A structured review of partial least squares in supply chain management research. Issue 4 (December 2015)

• Record Type: Journal Article
• Authors: Kaufmann, Lutz; Gaeckler, Julia
• Is Part Of: Journal of purchasing and supply management. Volume 21:Issue 4(2015:Dec.)
• Page Start: 259
• Page End: 272
• Publication Date: 2015-12
• Subjects: Partial least squares -- Structural equation modeling -- Supply chain management; Industrial procurement -- Europe -- Management -- Periodicals; Purchasing -- Europe -- Periodicals; Purchasing -- Europe -- Management -- Periodicals; Materials management -- Europe -- Periodicals; Industrial procurement -- Management; Materials management; Purchasing; Purchasing -- Management; Europe; Periodicals; 658.7205
• DOI: 10.1016/j.pursup.2015.04.005
• Languages: English
• ISSNs: 1478-4092

Abstract

Abstract: The application of structural equation modeling (SEM) in the supply chain management (SCM) context has experienced increasing popularity in recent years. Although most researchers are well equipped with a basic understanding of the traditional covariance-based SEM (CBSEM) techniques, they are less familiar with the appropriate use of partial least squares (PLS) SEM. To fill this gap, the current paper critically reviews the use of PLS in 75 articles published in leading SCM journals from 2002 until 2013. The review indicates the potential of PLS, but also its limitations. A comparison across PLS reviews from various disciplines suggests that SCM research applies the same or even higher reporting standards in performing a PLS analysis and reporting the results than other disciplines (e.g., marketing or strategic management) that use PLS. However, SCM researchers often do not fully exploit the method's capabilities, and sometimes they even misapply it. This review thus offers guidelines for the appropriate application of PLS for future SCM research. Highlights: This article critically reviews the use of partial least squares in SCM research. A comparison across various research fields yields positive results for SCM research. However, the application of PLS and the interpretation of results can be improved. This article offers subject-specific guidance for the use of PLS in SCM research.