Mapping Educational Inequities Using Bayesian Networks and Spatial Analysis of Social Determinants

• In Event: The Nuance of Patterns: Methods for Uncovering Intricate Quantitative Relationships

Abstract

This study presents a novel methodological framework combining Bayesian network modeling and spatial analysis to examine how social determinants of health (SDOH) condition educational outcomes across Kansas counties. Using county-level public data, we model probabilistic relationships among variables such as food insecurity, parental education, and neighborhood safety, while also identifying geographic clusters of systemic disadvantage. Preliminary findings reveal key dependencies—e.g., dropout rates linked to unemployment and low parental education—and spatial hotspots of inequity. By integrating spatial context into Bayesian models, this research advances equity-focused education research, providing actionable insights for policy and planning. The study offers a reproducible, theory-driven method aligned with AERA's call to “unforget histories” and imagine data-informed futures for educational justice.

Author

• Merve Akin-Tas, University of Kansas