Replication Data and Code for: "From Plans to Action: Measuring AI Governance Capacity and Structural Asymmetry Across Latin America"

This repository contains the complete replication package for the paper "From Plans to Action: Measuring AI Governance Capacity and Structural Asymmetry Across Latin America — A Regression-Based Analysis Using the ILIA 2025" (submitted to Data & Policy, Cambridge University Press). Contents: (1) the analytical dataset (19 Latin American countries × 4 ILIA 2025 sub-scores), verified line-by-line against the primary source's country-level fact sheets; (2) a self-contained Python script reproducing every statistic reported in the paper, including the mechanical-artifact check, the regression of Governance on Enabling Factors and R&D Adoption, 5,000-replication bootstrap confidence intervals, and the out-of-sample diagnostic for El Salvador; (3) a README with full documentation, including disclosure of a known data-quality anomaly in the primary source (El Salvador's governance score). Primary data source: Soto, Á., Durán, R., Moreno, A., Adasme, S., Rovira, S., Jordán, V., & Poveda, L. (Coords.) (2025). Índice Latinoamericano de Inteligencia Artificial (ILIA) 2025. CEPAL and CENIA. https://indicelatam.cl