With Great Power Comes Great Responsibility

Abstract

Partial least squares (PLS) offers multiple advantages as a composite-based structural equation modeling (SEM) technique. PLS enables scholars to examine the measurement model and structural model simultaneously and often requires fewer assumptions than factor-based SEM techniques. For these reasons and more, PLS offers great power for researchers who wish to use a SEM-based approach to evaluate a research model. However, with the great power of PLS also comes great responsibility. Scholars should determine if PLS is appropriate to use within their context, and scholars should explain their rationale for employing PLS for data analysis. Recognizing the power and responsibility associated with PLS is important since many scholars have called for an abandonment of PLS within the information systems discipline and beyond. We reviewed articles from four premier journals within the information systems field from 2017-2020 that use PLS as an analysis technique. Based on this review, we identify recommendations for scholars seeking to embrace the power and responsibility of using composite-based SEM to analyze research models.