Get Started
openwashdata publishes data packages with you and teaches the skills to do it yourself. Start here.
Is your dataset ready to publish?
Tick what you already have. The first five items are what a review needs before it starts; the last two are what publishing adds. Nothing you tick leaves this page.
Where you stand
Start with the raw data.
A short version of the items that washr::check_publication_readiness() tests. Run it inside your package for the full list.
Questions people ask before their first package
Who can publish through openwashdata?
Anyone connected to the greater WASH sector: utilities, NGOs, researchers, students, consultants. Public health, solid waste, faecal sludge management and air quality count too. The data has to be yours to share, or shared with the permission of whoever owns it.
Do I need to know R?
No. If you can send us a spreadsheet and answer questions about the columns, we can publish with you. If you want to do it yourself, the washr package writes every file with a fixed shape, and the free course teaches the rest.
What about personal data?
Nothing is published before a person has checked every dataset for direct identifiers (names, phone numbers, IDs, household GPS), for household or person level rows, and for sensitive contexts. The check covers the git history as well, because a removed column stays recoverable. If data is sensitive, we help you aggregate it or publish only the dictionary.
Who owns the data, and under which licence?
You do. You stay the author and the copyright holder. Packages are published under CC BY 4.0 by default, which lets anyone reuse the data as long as they credit you. The DOI and the citation make that credit straightforward to give.
How is a package reviewed?
Against a versioned review standard in four areas: metadata, data, documentation and tests. A script checks the mechanical items, a reviewer checks the rest, and each area gets one GitHub issue and one pull request. A person approves every step, and the package records the standard it was reviewed against.
How do I cite a package?
Every package website has a citation block, and the Zenodo deposit gives it a DOI. In R, citation("packagename") prints the reference. The data catalogue links every DOI.
How long does it take?
With the raw data, a dictionary and a check for personal data in hand, a package has gone from spreadsheet to reviewed publication in a day. We wrote up one such day. Most take a few weeks, because the prose about the data takes thought.