NORED925: Structural equation modeling: An introduction

The main goal of the course is to introduce PhD students to the theoretical foundations and practical applications of Structural Equation Modeling (SEM) as a framework for testing complex theoretical relationships in the social and educational sciences. Throughout the course, the PhD students will develop their understanding of how latent variables and structural models can be used to integrate measurement theory with theory building. In addition, the course aims to develop the PhD students’ competence in designing, estimating, and evaluating SEMs using RStudios (no prior coding experience needed). Students will learn to interpret model parameters and fit indices critically, with particular emphasis on the limitations of statistical significance testing and the responsible use of fit criteria in empirical research.

By the end of the course, PhD students should be able to apply SEM thoughtfully in their own research, communicate results transparently, and evaluate published SEM studies with a high degree of methodological literacy. 

Knowledge

The student is able to…

• explain the theoretical foundations and basic principles of structural equation modeling (SEM), including measurement models, structural models, and model identification.
• describe the statistical logic of model estimation and fit assessment in SEM, including the role of χ², p-values, and common fit indices (CFI, TLI, RMSEA).
• account for the limitations, assumptions, and potential sources of bias in SEM, such as sample size sensitivity, model complexity, and specification error.

Skills

The student…

• is able to design and estimate a structural equation model using appropriate software, specify latent constructs, and interpret model parameters and path coefficients.
• is able to interpret and critically evaluate model fit indices and significance tests, assessing both statistical and substantive model quality.
• is able to identify and discuss common misinterpretations of significance and fit statistics in published SEM research, and propose justified alternatives for improved reporting and transparency.

General competence

The student…

• is able to critically assess the use of SEM in research, integrating statistical results with theoretical reasoning and practical implications.
• is able to communicate SEM results clearly and responsibly, demonstrating methodological awareness and adherence to good scientific practice.
• is able to reflect on the limitations of statistical significance and model fit criteria and use these reflections to inform responsible model evaluation and reporting in academic contexts.