D.Talks: Big Data in Development

By Lily Lyman, Angela Rastegar, and Dan Zook

In 2011, the world produced more data than in all previous years combined. Collected through digital platforms such as social media, mobile phones, and other technologies, “big data” – or the massive amount of digital information that is created in real-time as a by-product of people’s activities and transactions globally – presents a unique opportunity for companies to understand what people want and how they think.

Robert Kirkpatrick, Executive Director of the UN Global Pulse, presents at D. Talks in New York

It also presents great opportunity in international development, which, historically, has been data-starved. Last Thursday, Dalberg hosted guest speaker Robert Kirkpatrick, Director of the UN Global Pulse, for its inaugural D. Talks event in New York City. The UN Global Pulse, an initiative of the UN Secretary General, is harnessing the opportunities afforded by big data and real-time analytic technologies to help understand changes in human well-being. Mr. Kirkpatrick presented some of these exciting developments and global trends, and discussed how they can better inform and guide decision-makers in the global development space.

The Global Pulse team has discovered numerous examples in which this “new data” relatively accurately predicts trends in development issues, including the food prices, unemployment rates, and rise in conflict levels. For example, one of Global Pulse’s first data-mining research projects illustrated that changes in sentiment expressed by people in social media (particularly, blogs, online news, and discussion forums) can be a several-month leading indicator of unemployment spikes in the US and Ireland. However, a strong understanding of local context is essential to accurately correlating behavior changes captured with “new data” with economic trends. Global Pulse is establishing Pulse Labs in several countries around the world, including Uganda and Indonesia, which will serve as hubs for testing new techniques for collecting and analyzing new data sources that could be useful for development. These labs sit within UN country teams and work with a range of local actors, including governments, civil society organizations, and NGOs.

Mr. Kirkpatrick said that real-time data analysis is beneficial for the monitoring and evaluation of policies, programs, and projects in global development. Traditional M&E programs involve piloting a program several years and then analyzing results or reviewing existing metrics monthly or annually. ”New data” sources can be used to identify successes or challenges in real-time, allowing for immediate course corrections and program reforms.

This approach to data offers an opportunity for the private sector that Kirkpatrick called “Data Philanthropy.” Large private sector companies who own much of the valuable data can partner with development players such as the UN Global Pulse to share their data and coordinate on what type of information is captured. Since the data is used to better understand markets and influence policies, collaboration can benefit both development agencies and the private sector.

Mr. Kirkpatrick was Dalberg’s first D. Talks speaker, a new initiative that works to bring together the development community in each city where Dalberg is located. D. Talks aims to drive dialogue and critical thinking on global development issues and provide networking opportunities within the community.

Lily Lyman (Consultant), Angela Rastegar (Senior Consultant) and Dan Zook (Project Leader) are all based in Dalberg’s New York office, and serve as the New York D. Talks coordinators. For more information on the D. Talks program, contact lucy.mele@dalberg.com.

This entry was posted in Dalberg Community and tagged , , , , , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>