Evaluating Textbooks Through Concept Network Analysis

Textbooks serve as the primary basis for instructional activities, with their knowledge organization directly shaping teaching and learning pathways. We establish a framework for quantitative textbook evaluation based on concept network analysis and apply it to five elementary mathematics textbooks. Concept networks are constructed from each textbook, their topological properties are analyzed, and an information spreading model is applied to evaluate learning efficiency. Three networks exhibit high modularity and long path lengths, with influential nodes centered on observation-based learning and computational procedures. The other two exhibit high density, short path lengths, and high clustering, with influential nodes centered on general mathematical thinking such as abstraction and deductive reasoning. The spreading model demonstrates that textbooks organized around advanced knowledge achieve substantially higher efficiency. These findings suggest concept network analysis can serve as an effective tool for textbook evaluation.