Open Data and Crowdsourcing in International Development

Aid and development work runs on information: who is doing what, where, with what money, and with what result. For a long time much of it sat in separate databases, behind subscriptions, or in reports few people read. Two ideas pushed against that. Open data means publishing information so anyone can use it. Crowdsourcing means collecting information from many people instead of a few experts. Three dated examples show how both have played out.

A BarCamp on aid transparency (2009)

On June 16, 2009, Giulio Quaggiotto announced on the World Bank’s blog that the Bank would host an International Development Data BarCamp on July 10. A BarCamp is an informal, open meeting where participants set the agenda. The announcement said the aim was to brainstorm about making aid more transparent, improving access to data, and making development knowledge and tools easier to find on the internet. The stated focus was being able to see who is doing what, where they work, what they have learned and where the money comes from.

The World Bank opens its data (2010)

On April 20, 2010, the World Bank Group launched its Open Data Initiative. It gave the public free access to more than 2,000 sets of statistics that had previously been restricted to paying commercial users. World Bank President Robert Zoellick said statistics tell the story of people in developing and emerging countries and can help overcome poverty. The practical change was simple. A researcher, journalist, student or small organization could now download figures on living standards without paying for access.

Crowdsourcing in a crisis: Ushahidi

Ushahidi started in Kenya after the disputed 2007 presidential election. Its founders, Erik Hersman, Juliana Rotich, Ory Okolloh and David Kobia, built a way to collect eyewitness reports of violence by email and text message and place them on a Google map. The name is Swahili for “testimony” or “witness.” The platform is open source software that gathers user-generated reports and maps them.

After the January 2010 Haiti earthquake, Patrick Meier coordinated a deployment that brought together Ushahidi, universities and humanitarian groups. As summarized on Wikipedia, nearly 40,000 independent reports reached the Ushahidi Haiti Project, and nearly 4,000 distinct events were plotted on the map.

What these examples have in common

• The data existed before the tool. In each case the valuable part was information that someone already held or could gather.
• Access, not storage, was the barrier. The Bank’s data was not new in 2010; being free to use was.
• Many hands. The BarCamp and Ushahidi both rely on many participants contributing what they know.

Limits worth knowing

Open data does not explain itself. Raw figures need context, and a map of reports shows what people sent in, which is not the same as a complete count of what happened. Crowdsourced reports also need checking before anyone acts on them. Plan for that work in advance.

Where to start

If you work with a small organization, you do not need a platform. Look for public datasets in your field, such as the World Bank’s data site, and ask what question a figure could help you answer. If you collect information yourself, decide early whether you will publish it, in what format, and who will keep it up to date.

Sources

• Giulio Quaggiotto, “Coming your way: International Development Data Barcamp @ the World Bank,” World Bank blog, June 16, 2009: https://blogs.worldbank.org/en/psd/coming-your-way-international-development-data-barcamp-the-world-bank
• World Bank Group, “World Bank Group launches Open Data Initiative,” April 20, 2010 (World Bank timeline entry): https://timeline.worldbank.org/en/timeline/eventdetail/2046
• Ushahidi, overview and Haiti 2010 deployment (Wikipedia): https://en.wikipedia.org/wiki/Ushahidi