About

Sources

U.S. Department of Education College Scorecard (2026-06 release) · IPEDS 2023 · Wikidata & Wikimedia Commons & Wikipedia (extracted 2026-07-09/10 (alumni panel), 2026-07-16/17 (facts, portraits)). All public data; alumni records are CC0.

Methods

1. 01

   Stage the Scorecard corpus

   The 2026-06 College Scorecard release is read directly from the raw archive: the most-recent institution snapshot (42 columns), all thirty annual files 1996–2026 for trend lines, and the field-of-study files for program-level earnings and debt.

2. 02

   Join the IPEDS directory

   IPEDS 2023 supplies the identity scaffold — coordinates, county, NCES locale, sector, level, and fall enrollment — joined to the Scorecard on unitid.

3. 03

   Build the unitid ↔ Wikidata crosswalk

   Wikidata items holding an IPEDS ID are pulled by property query, then verified with a coordinate sanity check (campus location vs. Wikidata's) to catch system-vs-flagship and same-name mismatches.

4. 04

   Extract the alumni panel

   For the 89 Tier-1 campuses, every person Wikidata records as educated there: counts, field of work in thirteen buckets, gender by birth decade, occupations, and the most-covered exemplars.

5. 05

   Fetch imagery with its licenses

   Campus photos and portraits come from Wikimedia Commons. Every referenced file gets a manifest row — artist, license, source link — and no image renders without one.

6. 06

   Add the Wikipedia layer

   Each Tier-1 school's lead extract (with revision ID for provenance) and monthly article pageviews since 2015, from the Wikimedia REST APIs.

7. 07

   Compute context

   Sector quantile grids (control × level) and within-sector percentiles for every headline metric, plus each school's eight nearest peers in its sector.

8. 08

   Assemble one payload per school

   Everything above lands in a single JSON per institution, with the caveats shipped inside the payload so no page can forget to render them.

9. 09

   Bake the site

   A fully static build renders all 6,273 profiles at build time. The published site makes no database or storage calls — what you read was checked in at build.

What the numbers mean

Median earnings 10 years after entry

Students who received federal aid, measured from the year they entered. Students who never received federal aid are not in this number.

Average net price after grants

Average yearly price after grants and scholarships, for students who received federal aid.

Sticker price (average yearly cost of attendance)

The published average cost for all students, before any aid.

Not reported

Not reported: suppressed for privacy or never collected. It is not zero, and nothing has been filled in.

The rules this site reads by

• Suppressed is suppressed. Privacy-suppressed and uncollected values render as “not reported” — never zero, never imputed.
• A gap is not a zero. Trend lines break where a measure wasn't collected; early-year nulls mean “not yet collected.”
• Earnings cover federally aided (Title IV) students only — students who never took federal aid are invisible to these figures.
• Percentiles are within-sector by default and labeled as such (“82nd percentile among private nonprofit 4-years”).
• Wikidata measures recorded notability — access × achievement × documentation — never placement rates or institutional quality.
• Geography is the campus address, not where students live or the area a school serves.

Why there are no rankings

The Scorecard measures money; Wikidata measures memory. Neither measures teaching, belonging, or what a particular student will make of a particular place — so this site orders schools alphabetically, geographically, or by the reader’s own question, never by a composite score.

Images

Photographs and portraits are hotlinked from Wikimedia Commons under free licenses. Free-licensed is not public domain: every image on this site carries its artist and license, linked to the source file.