Beyond explanation: the evolving role of PLS-SEM in marketing analytics

Hair, Joseph F., Ringle, Christian M., Sarstedt, Marko
Journal of Marketing Analytics · Springer Nature · 2026-11-06 · DOI: 10.1057/s41270-026-00565-0

Аннотация

Partial least squares structural equation modeling (PLS-SEM) has developed into a broad methodological framework for marketing analytics while retaining its longstanding causal-predictive orientation. This editorial introduces a curated collection of ten articles published in the Journal of Marketing Analytics that illustrate how methodological advances extend the evidence generated by PLS-SEM-centered research. We organize the collection around six analytical stages: estimation, inference, prediction, necessity, prioritization, and robustness. The articles demonstrate how researchers can account for model selection uncertainty, examine conditional mediation, assess out-of-sample predictive power, distinguish sufficiency from necessity, and integrate importance, performance, and necessity to inform managerial priorities. They also illustrate the requirements for comparing structural relationships across groups and time. We argue that methodological progress lies not in accumulating analytical procedures but in aligning research questions, designs, analyses, and substantive claims. This perspective guides researchers toward more informative and transparent PLS-SEM applications that connect theoretical reasoning with predictive assessment and managerial relevance.

Читать у издателя →

Похожие статьи