Oklahoma public schools (OSDE, NCES, CRDC) - school, district and city education indicators

Source & License

Two provenances, split by publisher and by what the numbers are:

• OSDESchools - Oklahoma State Department of Education. Three file families:
• OSTP (grades 3-8) and CCRA (grade 11) media-redacted results, one file per school year, 2016-17 forward. Counts per site x grade x subject: valid N, Below Basic, Basic, Proficient, Advanced. Cells under the reporting threshold are masked (***).
• Oklahoma School Report Card bulk CSVs at oklaschools.com/archive/{2018..2025}/{file}.csv (af-academic-achievement, af-chronic-absenteeism, af-graduation, af-postsecondary-opportunities, rc-attendance, schools, districts). Rates by site x grade x subject x student group; no counts, no tested N.
• Oct 1 enrollment workbooks (site x grade x race x sex), 2015-16 forward.
• CRDCSchools - U.S. Department of Education, Office for Civil Rights, Civil Rights Data Collection public-use files, 2017-18, 2020-21, 2021-22 (biennial; 2023-24 not yet published). Out-of-school suspension counts by school and student group.

The school spine (NCES CCD directory, EDGE geocodes) and the TIGER2023 place polygons are inputs to membership and geography, not observation sources, so they carry no provenance of their own.

Both publishers are government bodies and the files are public records; the download pages state no license terms, so license above is recorded as public domain pending confirmation. OSDE's report-card files carry oklaschools' minimum-n suppression (groups under 10 students are not published) and the OSTP files carry *** masking; both are preserved as suppressed rows, never filled in.

Methodology & caveats

• These are the districts' students, not all of a city's children. Families in suburban districts that reach into the city (Jenks, Moore, Edmond, Mid-Del and others), private schools and exclusively virtual schools are outside this data. A check of member-district enrollment against the city's child population (Kids Count birth cohorts or ACS under-18 counts) is the intended validation of how much of the city's children the districts cover; it is not recorded in the repo - needs confirmation.
• Comparability breaks are flagged on the row and on the City+ card and are never smoothed across. Three mechanisms: an interior excluded year (shown on the chart, dropped from change and trend), a comparability note (a year with no value, explained), and a discontinuity (a series start).
• 2016-17: new academic standards and redesigned OSTP. 2017 forward is the modern series; earlier proficiency is not comparable.
• 2019-20: no state testing. 2020-21: results not published by OSDE (Kids Count / OK Policy Institute also treat both years as unavailable). Report cards for 2020 and 2021 lack achievement, growth and absenteeism.
• 2023-24 cut-score change (ELA and math only; science and history did not move): state elementary ELA proficiency runs 26.8% (2023) -> 48.0% (2024) -> 25.7% (2025); middle ELA 22.1 -> 39.1 -> 22.9. 2024 carries a note and is skipped when computing change. In the state's words (Oklahoma Office of Educational Quality and Accountability, FAQ, https://oklahoma.gov/oeqa/about-oeqa/faqs.html): "in 2024, the cut scores set through a new standard-setting process were no longer aligned to national benchmarks"; "more students appeared 'Proficient' or higher even though those scores did not align with college and career readiness standards"; CEQA's May 2025 decision restored cut scores "aligned with college and career readiness" and "with the National Assessment of Educational Progress (NAEP)"; "A student who was identified as Proficient in 2024 may be in a lower performance level in 2025, not due to academic decline, but because of a return to expectations that are aligned with national standards." Reporting: Oklahoma Voice, 2024-08-21, "Oklahoma's education department quietly made changes that will cause big jumps in state test scores"; Oklahoma Voice, 2025-05-21, "Oklahoma board raises rigor to restore 'truth and transparency' to student testing" (CEQA unanimous vote to restore the 2017-2023 cut scores).
• 2024-25: chronic absenteeism (and so regular attendance) redefined (medical exemptions); attendance cards carry a 2025 note.
• CRDC is biennial and its school universe changes between collections.
• High school ELA and math come from the report card (SAT/ACT-based CCRA grade 11 rates, 2018 forward); no count file exists, so those city figures are grade 11 enrollment-weighted and derived. The grade 11 count file carries only science and U.S. history.
• Headline change is latest year minus the earliest comparable year of the modern series (2017 for assessments, 2018 for report-card indicators), skipping excluded years.
• Public and charter schools only. Nothing is imputed, interpolated or smoothed; suppressed cells are stored as suppressed rows so coverage is always computable and disclosed.
• Rankings among the eight peer cities are shown on the City+ card but never decide a badge; the badge is level and ground vs the state (badge rule in education/docs/methodology.md and strategy/decisions.md).