<![CDATA[Partial Least Squares Structural Equation Modelling (PLS-SEM) is widely utilised owing to its flexibility in handling complex models, small sample sizes, and non-normal data distributions. However, many empirical studies continue to rely on outdated or incomplete evaluation procedures, which can compromise parameter precision and lead to biased conclusions. This study aims to enhance parameter precision in PLS-SEM by proposing a comprehensive methodological refinement framework aligned with contemporary best practices. Using a systematic methodological review and conceptual synthesis of PLS-SEM studies published between 2014 and 2024, this research identifies recurring weaknesses in the assessment of measurement models, discriminant validity, structural models, and model fit indices. The findings demonstrate that parameter precision is substantially improved through flexible indicator assessment, rigorous discriminant validity testing, comprehensive structural model evaluation, and modern model fit diagnostics. The proposed refinement framework integrates updated thresholds and evaluation metrics across the outer model and inner model, emphasising predictive relevance and robustness over rigid cut-off rules. By consolidating these methodological improvements into a unified framework, this study bridges the gap between advanced methodological guidelines and prevailing empirical practice. The results provide clear guidance for researchers across disciplines in obtaining more accurate, reliable, and unbiased parameter estimates when applying PLS-SEM, thereby strengthening both theoretical inference and practical decision-making.]]>
Copyrights © 2026
