PhiladelphiaMethodologymedium
Philly PSSA: 2023-24 vs 2024-25 year-over-year — which schools moved most
Thesis
With both 2023-24 and 2024-25 SDP PSSA in our DB (loaded for the SPREE spot check), we can now produce year-over-year deltas at the school × metric × subgroup grain. The Movers page surfaces this; the methodology question is whether 1-year movements are meaningful signal or just measurement noise. NYC's pattern (from the small-schools-legacy story) suggests school-level year-over-year is mostly noise.
Supporting findings
- Two years of SDP PSSA data: 2023-24 (13k facts) and 2024-25 (13k facts).
- Year-over-year delta computable for matched schools.
- Single-year deltas at the school level are typically very noisy (per NYC pattern).
- Movers UI surfaces top gainers and losers but should be read as a watch list, not a definitive ranking.
Reporting directions
- Compute the distribution of year-over-year deltas. Most should be near zero.
- Flag schools with statistically-large deltas (Δ > 2sd) for further reporting.
- Backfill more years (2021-22, 2022-23) to give the trend more context.
Methodology & replication recipe — the queries + sources behind the findings above.
More on methodology
- Philly publishes both Acct and Actual PSSA cuts. They're byte-identical.
- SPREE agrees with our DB on 616/616 cells — Philly's data pipeline is structurally clean
- Computed-value (iv) validation: OSS + PSES rollups agree to the cell with the raw source
- With 287 peer-grouped schools and k=40, the rankings barely move at k=20