Testing Two Generations of SEM
Abstract
This project compares classic structural equation modeling (SEM) with a newer computational-graph approach (cgSEM). Using simulations and an empirical dataset, we examine how the two frameworks differ in convergence behavior, admissible-solution rates, bias, model fit, and runtime across a range of modeling conditions, and identify when they yield practically similar results versus when their estimates diverge. All code and supporting materials are included to facilitate reproduction and extensions of the work.
Fields of Study & Subjects
Psychometric Methodologies and Testing
Model-Driven Software Engineering Techniques
Model Reduction and Neural Networks
Publication details
- Journal / source
- OSF Preprints (OSF Preprints)
- Publication date
- 1 January 2026
- Document type
- Preprint