The open dataset
Every feed record's latest grade and measured category scores in one file, updated on completed scoring runs and free to download and cite. The same scores behind the agency pages, published flat so a researcher or transit program can analyze every feed this site currently covers at once.
Get the data
One row per feed record: id, name, date, letter grade, overall score, available category scores, and days until the service calendar expires. Built from the same dated artifacts the site runs on, so it is reproducible and versioned.
Download CSV Download JSON United States NTD portfolio (JSON) Accessibility coverage (JSON)
Browse the same data as pages: United States NTD readiness, accessibility data coverage, and a procurement clause agencies can put in a vendor contract. The full endpoint list is at api/v1/index.json.
United States NTD readiness, by state
This separate view covers United States feed records. Since Report Year 2023, every National Transit Database reporter with fixed-route or deviated-fixed-route service has to publish a valid, current GTFS feed and certify it each year. This is how many tracked U.S. feed scorecards look ready, nationally and per state. Readiness mirrors three pillars (published, valid, current); it is a heads-up, not an official determination.
| State | Ready | Needs attention | Not ready | Total |
|---|
Download ntd.json for the full United States national and per-state breakdown.
How to use and cite it
Free to use, share, and build on. If you cite it, please link gtfsscorecard.org and note the date you pulled the data, since it can change after a completed scoring run. Check pipeline status for the latest run time. The scoring method, with citations, is in docs/rubric.md, how it maps to the standards is in docs/crosswalk.md, and the machine-readable weights are at scoring.json.
The CC BY 4.0 license covers this scorecard's reports and derived dataset. It does not relicense or redistribute the underlying GTFS files. Each feed keeps its publisher's own terms, recorded in its scorecard artifact.
This is a data-quality lens, not an official compliance determination. The grade is computed on top of the canonical MobilityData GTFS validator.
If you are reporting on a whole program
The dataset above stays free for that, including in a deliverable you are paid for. What is sold is the finished document, not the numbers: the program report bundle renders every agency you support as one archive of branded board reports, built from these same published scorecards. Prices are on that page, and a purchase buys no influence over grades, methodology, or which agencies are listed.
If you need the whole corpus as a table over time
The file above is the latest check per feed record. Every dated check stays free as
well, one file per agency per date under /data/artifacts/, and so does the
open-source command that turns them into tables. What is sold is the built artifact: the
scorecard history tables, every dated check and every finding
as two Parquet tables with a data dictionary, provenance, and license, rebuilt monthly and
licensed to one organization. The price is on that page, and a purchase buys no influence
over grades, methodology, or which agencies are listed.