM applications.