Testing moderating effects in PLS path models with composite variables

Georg Fassott, Jörg Henseler, Pedro S. Coelho

Research output: Contribution to journalArticleAcademicpeer-review

44 Citations (Scopus)

Abstract

Purpose The purpose of this paper is to explain how to model moderating effects of composites using partial least squares (PLS) path modeling. It provides the methodological underpinning of moderating effects in general and describes the various approaches for forming the interaction term, i.e., the product indicator approach, the two-stage approach, and orthogonalization. Design/methodology/approach The paper discusses the use of standardized vs unstandardized construct scores and introduces spotlight analysis as a useful way to report findings. Findings Researchers should rely on unstandardized estimates when analyzing moderating effects. Centering or orthogonalization can help improve the interpretability of path coefficients. Practical implications PLS software implementations should facilitate unstandardized estimates. Originality/value This paper formulates step by step guidelines for analyzing moderating effects of composites using PLS path modeling. It is the first to propose spotlight analysis for PLS path modeling.
Original languageEnglish
Pages (from-to)1887-1900
JournalIndustrial management & data systems
Volume116
Issue number9
DOIs
Publication statusPublished - 2016

Keywords

  • IR-101907
  • METIS-318585

Fingerprint Dive into the research topics of 'Testing moderating effects in PLS path models with composite variables'. Together they form a unique fingerprint.

  • Cite this