Six years after the pandemic, two in three district-run schools are at least 10% smaller and one in seven has lost a third

Summary

For district-run schools with October membership in both 2019-20 and 2025-26, computed percent change, its distribution and the shares below thresholds; the grade pipeline is kindergarten ÷ grade 5 from the per-grade columns.

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.

• Cached outputdata/analysis/miami-enrollment.json

  The enrollment analysis packet: trend, school change, utilization, lift table, spending versus size.

• External datasetFLDOE Membership by District and School by Grade (Survey 2)

Steps

1. Change per school.

   SELECT a.school_number, a.value AS e2019, b.value AS e2025, (b.value / a.value - 1) * 100 AS pct
   FROM miami_school_year_metrics a JOIN miami_school_year_metrics b USING (school_number) JOIN miami_schools s ON s.school_number = a.school_number
   WHERE a.metric_key = 'enrollment_oct' AND b.metric_key = 'enrollment_oct' AND a.year = '2019-20' AND b.year = '2025-26' AND a.subgroup = 'ALL' AND b.subgroup = 'ALL'
     AND NOT a.suppressed AND NOT b.suppressed AND a.value > 0 AND NOT s.is_charter AND s.school_type = 'REGULAR';

2. Distribution and by-level summary (packet school_change); pipeline ratios (packet pipeline_ratios).

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.