Idiographic Networks: A Tutorial on Graphical Vector Autoregression and Unified Structural Equation Modeling

Abstract

This tutorial introduces the application of advanced network analysis methods, specifically Graphical Vector Autoregression (graphicalVAR) and Unified Structural Equation Modeling (uSEM), to model learning processes as complex, dynamic systems. These approaches allow exploring both temporal and contemporaneous relationships among variables within individual learners over time. The chapter begins by conceptualizing learning as a networked system and reviewing relevant literature, discussing the advantages of probabilistic network models in education. Then, a step-by-step tutorial in the R programming language is presented so readers can learn to estimate idiographic models, visualize dynamic relationships, and interpret. As such, this tutorial aims to provide researchers with tools to analyze multivariate time-series data, which is a necessary step for truly personalized interventions in educational research © 2026 The Editor(s) (if applicable) and The Author(s).