REVISITING THE MODELS OF STATISTICAL KNOWLEDGE FOR TEACHING: WHAT MATHEMATICS TEACHERS NEED TO KNOW FOR K-12 DATA SCIENCE EDUCATION

Recent discussions in statistics education emphasize the importance of supporting reasoning about variability and uncertainty when data are interpreted and used for judgment. In such contexts, teachers' statistical knowledge involves not only the correct use of procedures but also the interpretation of variation in relation to instructional goals. These demands are particularly salient in discussions on data science education, where statistical ideas are mobilized to support data interpretation and decision-making under uncertainty. Building on previous research on statistical knowledge for teaching (SKT) and key developmental understanding (KDU), this study proposes a theoretical restructuring of SKT based on variability. Variability is proposed as an anchoring KDU within a broader network of statistical ideas, one that organizes teachers' interpretations of statistical content, student reasoning, and pedagogical judgment. Rather than treating statistical concepts as parallel components of teacher knowledge, the proposed perspective highlights the asymmetrical role of variability in structuring developmental relationships among SKT components. By articulating a variability-centered developmental model of SKT, this study extends existing accounts of teachers' statistical knowledge and clarifies their relevance in instructional contexts, where data interpretation and uncertainty are central. The proposed framework provides a conceptual basis for examining how teachers' statistical knowledge develops and transforms in relation to the goals of K-12 data science education.