Data for Development
The Data for Development initiative was active from 2017 to 2023.
The recent explosion in rich data sources and powerful technologies for analyzing them is radically reshaping development research and strategies for building sustainable economies around the world. For example, satellite imagery combined with call data records can track disease outbreaks in even the remotest villages. Social media data are advancing what is known about the genesis of political protests and government responses to them.
The Data for Development (DDI) initiative at the King Center, home to some of the world’s leading development researchers, was uniquely positioned to explore the relevance of these often low-cost, unconventional data streams and to help academics, policymakers, and business leaders craft solutions with real-world impacts on global poverty.
Four Pillars
- Research and collaboration - Enabling and supporting sustained interactions between experts in both data science and development are crucial to harnessing the enormous potential of new data streams. Faculty-led research efforts, for instance, were using high-resolution satellite imagery and machine learning to study “hidden” pockets of poverty globally. A separate project was designing a scalable, low-cost method for accurately predicting crop yields amid climate change.
- Student training - A project-based class brought together Stanford undergraduates studying machine learning and development, while a graduate summer school course convened students in the social sciences from around the world.
- External partnerships - As private-sector companies and public agencies increasingly open up their data to development researchers, the Initiative hosted an annual meeting to foster communication and the sharing of ideas among partners, which include Google, Facebook, and the World Bank.
- Data creation and curation - The Initiative identified the best methods for generating datasets that lead to scalable development programs and was constructing an open-source web platform for development experts worldwide.
Leadership
- David Lobell
DDI Co-Director
School of Earth System Science - Jeremy Weinstein
DDI Co-Director
Department of Political Science - Marshall Burke
School of Earth System Science - Pascaline Dupas
Department of Economics - Stefano Ermon
Department of Computer Science
Affiliated Faculty
- Emma Brunskill
Department of Computer Science - Jennifer Pan
Department of Communications
For more information about the Data for Development initiative, please contact King Center Executive Director Jessica Leino at jleino@stanford.edu.
Putting machine learning to work for development
Researchers of the Data for Development initiative created a new tool to measure poverty by combining machine learning with street-level imagery from 48 countries.
Read the story