How we validate the 509 dataset
The same accounting as the white paper, as a turn-the-page field report. Drag a corner to flip. Charts are drawn from the audit itself.
How we validate the 509 dataset
What we corrected, why, and what it changes — the audit behind every number on Exhibit 509.
Each field is checked against its own source of truth
Not every number on a school's page comes from the same document, so we never validate them the same way. The governing rule is simple, and it is the whole method in one line: each field is reconciled against the authoritative ABA source for that field — and nowhere else.
| The 509 PDFs | Admissions (LSAT/GPA, plus GRE & JD-NEXT where reported), enrollment & volume, tuition, fees, scholarships (shares & grant-dollar amounts), bar passage |
| A separate ABA workbook | Employment outcomes — not carried in the 509 PDF at all |
That last row matters: anyone who says a school's employment numbers are "validated against the 509 PDF" is mistaken. The PDF has no such section.
25,457 cells, checked against the PDF they came from
25,440 of 25,457 cells agree with the source PDFs. The 17 that don't are documented artifacts of the audit's own parser — not errors in the data. The school crosswalk audited clean: 0 of 210 mappings diverged.
3,009 reports across fifteen years
PDF reports held per year, 2011–2025 (labelled in odd years). Counts track how many schools were ABA-accredited and reporting that year; the gentle decline reflects closures and mergers, not missing data.
The biggest fix: a one-year enrollment slip
For 2018–2025, the JD-enrollment series had been pinned to the prior year's population instead of the present-year "as of October" census on the front of the report. We re-sourced every modern cell to the right year and added a build guard so it can't silently return.
Correction volume by class of fix. Every change is a re-sourcing or re-alignment — never a fabrication, and every blank stays blank.
The full white paper
The complete accounting — the audit table, every class of correction, which year each field describes, and the limits we keep visible — is on the web.
exhibit509.com/explained/how-we-validate-the-509-dataset
An interactive draft. Drag a page corner, use the buttons, or swipe on mobile.