Federal transit safety datasets describe reported agency-level events and summaries; they are not a substitute for the records of one incident. They can help a reader understand what a dataset counts, but they do not establish why an event occurred or who was responsible.

What the National Transit Database publishes

The Federal Transit Administration’s National Transit Database (NTD) releases several safety and security files. Its safety-data overview distinguishes time-series summaries from detailed major-event data and describes when the files are updated. The dataset, agency, mode, calendar year, and event category all affect what a result means.

Major-event detail and non-major monthly summaries

The FTA describes a Major Safety Events file with one entry for each reported event that meets the major reporting threshold. Those entries can include event date, time, agency, mode, location, and a description. Non-major safety data are different: they summarize reportable occurrences and immediately transported injuries by month, mode, and location rather than supplying a public line-by-line incident record.

The current NTD Safety and Security Reporting Policy Manual identifies the Major Event Report (S&S-40) and the Non-Major Monthly Summary Report (S&S-50). The FTA’s forms page explains that the major form captures detailed information for each major incident, while the non-major form summarizes less severe events per mode per month.

A careful way to filter and compare data

Before interpreting a number, identify the agency, transit mode, reporting year, month or calendar-year field, event category, and whether the file includes major events, non-major events, or both. The NTD releases monthly safety and security data within three months of being reported, so the newest events may not yet appear in the published files.

  • Compare like with like: the same agency, transit mode, dataset, and time period.
  • Read the file notes before comparing years because reporting thresholds and historical adjustments can affect totals.
  • Distinguish a count of events from a count of injuries or fatalities; they are separate measures.
  • Use the NTD glossary when a data field or mode label is unclear rather than guessing from a column name.

What aggregate transit data cannot establish

A dataset result does not decide fault, determine the cause of a crash or onboard event, prove that every relevant record has been released, or show that no event occurred when a result is absent. Reporting thresholds, timing, the level of detail in a file, and the difference between aggregate summaries and incident-specific records all matter.

When the transit operator or record holder is unknown, the related guide to public-transportation entity and record questions addresses that separate identification step. The appropriate records for an individual event depend on the operator, vehicle or rail service, location, date, and the agencies involved.

Information about a Charlotte car-accident matter involving a transit vehicle is available when a roadway collision needs a fact-specific review. Federal aggregate data alone cannot determine responsibility, causation, coverage, or damages for one event.

Sources