01 / The finding
School Performance and Neighborhood Economic Need
Each of the city's 32 community school districts reports what share of its students reached proficiency on the state's grades 3–8 exams. When we set those results against how much economic hardship each district's students face, the pattern is hard to miss.
Schools are reflections of their larger communities, as long as we have impoverished communities, we will have impoverished schools.
02 / District by district
Economic need vs. academic proficiency
Every dot is one community school district. Dot size shows enrollment. Tap or hover for the district's numbers.
Proficiency = share of grades 3–8 tests scored Level 3 or 4, all grades combined, 2025–26, as published per district by NYC Public Schools. District schools only; charter schools and District 75 are not included. Economic Need Index = NYCPS estimate of the share of students facing economic hardship; where a student is not individually flagged, it uses the poverty rate of their census tract. The dashed line and the reading and math shares in section 01 are least-squares calculations by SquareOne from those two published files (correlation: ELA r = −0.87, math r = −0.82).
03 / Same students, different economics
Fourteen districts are more than 80% Black and Hispanic. Their results split by income.
These fourteen districts look nearly identical on race, but their academic results are at least 20 points apart.
Districts where Black and Hispanic students together make up 80% or more of enrollment, ordered by Economic Need Index. Need and enrollment shares from the NYCPS demographic snapshot; proficiency as published in the district ELA file.
04 / The districts that break the pattern
Poverty sets the baseline, but there are exceptions.
Two pairs of districts carry nearly identical economic need with varying student scores.
Bedford-Stuyvesant vs. Fordham
| District | Need | Black + Hispanic | ELA | Math |
|---|---|---|---|---|
| D16 Bed-Stuy | 83.9 | 88.7% | 56.3% | 60.0% |
| D10 Fordham | 83.4 | 83.6% | 38.7% | 42.1% |
Same level of hardship, nearly the same demographics, and an 18-point gap in reading.
Sunset Park vs. Corona
| District | Need | Black + Hispanic | ELA | Math |
|---|---|---|---|---|
| D20 Sunset Park | 76.2 | 30.6% | 59.3% | 72.0% |
| D24 Corona | 75.9 | 68.7% | 40.0% | 49.0% |
Two working-class immigrant districts with very different demographics, a third of a point apart on need, 23 points apart in math.
What this page does not claim: that poverty causes low scores, that any district's results are fixed, or that school quality is irrelevant. It shows one year of district-level results alongside one measure of economic need. Districts are not students: a child in any district can and does outperform the pattern around them.
05 / Sources
Where these numbers come from
- Grades 3–8 ELA results by district, 2025–26. NYC Public Schools, district-ela-results-public.xlsx, "ELA - All" tab, All Grades rows. Data as of August 3, 2026. infohub.nyced.org / Test Results →
- Grades 3–8 math results by district, 2025–26. NYC Public Schools, district-math-results-public.xlsx, "Math - All" tab, All Grades rows. infohub.nyced.org / Test Results →
- Economic Need Index, enrollment, and racial makeup by district, 2025–26. NYC Public Schools, demographic-snapshot-2021-22-to-2025-26-public.xlsx, "District" tab. Need and poverty figures as of June 4, 2026. infohub.nyced.org / Information and Data →
- Where students go to school, 2025–26. NYC Public Schools, demographic-snapshot-2021-22-to-2025-26-public.xlsx, "School" tab. SquareOne analysis of 1,550 district schools, excluding charter schools and Districts 75 and 79: 347,032 of 482,808 Black and Hispanic students attend a school with an Economic Need Index of 75% or higher, against 30,371 of 124,211 white students. infohub.nyced.org / Information and Data →
- In response to. Nikole Hannah-Jones, "In Fighting for Every Black Child, Did I Betray My Own?", The New York Times Magazine, September 20, 2026. nytimes.com →
All three data files were downloaded on September 20, 2026. Figures on this page are reported exactly as published, except the trend line, the correlation figures in section 01, the school concentration figures in section 01, and the shares in section 03, which SquareOne calculated from those files.