ny_times_us_counties
Data license: LICENSE · Data source: The New York Times · About: simonw/covid-19-datasette
6 rows where cases = 28, county = "Bradford" and state = "Pennsylvania" sorted by date descending
This data as json, yaml, Notebook, copyable, CSV (advanced)
Suggested facets: date (date)
Link | rowid | date ▲ | county | state | fips | cases | deaths |
---|---|---|---|---|---|---|---|
97358 | 97358 | 2020-04-28 | Bradford | Pennsylvania | 42015 | 28 | 5 |
94519 | 94519 | 2020-04-27 | Bradford | Pennsylvania | 42015 | 28 | 5 |
91692 | 91692 | 2020-04-26 | Bradford | Pennsylvania | 42015 | 28 | 5 |
88876 | 88876 | 2020-04-25 | Bradford | Pennsylvania | 42015 | 28 | 5 |
86062 | 86062 | 2020-04-24 | Bradford | Pennsylvania | 42015 | 28 | 5 |
83258 | 83258 | 2020-04-23 | Bradford | Pennsylvania | 42015 | 28 | 5 |
Advanced export
JSON shape: default, array, newline-delimited
CREATE TABLE [ny_times_us_counties] ( [date] TEXT, [county] TEXT, [state] TEXT, [fips] INTEGER, [cases] INTEGER, [deaths] INTEGER ); CREATE INDEX [idx_ny_times_us_counties_state] ON [ny_times_us_counties] ([state]); CREATE INDEX [idx_ny_times_us_counties_county] ON [ny_times_us_counties] ([county]); CREATE INDEX [idx_ny_times_us_counties_fips] ON [ny_times_us_counties] ([fips]); CREATE INDEX [idx_ny_times_us_counties_date] ON [ny_times_us_counties] ([date] desc);