Partial Least Squares Structural Equation Modeling Applications

Partial least squares structural equation modeling (PLS-SEM) has emerged as a versatile multivariate technique for estimating complex cause-and-effect relationship models, especially when theoretical constructs are represented by multiple indicators. By combining principal component analysis with regression, PLS-SEM accommodates both measurement and structural model assessment, enabling simultaneous estimation of latent variable scores and path coefficients. Unlike covariance-based SEM, it makes minimal demands on sample size and distributional assumptions, rendering it well suited to exploratory research, prediction-oriented studies and contexts with formative indicators. Over the past decade, methodological refinements such as consistent PLS algorithms, bootstrap-based global fit tests and heterotrait-monotrait ratio assessments have strengthened its rigor and broadened its appeal across disciplines. Practitioners have applied PLS-SEM to fields as diverse as information systems, marketing, consumer behaviour, environmental science and public transport, often integrating it with complementary techniques to uncover antecedents and mediators of complex phenomena. Its predictive orientation has proved valuable for developing actionable insights in digital innovation, service quality management and policy evaluation. Recent methodological work has addressed interaction modelling, multigroup analysis and the assessment of composite constructs, establishing PLS-SEM as a robust framework for both confirmatory and exploratory investigations with practical significance.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Recent studies in information systems research have provided comprehensive guidelines for rigorous PLS-SEM applications in confirmatory and explanatory contexts. These guidelines outline the latest algorithmic enhancements, such as consistent PLS estimators for latent variables, improved global goodness-of-fit diagnostics and recommended reporting standards for path coefficients, composite reliabilities and validity measures. In business marketing management, a critical appraisal of two decades of PLS-SEM use highlighted recurring shortcomings in measurement assessment and structural reporting, and proposed best-practice checklists to strengthen model evaluation, encourage adoption of advanced analytic options and ensure transparent documentation of sample size justification, threshold criteria and robustness checks. Another strand of work has illustrated the integration of PLS-SEM with necessary condition analysis and fuzzy-set qualitative comparative analysis to deliver richer, configuration-based insights. By combining these methods, researchers have identified both predictive correlations and bottleneck conditions in service quality studies, demonstrating how the expanded analytical workflow can inform strategic decision-making in sectors such as urban transit and consumer services.

Partial Least Squares Structural Equation Modeling Applications publication trend

The graph below shows the total number of articles in partial least squares structural equation modeling applications across all publications each year (not limited to Nature Index journals).

Technical terms

Partial least squares structural equation modeling (PLS-SEM): A component-based estimation approach that models relationships among latent constructs and manifest indicators, prioritising prediction and accommodating small samples and non-normal data.

Latent variable: An unobserved theoretical construct represented indirectly by multiple measured indicators, used to capture underlying phenomena such as satisfaction or innovation capability.

Composite-based model: A modelling approach in which constructs are formed as weighted linear combinations of their indicators, reflecting emergent phenomena rather than common factors.

Consistent PLS algorithm: An enhanced estimation procedure that provides asymptotically unbiased parameter estimates and enables proper assessment of global model fit in PLS-SEM.

Bootstrap method: A resampling technique used to generate empirical sampling distributions of path coefficients and validate their statistical significance without strict distributional assumptions.

References

1. Unlocking potential: An integrated approach using PLS-SEM, NCA, and fsQCA for informed decision making. Journal of Retailing and Consumer Services (2023).
2. How to perform and report an impactful analysis using partial least squares: Guidelines for confirmatory and explanatory IS research. Information & Management (2020).
3. CB-SEM vs PLS-SEM methods for research in social sciences and technology forecasting. Technological Forecasting and Social Change (2021).
4. Consistent and asymptotically normal PLS estimators for linear structural equations. Computational Statistics & Data Analysis (2015).
5. ESTIMATING MODERATING EFFECTS IN PLS-SEM AND PLSc-SEM: INTERACTION TERM GENERATION*DATA TREATMENT. Journal of Applied Structural Equation Modeling (2018).
6. Improving PLS-SEM use for business marketing research. Industrial Marketing Management (2023).