From Foundations to Frontiers: the Development and Future of Structural Equation Modeling (sem)
Year: 2025
Type: article
Abstract: Structural Equation Modeling (SEM) is a versatile statistical framework for modeling complex relationships among observed and latent variables. This narrative review traces SEM's development from its foundational roots in factor and path analysis to its current integration with machine learning, Bayesian techniques, and big data analytics. We critically examine SEM’s methodological advancements, including Confirmatory Factor Analysis (CFA), Multi-Group SEM, Bayesian SEM, and Dynamic SEM, and highlight its diverse applications across psychology, education, business, and health sciences. Emphasis is placed on current challenges—such as data quality, sample size, and model mis-specification—and future opportunities enabled by open-source tools and AI integration. This review’s objective is explicitly aligned with the article’s purpose: to synthesize SEM’s historical development, current methodologies, challenges, and future directions into practical guidance for researchers using SEM in increasingly complex data environments. This paper provides a comprehensive synthesis for both novice and experienced researchers, offering guidance on using SEM effectively in an era of increasingly complex data environments. This review benefits readers by providing practical insights into applying SEM in their current research, guiding them to select suitable model-fitting tools and interpret results effectively.
Source: UMT-Poly Journal
Language: Thai
Field: Decision Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W7219386253
