ARTIFICIAL NEURAL NETWORK MODEL FOR EDUCATIONAL EQUITY: NONLINEAR DYNAMICS OF POLICY, INVESTMENT, AND INCLUSION

Abstract

This study develops and tests an Artificial Neural Network (ANN) model to analyze the nonlinear relationships among key constructs associated with educational equity in Latin America and the Caribbean. Drawing on a system of indicators related to access, retention, completion, public investment, intercultural inclusion, and policy coherence, the model integrates multidimensional data into a predictive architecture capable of capturing complex interactions. The ANN structure includes an input layer composed of standardized indicators, a hidden layer with optimized synaptic weights, and an output layer representing an educational equity index. Results show high predictive accuracy and reveal differentiated pathways through which economic investment, inclusion, and governance exert both reinforcing and constraining effects on equity outcomes. The findings highlight the central role of investment as a structural driver, the mediating function of inclusion, and the regulatory influence of systemic coherence. The study contributes to methodological innovation by combining psychometric validation with machine learning techniques, offering a robust framework for policy analysis and decision-making in complex educational systems.