Rethinking some of the rethinking of partial least squares

Abstract

Purpose Partial least squares structural equation modeling (PLS-SEM) is an important statistical technique in the toolbox of methods that researchers in marketing and other social sciences disciplines frequently use in their empirical analyses. The purpose of this paper is to shed light on several misconceptions that have emerged as a result of the proposed “new guidelines” for PLS-SEM. The authors discuss various aspects related to current debates on when or when not to use PLS-SEM, and which model evaluation metrics to apply. In addition, this paper summarizes several important methodological extensions of PLS-SEM researchers can use to improve the quality of their analyses, results and findings. Design/methodology/approach The paper merges literature from various disciplines, including marketing, strategic management, information systems, accounting and statistics, to present a state-of-the-art review of PLS-SEM. Based on these findings, the paper offers a point of orientation on how to consider and apply these latest developments when executing or assessing PLS-SEM-based research. Findings This paper offers guidance regarding situations that favor the use of PLS-SEM and discusses the need to consider certain model evaluation metrics. It also summarizes how to deal with endogeneity in PLS-SEM, and critically comments on the recent proposal to adjust PLS-SEM estimates to mimic common factor models that are the foundation of covariance-based SEM. Finally, this paper opposes characterizing common concepts and practices of PLS-SEM as “out-of-date” without providing well-substantiated alternatives and solutions. Research limitations/implications The paper paves the way for future discussions and suggests a way forward to reach consensus regarding situations that favor PLS-SEM use and its application. Practical implications This paper offers guidance on how to consider the latest methodological developments when executing or assessing PLS-SEM-based research. Originality/value This paper complements recently proposed “new guidelines” with the aim of offering a counter perspective on some strong claims made in the latest literature on PLS-SEM. It also clarifies some misconceptions regarding the application of PLS-SEM.

The Roles of The Physical Environment, Social Servicescape, Co-Created Value, and Customer Satisfaction in Determining Tourists’ Citizenship Behavior: Malaysian Cultural and Creative Industries

“…To test the convergent validity (defined as "the degree to which a measure is positively correlated to alternative measures of the same construct"), the average variance extracted (AVE) was used in this study. This is an indication that all the values of the AVE, ranging from 0.556 to 0.934, were higher, compared to the recommended value of 0.50 [104,105]. For all the constructs, convergent validity was met successfully, and an adequate convergent validity was achieved, as presented in Appendix A.…”

“…The conceptual model was then analyzed with the Smart-PLS 3.2.8 software [103]. PLS-SEM data analysis was done through the two-stage technique and this present research utilized this technique recommended by [104,105]. Firstly, the measurement model was assessed to check the construct reliability and validity for indicator reliability and internal consistency.…”

“…In terms of the construct reliability, the composite reliability (CR) was used by this study to test the construct reliability, which gave the recommended value. Therefore, the values obtained, which ranged from 0.705 to 0.947, were more than 0.70 [104,105]. This is an adequate signifier that constructs reliability was achieved, as shown in Appendix A.…”

“…To test the reliability indicator, factor loading was checked. High loadings on a construct are indicators that the associated indicators appear to have a lot in common, in that the construct was able to capture them [104,105]. For factor loadings, values higher than 0.50 were classified as very significant [104,105].…”

“…High loadings on a construct are indicators that the associated indicators appear to have a lot in common, in that the construct was able to capture them [104,105]. For factor loadings, values higher than 0.50 were classified as very significant [104,105]. As seen in Appendix A, the loadings for all the items were more than the suggested value of 0.5, except for some items like (SFL7 = 0.325, SSA6 = 0.241, PAT5 = 0.231, MF3 = 0.221), which were lower than 0.50, and were therefore dropped due to the low loading (see Appendix A).…”