Ohio School Report Cards — District Details

Source. Ohio Department of Education and Workforce, Ohio School Report Cards, file 2025_District_Details.xlsx, sheet District_Details. Type. Primary source — machine-readable per-district subgroup data. Location. reportcardstorage.education.ohio.gov/data-download-2025/. Requires the read-only query token recorded in “Report card connector”.

What it contains. Long form — one row per district per student group, 5,820 rows over the same 607 rated traditional districts. Per row: Enrollment, Enrollment Percent, Attendance rate, Mobility Rate, Chronic Absenteeism Rate. Eleven groups, of which All Students, Economic Disadvantage, English Learner, and Students with Disabilities are the ones this corpus reads.

Why it matters here. These are the need covariates, published for the same school year as the outcomes, by the same publisher, on the same IRN key. They are what made a need-adjusted model possible — the step OCG White Paper 013 named as its own priority and could not take.

THE ECONOMIC-DISADVANTAGE SHARE HERE IS NOT THE CUPP REPORT’S. This is the most important caveat in the file and it is not stated on the file.

	this file (2024-25)	Cupp Report (FY2024)
median share	49.5%	47.0%
districts at exactly 100%	87	37
districts at 95% or more	197	—
correlation with the Performance Index	−0.734	−0.846

The two correlate with each other at only +0.823, and individual districts disagree wildly — one reports 94.4% here against 5.8% there. This is not a year of drift. This measure is top-coded by community eligibility: where a district qualifies for universal free meals, every enrolled student is counted economically disadvantaged, so the variable saturates at 100% and stops discriminating among the poorest districts. Akron City reports an Economic Disadvantage enrollment exactly equal to its All Students enrollment. verified

The censoring shows up exactly where censoring should: the saturated measure gives the weaker association with the outcome. This corpus uses the Cupp measure for its headline poverty findings and commits both. verified see crates/xcheck/tests/report_card_2425.rs

Access constraints. Freely available, no registration. XLSX.

Caveats:

• NC and <10 are a matched pair. Where a group has fewer than ten students the enrollment cell reads <10 and the percent reads NC — 216 districts for English Learner. These are suppressed small counts, not absent data and not zero. Any share imputed at zero for them is a stated assumption; the corpus makes that assumption for English-learner share, notes that ten students in a district of ~1,500 is under 0.7%, and re-runs the model on complete cases as a check. verified
• Attendance, mobility, and chronic absenteeism are outcomes, not covariates. They sit in the same rows and are tempting. Putting them on the right-hand side of a spending-outcome model controls away a mediator and biases the coefficient of interest toward zero. This corpus does not use them as controls. inference
• Long form. Keyed by (IRN, student group). A join on IRN alone silently multiplies rows.
• Group coverage varies — 607 districts report Students with Disabilities, 606 Economic Disadvantage, 303 a numeric English Learner share, 119 Migrant.

Feeds connector #

dew-report-card, source key district-details-2425.