Search results

The purpose of this paper is to discuss the pros and cons of partial least squares approach to structural equation modeling (PLS-SEM). The topics include bias, consistency, maximization of R², reliability and model validation.

The approach in this study is descriptive, and the method consists of logical arguments and analysis that are supported by results in references.

Several optimal properties of the PLS-SEM methodology are clarified. A proposal for transforming PLS-SEM mode A to mode B is highlighted, and the transformed mode possesses the desired properties of both modes A and B. Issues with the application of regression analysis using composite scores are also discussed. The strength of PLS-SEM is also compared against that of covariance-based SEM.

Additional studies on PLS-SEM are needed when the population structure contains cross-loadings and/or correlated errors.

PLS-SEM may have inflated type I errors and R² values even with normally distributed data.

The content of this paper is new, and there does not exist such an in-depth discussion of the pros and cons of PLS-SEM methodology in the literature.

This paper aims to discuss recent criticism related to partial least squares structural equation modeling (PLS-SEM).

Using a combination of literature reviews, empirical examples, and simulation evidence, this research demonstrates that critical accounts of PLS-SEM paint an overly negative picture of PLS-SEM’s capabilities.

Criticisms of PLS-SEM often generalize from boundary conditions with little practical relevance to the method’s general performance, and disregard the metrics and analyses (e.g., Type I error assessment) that are important when assessing the method’s efficacy.

We believe the alleged “fallacies” and “untold facts” have already been addressed in prior research and that the discussion should shift toward constructive avenues by exploring future research areas that are relevant to PLS-SEM applications.

All statistical methods, including PLS-SEM, have strengths and weaknesses. Researchers need to consider established guidelines and recent advancements when using the method, especially given the fast pace of developments in the field.

This research addresses criticisms of PLS-SEM and offers researchers, reviewers, and journal editors a more constructive view of its capabilities.

Rigorous applications of analytical tools in information systems (IS) research are important for developing new knowledge and innovations in the field. Emerging tools provide building blocks for future inquiry, practice and innovation. This article summarizes the findings of an analysis of the adoption and reporting of partial least squares structural equation modeling (PLS-SEM) analytical tools by Industrial Management & Data Systems authors in the most recent five-year period.

Selected emerging advanced PLS-SEM analytical tools that have experienced limited adoption are highlighted to broaden awareness of their value to IS researchers.

PLS-SEM analytical tools that facilitate understanding increasingly complex theoretical models and deliver improved prediction assessment are now available. IS researchers should explore the opportunities to apply these new tools to more fully describe the contributions of their research.

Findings demonstrate the increasing acceptance of PLS-SEM as a useful alternative research methodology within IS. PLS-SEM is a preferred structural equation modeling (SEM) method in many research settings and will become even more widely applied when IS researchers are aware of and apply the new analytical tools.

Emerging PLS-SEM methodological developments will help IS researchers examine new theoretical concepts and relationships and publish their work. Researchers are encouraged to engage in more complete analyses by applying the applicable emerging tools.

Applications of PLS-SEM for prediction, theory testing and confirmation have increased in recent years. Information system scholars should continue to exercise sound practice by applying these new analytical tools where applicable. Recommended guidelines following Hair et al. (2019; 2022) are included.

The purpose of this paper is to describe potential flaws and pitfalls in the contemporary process of testing the theory of a research model in business research through the use of covariance-based structural equation modeling (CB-SEM).

This paper offers a foundation for discussion, debate and questioning regarding the contemporary process of testing the theory of a research model in business research through CB-SEM.

The contemporary process to test theory of a research model through CB-SEM in business research lacks to a large extent a stepwise and iterative process of an accumulation of knowledge to build sound and rigorous business theory that is both reliable and valid over time as well as across contexts.

This paper provides an awakening toward further debate and discussion on the relevance and suitability of the contemporary process to test the theory of a research model through CB-SEM in business research – is it science, quasi-science or just nonsense?

The primary implication of this paper is that its content will challenge most readers ' preconceptions of the topic and stimulate debate. Subsequently, it is the author’s hope that the content is thought-provoking and counterintuitive. Some scholars might reject the content, while others may find it valuable.

The paper intends to provide counterintuitive thoughts regarding the contemporary process of testing the theory of a research model in business research through the use of CB-SEM. CB-SEM offers potentially valuable merits in business research settings, if applied and performed properly.

Tourism research contains a large share of consumer behavior-orientated studies using multidimensional constructs (exogenous/endogenous). Accordingly, scholars have mainly made use of a two-step approach that can be referred to as PCA-MLR (principal component analysis and then ordinary least squares multiple linear regression analysis) to examine the relationships among exogenous and endogenous constructs in a statistical model. Although this two-step approach has contributed to the advancement of tourism research, it still suffers from a number of drawbacks which can readily be overcome by a so-called second-generation statistical tool, namely, partial least squares approach to structural equation modeling (PLS-SEM). The current chapter explains and illustrates (with an application to tourism data) the advantages (e.g., several layers of estimations, suiting small sample sizes, robustness to multicollinearity, model-based clustering, etc.) of PLS-SEM both from a statistical and practical point of view. Finally, an elucidation is also provided for suggesting PLS-SEM as an alternative to PCA-MLR instead of COV-SEM (covariance-based structural equation modeling). The chapter concludes by proposing that PLS-SEM is a reliable and flexible statistical approach that is of high value, in particular, for applied research.

In recent years, the use of structural equation modeling (SEM) has become widespread in tourism and hospitality research. Because there are two different approaches to SEM (i.e., covariance-based SEM and variance-based, partial least squares SEM), this brings challenges for researchers about which SEM to use and what to report in each SEM approach. Therefore, the purpose of this chapter is to discuss the differences between CB-SEM and PLS-SEM and to provide comprehensive guidelines for researchers on how to apply each SEM. Within this context, the authors first briefly summarize the fundamentals and advantages of using SEM. Then, the authors explain in detail the major issues that should be considered when selecting between CB-SEM and PLS-SEM. Finally, to ensure rigorous research practices, the authors provide step-by-step guidelines for the application of both CB-SEM and PLS-SEM.

Structural equation modeling (SEM) is a powerful multivariate statistical technique that requires careful application. The use of SEM in international business research has substantially increased recently, necessitating a critical evaluation of its use in the field. Through an analysis of 148 articles in the international business (IB) literature, we detail the state of current use of SEM in IB research and compare its use to the established best practices. In many instances, SEM's use in IB has been faulty, suggesting that authors may have drawn incorrect conclusions. To expand the IB field's knowledge base, methodological accuracy is essential. Based on our review of the technique's use in IB research coupled with the established practices in the social science literature, we provide practical suggestions for better applying SEM in the IB literature.

Although recent decades have seen increasing calls for fundamental change in the teaching of Science, Engineering, and Mathematics (SEM), efforts to more broadly propagate proven innovations have met with only modest success despite (i) numerous national reports calling for changes, (ii) considerable funding that has been invested in SEM education improvements, and (iii) the growing body of literature on the superior efficacy of many curricular innovations. This chapter suggests that SEM innovators, while expert in their fields, may need to thoughtfully consider research and literature on change, both within higher education and including broader work on organizational change. From a review of the literature on change in higher education, two particular challenges are identified: goal ambiguity and narrow focus of change initiatives. To address these challenges, the authors offer a conceptual framework for decisions that SEM educational change agents make as they design and implement their change initiatives. Within this framework, they offer options and combinations of options that change agents might consider. Given the breadth and complexity of the literature and challenges of change, SEM educational change agents might consider forming collaborations to which they would contribute their disciplinary expertise in one of the three research communities. They might team with individuals who bring requisite expertise from other research communities or with respect to individual and organizational change. Such partnerships might develop approaches that would concurrently address multiple foci. Collaborations that included expertise in individual and organizational change would also be better prepared to navigate complexities of institutional change.

This chapter reviews the methods available to hospitality and tourism researchers to perform moderation analysis with continuous variables in partial least squares structural equation modeling (PLS-SEM), with the objective of enhancing understanding and encouraging the use of these techniques in future papers. The product term method is presented first, followed by an empirical example/application in the context of hospitality and tourism. Two extensions, namely the two-stage approach that can help cope with formative and higher-order constructs, and the orthogonalizing approach that can help generate more accurate results and overcome multicollinearity among tourism variables in the presence of a continuous moderator variable, are then presented and discussed. The chapter concludes by presenting guidelines and recommendations for improving the use of interaction effects in analyses of tourism variables, as well as highlighting ongoing developments in both the product term method and PLS-SEM software.