Beware of the Woozle effect and belief perseverance in the PLS-SEM literature!

Abstract

Partial least squares structural equation modeling (PLS-SEM) is a popular method of data analysis in marketing, information systems research, and related fields. Unfortunately, the literature on PLS-SEM contains a number of misstatements that ascribe characteristics to PLS-SEM that it does not possess. In our study, we consider why these misstatements continue to be made and reinforced. We show how inaccuracies, omissions, repetitions of past misconceptions, and the introduction of additional misconceptions lead to the so-called Woozle effect. As an example, we use perhaps the most serious misconception about PLS-SEM namely its alleged suitability for estimating the parameters of reflective measurement models. The Woozle effect is a cumulative process by which falsehoods become established as fact, and the self-correcting mechanisms of science are suspended. Through a literature review, we identify a number of factors that are likely to have contributed to the Woozle effect in the PLS-SEM literature. For the Woozle effect to disappear, researchers need to acknowledge that PLS-SEM, in its current incarnation, is not suitable for scientific work with reflective measurement models.