Replication Data for: From Child Cohorts to Public Schools: Enrollment Tracks Demographic Change More Closely Than Staffing Across U.S. Counties, 2009-2023
Description
Replication materials for a national county-year study comparing how resident child cohorts align with grades 1-5 public enrollment and elementary teacher full-time-equivalent staffing from 2009 through 2023. The archive contains analysis-ready public-data-derived panels, Python code, generated tables and figures, source documentation, and file checksums. It reproduces paired common-sample models, age-timing checks, county-trend and first-difference specifications, common-support horizon profiles, system-size comparisons, geographic validation, and out-of-sample forecasts. The files contain aggregate geographic and school records and no individual student, teacher, household, or birth records. No API key or credential is included.
Files
Steps to reproduce
Use Python 3.11 or later and install the versions in requirements-prpr.txt. From the repository root run: python 03_scripts/run-prpr-analysis.py --clean-output The master runner stops if the central sample counts, horizon support, or estimand schema do not match the manuscript. Upstream public-source acquisition requirements and the raw-source reconstruction boundary are documented in the repository README and source notes.