The 10 schools doing the best for the most-challenged kids
A reproducibility log: the data this analysis touched, the queries it ran, the outside sources it leaned on, and the steps a third party would follow to re-run it.
Summary
Filtered to highest-need-decile schools (proclivity_decile=10) and ranked by peer-adjusted academic residual descending. Picked the top 10 for profile.
Data sources
- Database tableschools
One row per New York City public school (DBN, name, district, borough, school_type, grade_band, latest_enrollment, proclivity_decile, proclivity_score, closed_at, admission_category).
- Database tableschool_year_metrics
Long-format per-school per-year metric facts (school_dbn, year, metric_key, subgroup, value, suppressed). Loaded from DOE/NYSED public files via scripts/loaders/*.
- Database tablepeer_groups + school_peers
K-nearest-neighbor peer groups derived from grade band + admission category (hard match) plus demographics + topic similarity (weighted). Default k=40. Built in scripts/derive/peer-groups.ts.
Steps
For each decile-10 school, compute (school's composite) − (peer-group-40 average); rank.
Caveats
Peer-adjusted residuals depend on the peer-group definition. Schools that 'punch above' may still score below the citywide average in absolute terms.
Reproduce
Clone the repo, set DATABASE_URL to a Postgres with the project schema loaded, run `npx tsx scripts/loaders/<source>.ts` for any not-yet-loaded data, then issue the queries in the Steps section. The story page also lists the exact `metric_key`/`subgroup`/`year` filters used. Searchable by the answer's headline number — every figure is recomputable from the queries shown.
The recipe lives at data/stories/recipes.ts in the repo. Corrections welcome.