open_data

Guidance for publishing open data

This guidance is intended to help ensure that open data published by Trafford Data Lab is extracted in a reproducible way, provided with accurate metadata and available in an open format.

Files

Each open dataset needs to be stored in its own folder in the open_data GitHub repo with:

• a README file which provides metadata (licence, attribution etc) for the dataset
• a pre-processing R script which shows how the data was retrieved, cleaned, and structured. The script must be scalable so that other local authorities can retrieve data for their area.
• a CSV, and if geospatial data are available, a GeoJSON file
• a preview page with an interactive table displaying the data, concise metadata and links to download the complete data and the pre-processing script.

Example folder structure

dataset/
├── README.md
├── index.Rmd
├── index.html
├── pre-processing.R
├── trafford_dataset_name.csv
├── trafford_dataset_name.geojson

Variables

Data that has been aggregated to a geographical area must supply the area_name and area_code for each observation. These refer to the names and codes used by the Office for National Statistics for statistical and administrative geographies.

Where available, the coordinates of a location or event need to be provided using the WGS84 Geographic Coordinate System (i.e. Longitude, Latitude). The coordinates need to be consistently labelled as lon and lat in that order.

This site is open source. Improve this page.