| layout | title | nav_order | has_children |
|---|---|---|---|
default |
CSV Bulk Download |
6 |
false |
We provide access to some of our data in a relational format in a public Google Cloud Storage bucket, which is available for CSV download. The tables are constructed such that each row represents a Place and each column represents a Statistical Variable.
These relational tables are organized by vertical, each within a different zip folder, which can be downloaded from the links below:
- Agriculture
- Climate
- Crime
- Demographics
- Economics
- Education
- Employment
- Energy
- Health
- Household
- Housing
Each vertical zip folder contains tables for various Place categories: all (all places), us (US places), non_us (non-US places), county (US counties), and zip (US zip codes). For each vertical and Place category, there are three types of tables:
value: Each cell contains the value of the latest observation for a given Statistical Variable and Place.date: Each cell contains the date of the latest observation for a given Statistical Variable and Place.provenance: Each cell contains the provenance URL of the latest observation for a given Statistical Variable and Place, as well as the measurement method, if provided. Measurement methods that are prefixed withdcAggregate/represent Data Commons aggregated values.
The table names follow the pattern [vertical]_[place_category]_[type] and are sharded into multiple CSV files. (For example, the file demographics_all_date-00000-of-00456.csv contains a portion of the observation dates for demographics Statistical Variables and all Places. In this case, the table has been sharded into 456 files.)
The corresponding value, date, and provenance tables can be joined using the first three columns, which contain information about the place:
place_name: The name(s) of the Place.place_dcid: The Data Commons ID for the Place.place_type: The type(s) of the Place.
Below is a subset of the housing_county_value table:
| place_name | place_dcid | place_type | Count_HousingUnit | Count_HousingUnit_NoCashRent | ... |
|---|---|---|---|---|---|
| Nuckolls County | geoId/31129 | County | 2445 | 74 | ... |
| Wells County | geoId/38103 | County | 2422 | 74 | ... |
| ... | ... | ... | ... | ... | ... |
And the corresponding subset of the housing_county_date table:
| place_name | place_dcid | place_type | Count_HousingUnit | Count_HousingUnit_NoCashRent | ... |
|---|---|---|---|---|---|
| Nuckolls County | geoId/31129 | County | 2019 | 2019 | ... |
| Wells County | geoId/38103 | County | 2019 | 2019 | ... |
| ... | ... | ... | ... | ... | ... |
And for the housing_county_provenance table:
| place_name | place_dcid | place_type | Count_HousingUnit | Count_HousingUnit_NoCashRent | ... |
|---|---|---|---|---|---|
| Nuckolls County | geoId/31129 | County | https://www.census.gov/|CensusACS5yrSurvey | https://www.census.gov/|CensusACS5yrSurvey | ... |
| Wells County | geoId/38103 | County | https://www.census.gov/|CensusACS5yrSurvey | https://www.census.gov/|CensusACS5yrSurvey | ... |
| ... | ... | ... | ... | ... | ... |
The provenance value https://www.census.gov/|CensusACS5yrSurvey indicates that the observation comes from https://www.census.gov/ using the CensusACS5yrSurvey measurement method.