---
pretty_name: NYT COVID-19 — US county daily
license: other
size_categories:
- 1M<n<10M
tags:
- cadence-daily
- geo-us
- surveillance-respiratory
- pathogen-sars-cov-2
- tier-3
- availability-open
schema_version: '0.1'
source_id: nyt-covid
source_url: https://github.com/nytimes/covid-19-data
manifest_section: §16.5
surveillance_category: respiratory
pathogens:
- sars-cov-2
availability: open
availability_notes: NYT custom license — non-commercial use ok, commercial use requires permission.
access_type: github
tier: 3
cadence: daily
geography_levels:
- subnational-county
geography_countries:
- US
gold_standard_for: []
vintaged_version_of: null
succeeds: null
derived_from: []
value_columns:
- name: cases
  unit: cases (cumulative)
  value_type: cumulative
  description: Cumulative confirmed + probable COVID-19 cases since pandemic start
  aggregation: max
- name: deaths
  unit: deaths (cumulative)
  value_type: cumulative
  description: Cumulative confirmed + probable COVID-19 deaths since pandemic start
  aggregation: max
notes:
  extra_columns:
  - column: location_name
    description: NYT-supplied county + state name. Useful for traceability when a FIPS code resolves ambiguously (e.g. NYC's
      five boroughs collapsed into a single 'New York City' row).
  interpretation_caveats:
  - column: cases
    caveat: 'Cumulative — diff to recover incident. NYT collapses the five NYC boroughs (FIPS 36005, 36047, 36061, 36081,
      36085)

      into a synthetic ''New York City'' row that this ingest **drops** (no FIPS). Pre-2020-08 some rural counties were

      bundled with neighbours; the README documents the carve-outs.

      '
  general: 'Archived March 2023. Useful as an early-pandemic baseline at county granularity

    when nothing finer than state was available from federal sources.

    '
extra:
  source_columns_omitted: 0
  archive_note: Series stops 2023-03-23; the upstream maintained the file but no longer updates it.
computed:
  last_ingested: '2026-04-26T12:49:03Z'
  row_count: 2479154
  time_coverage:
  - start: '2020-01-21'
    end: '2022-05-13'
  geography_unit_count: 3220
  observed_cadence_days: 1
  missing_gaps: []
  data_hash: a040b3fa14225d93
---

# NYT COVID-19 — US county daily

Archived March 2023. Useful as an early-pandemic baseline at county granularity
when nothing finer than state was available from federal sources.

**Source:** <https://github.com/nytimes/covid-19-data>

## Coverage

- **Time:** 2020-01-21 → 2022-05-13
- **Cadence:** `daily` (observed median spacing: 1 days)
- **Geography levels:** `subnational-county` — 3220 unique location IDs
- **Countries:** US
- **Pathogens:** `sars-cov-2`
- **Surveillance category:** `respiratory`
- **Rows:** 2,479,154

## Columns

| Column | Unit | value_type | Aggregation | Description |
|--------|------|------------|-------------|-------------|
| `cases` | cases (cumulative) | `cumulative` | `max` | Cumulative confirmed + probable COVID-19 cases since pandemic start |
| `deaths` | deaths (cumulative) | `cumulative` | `max` | Cumulative confirmed + probable COVID-19 deaths since pandemic start |

### Additional data columns

- **`location_name`** — NYT-supplied county + state name. Useful for traceability when a FIPS code resolves ambiguously (e.g. NYC's five boroughs collapsed into a single 'New York City' row).

## Interpretation caveats

Things that may differ from how other sources define a similar measure. If you're combining this dataset with another, read these first.

- **`cases`** — Cumulative — diff to recover incident. NYT collapses the five NYC boroughs (FIPS 36005, 36047, 36061, 36081, 36085)
into a synthetic 'New York City' row that this ingest **drops** (no FIPS). Pre-2020-08 some rural counties were
bundled with neighbours; the README documents the carve-outs.


## Access

- **Availability:** `open`
- **Access type:** `github`
- **License:** other
- **Tier:** 3

---

*Schema version `0.1` · Last ingested 2026-04-26T12:49:03Z · `source_id: nyt-covid` · Manifest section §16.5*