Partial Least Squares Structural Equation Modeling (PLS-SEM) in business research: A simple guide for novice researchers

Abstract

This review provides a comprehensive, step-by-step guide to the application of partial least squares structural equation modeling (PLS-SEM) for novice researchers. This is a conceptual and literature-based review that focuses on best practices and PLS-SEM literature. It highlights the rationale for using PLS-SEM, sample size, software tools, and essential metrics in PLS-SEM analysis. Drawing on best practices and recent literature, the review offers a framework for conducting and reporting PLS-SEM analysis. The review presents essential such as outer loadings, Cronbach’s alpha coefficients, average variance extracted (AVE), composite reliability, cross-loadings, Heterotrait-Monotrait ratio of correlations (HTMT), the Fornell-Larcker criterion, variance inflation factor (VIF), and redundancy analysis. Moreover, for more consistent results, the paper emphasizes on researchers to employ 10,000 bootstrap subsamples and Bias-corrected and accelerated (BCa) bootstrap in assessing the structural model. Insights regarding path coefficients, p-values, R-square (R2), f-square (f2), and Q-square (Q2), are also presented. Furthermore, the review underscores the trade-off between predictive power and model fit when applying PLS-SEM. The presented practical insights alert novice researchers in avoiding common pitfalls and enhance the methodological rigor of empirical research that utilizes PLS-SEM. This step-by-step guide supports early-career researchers and contributes to the ongoing debates on improving methodological clarity and transparency.