The federal chronic absenteeism number for Miami-Dade elementary schools is lower than the state's because it leaves out pre-kindergarten
A reproducibility log: the data this analysis touched, the queries it ran, and how a third party would re-run it.
Summary
Compared the state's absent-10%+ count (rate × denominator) with the EDFacts DG814 count for 2022-23 by NCES id, split by whether the school has October pre-K membership; regressed the gap on pre-K membership.
Data sources
- Database tablemiami_school_year_metrics
Long-format facts: (school_number, year, metric_key, subgroup) with value, suppressed, denominator, peer and district percentiles, and the residual-z columns.
- External datasetEdDataExpress FS195/DG814 chronic absenteeism, SY2022-23
- Loader script
pipeline_miami/verify/cross_publisher_absenteeism.py
Steps
python -m pipeline_miami.verify.cross_publisher_absenteeism → docs/qa_reports/miami/__cross_publisher_absenteeism.md
Caveats
EDFacts perturbs small cells by ±1; the state file counts students enrolled 10 or more days. The regression uses an annual count gap against October PK membership; its slope is not a direct estimate of PK chronic-absence prevalence.
Reproduce
Follow docs/cities/miami/07_runbook.md: `make miami-discover-all`, parse the PDFs, run the loaders in order, `make miami-verify`, then the analysis scripts. The queries below run directly against the local Postgres.
The recipe lives at data/cities/miami/stories/recipes.ts in the repo.