Unforgetting the Past, Imagining the Future: A Machine Learning Analysis of Virginia School Finance Patterns

• In Event: Money Matters: Finance, Markets, and Equity in School Choice Systems

Abstract

This study used cluster analysis and random forest machine learning to classify Virginia public school districts (N = 132) by local, state, and federal revenue and measures of equity and adequacy. Three distinct clusters were identified—locally reliant, externally reliant, and balanced—that correspond to differing revenue source patterns. Results from a MANOVA and subsequent ANOVAs revealed significant differences in local ability to pay, adequacy gap, child poverty rate, and proportions of revenue from each source. Findings suggest a reinforcing feedback loop whereby revenue patterns tied to structural wealth may perpetuate disparities in school finance equity and adequacy. This research offers a novel, systems-level approach to understanding these relationships, providing valuable insight to help scholars and policymakers imagine more equitable futures.

Author

• Luke Parker, Virginia Commonwealth University