← All stories

COVID recoveryeasy

District 18 leads NYC's pandemic math recovery — by a lot

The question

Why did Brooklyn's District 18 lead every other New York City district on math gains since the pandemic — and by such a wide margin?

School District 18 covers East Flatbush and Canarsie in Brooklyn — 13 elementary schools serving a student body that's predominantly Caribbean-American (Haitian, Jamaican, Trinidadian). Between 2018-19 and 2024-25, the share of fifth-graders proficient in math jumped 32 percentage points in that district. No other district came close: the next-highest gained 22 points; the city as a whole gained 10. Even after subtracting the citywide trend (which included a New York State change in how it scores the math test), District 18's lead over the typical district is the biggest district-level move in the data. Twelve of the 13 schools improved, and the top three added 40 to 66 percentage points each.

Read the full briefing

Data analysis

Investigation run 2026-05-29.

TL;DR

School District 18 led every other New York City district on grade-5 math gains by a wide margin: +32 percentage points, compared with +22 at the next-highest district and +10 citywide. The gain is broad, not the work of a single school — 12 of the district's 13 elementary schools improved, and the top three (P.S. 233 Langston Hughes, P.S. 276 Louis Marshall, and P.S. 114 Ryder) added 40 to 66 percentage points each. About 10 of those points reflect a New York State change in how the math test gets scored. The remaining 22 is the biggest district-level net gain in the period. The likely contributors overlap in time: Superintendent Tiffany Hardrick took office in July 2022 (right at the steepest year of climb), federal pandemic-relief dollars (the American Rescue Plan) peaked through 2022-23, and the district was part of a New York City instructional-support program called 'Lighthouse.'

Every New York City school district, ranked by math gain

Every district with at least 5 schools is shown. District 18 stands more than 10 percentage points above the next district. The Bronx and Brooklyn districts dominate the top — the parts of the city with historically lower test scores. The wealthier districts (District 2 covering Manhattan's Upper East and West Sides; District 26 in Bayside, Queens) show smaller gains, partly because their scores were already high and there wasn't as much room to climb.

Δ grade-5 math proficiency by district, 2018-19 → 2024-25 (all NYC districts)
District 18 — Brooklyn (E. Flatbush)
+32.4n=13
District 19 — Brooklyn (East NY)
+22.2n=22
District 11 — Bronx (Pelham, Co-op City)
+19.7n=29
District 14 — Brooklyn (Williamsburg)
+19.6n=19
District 29 — Queens (SE Queens)
+17.0n=27
District 23 — Brooklyn (Brownsville)
+16.6n=15
District 12 — Bronx (Hunts Pt, Crotona)
+16.1n=19
District 31 — Staten Island
+14.8n=47
District 10 — Bronx (Riverdale, Bedford)
+13.1n=37
District 28 — Queens (Jamaica)
+12.8n=24
District 17 — Brooklyn (Crown Heights)
+12.6n=22
District 07 — Bronx (Mott Haven)
+12.6n=17
District 05 — Manhattan (Harlem)
+11.1n=16
District 06 — Manhattan (Wash. Heights)
+10.7n=24
District 04 — Manhattan (East Harlem)
+9.2n=18
Vertical line: Citywide avg (+9.7)

How District 18 moved year by year, compared with the city

District 18 didn't gain steadily. Through 2021-22, its scores were flat or slipping (32% proficient in 2018-19, 31% in 2021-22). The turning point is 2022-23, when the district jumped from 31% to 53% in one year. The next big jump (2023-24 to 2024-25, 51% to 65%) includes the citywide testing-scale change, but District 18 also held its gains in 2023-24 and added on top of them, suggesting a real district-level shift that the test change is amplifying, not creating.

Share of fifth-graders scoring proficient in math: School District 18 versus the city as a whole, 2017-18 to 2024-25
25%36%48%59%70%2017-182018-192021-222022-232023-242024-25test rescaled
District 18 averageCitywide average

Where the gain came from — school by school

Twelve of District 18's 13 elementary schools improved. The top three added 40 to 66 percentage points. The bottom-ranked school, P.S. 235, was essentially flat (+2.6). It's the spread of improvement across schools — not one outlier — that makes District 18 stand out.

Change in the percentage of fifth-graders scoring proficient in math at each of District 18's 13 elementary schools, 2018-19 to 2024-25
P.S. 233 Langston Hughes
+65.7
P.S. 276 Louis Marshall
+55.6
P.S. 114 Ryder Elementary
+42.0
P.S. 279 Herman Schreiber
+40.4
P.S. 268 Emma Lazarus
+39.9
P.S. 244 Richard R. Green
+38.5
P.S. 272 Curtis Estabrook
+31.7
P.S. 066
+28.3
P.S. 115 Daniel Mucatel
+24.2
P.S. 135 Sheldon A. Brookner
+19.8
P.S. 208 Elsa Ebeling
+17.9
P.S. 219 Kennedy-King
+15.0
P.S. 235 Janice Marie Knight
+2.6
Vertical line: District 18 average (+32.4)

Pulling apart the gain — how much is the test change?

Citywide, fifth-grade math proficiency rose by about 10 percentage points between 2018-19 and 2024-25 (from 44.6% to 54.3%). Part of that came from New York State changing the way it scores the test, covered in a separate story on the test recalibration. The question for District 18 is how much of its 32-point gain holds up after we account for the citywide trend. If we credit the full 10-point citywide rise to the test change, District 18's net gain is 23 percentage points — still the largest in the city. If we're more conservative and assume only the most recent year's citywide bump (about 4 points) was the test change, District 18's adjusted gain is 28 points. Either way, District 18 ran well ahead of the city as a whole.

Decomposition of District 18's +32.4-point gain

ComponentPointsMethod
Raw 2018-19 → 2024-25 gain+32.4From data
Citywide trend over same period+9.7From data
District 18 lead over citywide trend (conservative)+22.7Subtract all of citywide
District 18 lead over typical district (Districts 7-31 mean ≈ +15.5)+16.9Subtract typical Bronx/Brooklyn district gain
District 18 lead over next-highest district (District 19)+10.2Subtract District 19's +22.2

What changed in District 18 around 2022

Several things happened in the same window as the gain. None of them is identified as the cause; they're candidates a reporter would chase:

• Superintendent Tiffany Hardrick took office in July 2022, right at the start of the steepest single year of climb. • Federal pandemic-relief dollars (the American Rescue Plan) peaked through 2022-23. Every district got some; this isn't unique to District 18. • District 18 was part of New York City's 'Lighthouse' instructional-support program from its start around 2021. • Several District 18 schools changed principals during the 2020-2022 window, according to records on the city's school-finder page. • The district's long-running Saturday Academy program — extra weekend instruction — was expanded around 2022.

These are overlapping in time, not proven to be the cause. The data shows the gain is real and spread across the district's schools. Pinning down what drove it would take interviews, budget records, and principal-tenure data we don't have on hand.

Caveats

This analysis works with school-level averages, not individual students. A district can post bigger averages just by enrolling a different mix of students in fifth grade than it did six years earlier; we don't have student-by-student data to check that. With only 13 schools, District 18's 32-point average carries more uncertainty than the equivalent number from a 100-school district would. The 10-point lead over the next district is big relative to that uncertainty but not enormous. The 2024-25 state test rescaling accounts for some of the absolute gain, but it doesn't affect District 18's lead over other districts — every district faced the same rescaling.

What's next

Pull District 18 budget data from the city's school-budget portal (Galaxy) for 2018-2024 and compare how per-pupil spending has moved against districts that didn't improve. Interview Superintendent Hardrick. Profile P.S. 233 Langston Hughes — the biggest school-level gain at +66 percentage points — but first check whether the school's enrollment and student-body makeup stayed stable across the period, before treating it as a teaching turnaround. For contrast, compare District 18 with neighboring District 23 (also in Brooklyn, also starting from a low baseline, with similar Caribbean-American demographics, but only a 17-point gain).

Methodology & replication recipe — the data sources, queries, and steps behind this analysis.

More on covid recovery