The 25 schools improving fastest right now
The question
Which New York City elementary schools posted the biggest single-year math gains in 2024-25, and how much of those gains is real teaching versus a test-scoring change?
Twelve New York City elementary schools added 40 or more percentage points to their fifth-grade math proficiency rate in a single year (2023-24 to 2024-25). P.S. 21 Philip H. Sheridan in the Bronx led, climbing from 23% to 90%. Almost all the top movers serve higher-need student bodies (challenge groups 5-10 on the 1-to-10 scale) and most are zoned Brooklyn or Bronx schools. Some of the single-year movement comes from New York State changing how it scores the test in 2024-25 (about 4 percentage points citywide), but these schools exceeded that change by 30 percentage points or more. Single-year jumps often regress; confirming with the 2025-26 results will matter.
Data analysis
Investigation run 2026-05-30.
TL;DR
Twelve New York City elementary schools added 40 or more percentage points to their fifth-grade math proficiency in a single year (2023-24 to 2024-25). P.S. 21 Philip H. Sheridan in the Bronx led at +67 points (23% to 90%). Almost all the top movers serve high-need student bodies (challenge groups 8-10) and are zoned Brooklyn or Bronx schools. The state test rescaling in 2024-25 lifted the citywide third-grade math average by 13 percentage points; these schools exceeded the citywide bump by 30 points or more.
Top 12 single-year movers, fifth-grade math
| School | Borough | Challenge group | 2023-24 | 2024-25 | Change |
|---|---|---|---|---|---|
| PS 21 Philip H. Sheridan | Bronx | 9 | 22.8 | 90 | +67.2 |
| PS 277 | Bronx | 5 | 24.3 | 83.9 | +59.5 |
| PS 90 Edna Cohen | Brooklyn | 8 | 27 | 83 | +56.0 |
| PS 5 Dr. Ronald McNair | Brooklyn | 5 | 44.4 | 100 | +55.6 |
| PS 335 Granville T. Woods | Brooklyn | 4 | 12.5 | 64 | +51.5 |
| PS 147 Isaac Remsen | Brooklyn | 8 | 24 | 72.7 | +48.7 |
| PS 16 Leonard Dunkly | Brooklyn | 7 | 10 | 57.1 | +47.1 |
| Wave Preparatory ES | Queens | 8 | 37.9 | 82.9 | +45.0 |
| PS 201 Discovery School | Queens | 7 | 43.9 | 88 | +44.1 |
| PS X015 Environmental Learning | Bronx | 8 | 39.5 | 83.3 | +43.9 |
| PS 257 John F. Hylan | Brooklyn | 10 | 29.7 | 70.6 | +40.9 |
| PS 16 John J. Driscoll | Staten Island | 9 | 25.4 | 66 | +40.6 |
How much of this is real?
The state rescaled the math test in 2024-25, and citywide fifth-grade math went up by roughly 6 to 13 percentage points depending on the grade. A school that gained 13 points might have gained zero in actual classroom learning. But the schools above gained 40 to 67 points — well above what the rescaling alone could produce. These look like real improvements amplified by the test change.
Pattern check
11 of the 12 are zoned neighborhood schools (only P.S. 211 admits via open application). 10 of the 12 serve higher-need student bodies (challenge groups 5-10). None are charters. None use screened or composite admissions. None are in wealthier neighborhoods. The improvement is concentrated at zoned schools serving higher-need populations.
P.S. 21 Philip H. Sheridan — the biggest single-year mover
P.S. 21 in the Bronx (challenge group 9) climbed from 23% proficient to 90% in one year. Even discounting the citywide test rescaling, it's a 50-percentage-point gain. What changed at P.S. 21 between 2023-24 and 2024-25 is the question to chase on the ground.
Caveats
Single-year movers often regress. The 2025-26 results will tell us whether these gains held. A 3-year sustained-improvement filter would be more robust, but most of these schools wouldn't survive it because they only have one year of climb on the record.
What's next
P.S. 21 Philip H. Sheridan is the lead. Visit, interview the principal, look at budget and staffing changes. If the gain holds in 2025-26, it's a major New York City school turnaround story.
Methodology & replication recipe — the data sources, queries, and steps behind this analysis.