AidData pioneers data-based policy research for World Bank, Department of State


From outside, the appearance of The College of William and Mary’s Global Research Institute is unassuming. The wealth of research produced by members of the GRI is masked by the residential-looking yellow three-story colonial house nestled near Colonial Williamsburg. 

Housed on the second floor of the GRI is AidData — an independent interdisciplinary research lab. Inside, a small cohort of staff and student research assistants are filling a crucial gap in policy analysis research in the 21st century, having managed more than 90 grants and contracts worth more than $50 million from a variety of multilateral organizations, government agencies and private foundations since its inception in 2004. 

Using data and technology, AidData studies resources, policy and investment around the world — aiding policymakers from Washington D.C., to London, to Brussels. Their Tracking Underreported Financial Flows methodology compiles information from data-poor environments to highlight the economic activity of countries which do not fully abide by global aid transparency measures.

Executive Director of AidData Dr. Bradley Parks described how the research lab provides previously unmet needs to the policymaking community, bridging a divide with academia.

“We are trying to do rigorous empirical research that is directly relevant to decisions that policymakers are trying to make.”

“We are trying to do rigorous empirical research that is directly relevant to decisions that policymakers are trying to make,” Parks said. “A good idea that started in Williamsburg has spread very rapidly and very broadly around the globe.”

Parks also explained how innovations emerging from AidData made ripples through pre-established international distribution networks, starting with a partnership with the World Bank.

In the summer of 2010, then-president of the World Bank Robert B. Zoellick set out a goal to be able to track World Bank projects across the globe and provide feedback in live time through a smartphone app. Afterwards, the World Bank approached a group of faculty members from the College, Brigham Young University and Georgetown University, who then delegated a project to twelve student research assistants for the summer.

The students were asked if they could pinpoint the precise coordinates of World Bank projects in a single country: Kenya. By the end of summer, this small group had georeferenced, or matched aerial images to precise coordinates, for every World Bank Project in not only Kenya, but the entire world.

“It really opened the international community’s eyes in terms of what you could do with data,” Parks said. “Suddenly people were able to see with high levels of locational precision exactly what was happening and where, around the globe. The World Bank was kind of an early adopter. But then what happened was that this geocoding innovation exercise ended up setting an international standard.”

Geocoding is a process in which researchers take descriptive information and match it to a precise geographical location — which AidData has since excelled at.

Following the success of the 2010 project, the International Aid Transparency Initiative, which seeks to increase transparency of how aid money is spent by all major bilateral and multilateral donors through a set of reporting guidelines, found itself without a reporting standard for geocoding projects. They used the methodology that those students created to set a geocoding standard. The number of donors soon increased for AidData.

The following summer, the Ministry of Finance of Malawi used AidData’s methodology to help track all incoming aid projects on the country’s aid management platform. The project’s success allowed the Ministry to successfully track how to rationally allocate scarce resources to ensure that aid was getting to those who needed it. 

The accomplishment in Malawi only increased AidData’s credibility — being able to successfully track how money was being spent across the world.

AidData’s Director of Policy Analysis Samantha Custer explained how the research lab is able to help policymakers answer real world questions using academic methods.

“I work pretty closely with all sorts of external actors, from governments to international organizations, trying to identify their questions and pair them with the best methods and data available to answer them,” Custer said. “AidData has become known as the place where you come when you have questions you want answered but that requires data that does not yet exist and methods that have not been created yet.”

Custer emphasizes the importance of analyzing both the demand and supply sides of development initiatives. She highlights the importance of listening to how leaders themselves in low and middle-income countries view and assess their respective countries’ development goals, in comparison to what may be suggested by experts. Triangulating this information enhances development agencies’ responsiveness and maximizes the impact and analysis of their own aid contributions. 

AidData has also been able to develop some of the leading edge indicators in sectors and spaces where objective information and evidence is in short supply.

“You can quantify and measure these things,” Custer said. “We have done that in everything from filing Chinese development finance to quantifying Russia’s media influence, to trying to understand the perception of leaders — these are areas where most people have not touched.”

After increasing credibility internationally, AidData’s innovative research has attracted direct collaboration with multilateral organizations such as the World Bank and International Monetary Fund, government agencies including the United States Agency for International Development and the US Department of State, as well as a number of foreign governments. AidData also collaborates with private foundations, recently receiving a $4.74 million investment from the Bill and Melinda Gates Foundation.

Custer cites the importance of collaborative discussions between AidData and its funders. In Custer’s eyes, AidData’s audience is two-fold: the first is the immediate audience which seeks to use the lab’s findings to inform a specific goal for its funders, the second is the broader use of the data in general, rooted in increasing transparency of information worldwide.

“We are super transparent in our methods and in how we collect the data that we do, what we do to the data when we receive it and how that is analyzed because we view that it is important for credibility and for use,” Custer said. “We want to make it as easy as possible for as many people to use.”

Senior Research Scientist Dr. Ammar Malik further explained the relevance of AidData’s research and its increasing importance in a changing world, particularly as an increasing amount of aid is being distributed to developing countries around the world. 

Malik joined AidData last year after previously working in data science for AstraZeneca, but his background was not originally in the private sector. With China expanding its reach and involvement with foreign aid projects, such as the Belt and Road Initiative, Malik’s interest grew surrounding the secrecy of China’s aid development.

“My PhD is in public policy,” Malik said. “I studied cities, urbanization and transportation. So I was always intrigued by the China angle, which was, ‘Why are the Chinese funding so much infrastructure in so many places? What kinds of effects are they having? Is anybody studying that? Is anybody even caring about the design of those projects?’”

Malik described the international aid transparency system which has existed following the Second World War, a set of practices among Western nations to publicly track the aid which has been sent to developing countries around the world.

“One organization that does this is the Organization for Economic Cooperation and Development ,” Malik said. “They have a creditor reporting system — it’s a gigantic database. Japan, Australia, and other countries all report their giving there. It’s downloadable, freely available to anybody.”

AidData Junior Program Manager Thai-Binh Elston ’21 laid out AidData’s role of tracking underreported financial flows in the context of the current state of international aid transparency. 

“The OECD has… certain reporting guidelines for their foreign aid. So it is more transparent and open to scrutiny,” Elston said. “Emerging donors like China, Saudi Arabia and Brazil do not report according to the same guidelines. So their foreign aid activities aren’t as transparent as other kind of ‘traditional donors.’”

The OECD currently has 38 member countries, including the US, United Kingdom, France, Germany, and more.

“China also doesn’t know exactly how much money they are spending. They have, to our knowledge, no internal system of reporting for overseas financing,” Elston said. “It’s all publicly available information. We are kind of triangulating all that information into one neat database.”

Malik continued to discuss the importance of AidData’s research in regards to China.

“In the last 20 years, [China] has now become the single largest provider of development assistance [in the world] — the volume of help that they provide is twice that of the US,” Malik said. “It’s a two-to-one ratio.”

“It does create a problem,” Malik said. “If you do not know what China is doing, how are you supposed to adjust your spending? So that is where we come in… China is not undertaking this reporting, but we already have a system, definitions, guidelines, etc. So we can just go in and find that information in official and open source documents… whatever we can find, and then put it back into the same format that is set up by the OECD and followed by many other countries around the world’”

“The cool thing is that this data is in high demand. Nobody does it at the level of detail, granularity and specificity as we do.”

“That’s what this is about at the end of the day,” Malik said. “The cool thing is that this data is in high demand. Nobody does it at the level of detail, granularity and specificity as we do. That’s shown by the fact that we have so many media requests coming in. And policymakers from the State Department to the Department of the Treasury are always asking us to give them insight on China’s behavior around the world.”

The faculty and staff of AidData do not work alone, however. Their key findings have resulted from the important contributions of the many undergraduate students who help collect and analyze data, in order to focus on what is truly meaningful and increase efficiency.

AidData has the largest student research assistant team at the GRI, with a little over 50 students currently working on projects.

Students researchers alongside Dr. Ammar Malik. COURTESY IMAGE // SEAN NGUYEN

Student Research Assistant Sean Nguyen ’24 works under the TUFF team for AidData. He has experience working on tracking Chinese foreign aid contributions around the world, some of AidData’s most policy-relevant work.

“Essentially every time there is a transaction between the Chinese government or a Chinese institution with another institution, regardless of where it occurs in the world, my responsibility is to track where that locational geographic instance is for the recipient,” Nguyen said.

For AidData that means analyzing publicly available documents, and also those which can be accessed in other means, to track where resources are being allocated and used around the world.

This includes analyzing media reports, annual budget reports, financial statements, as well as loan contracts between banks and state and private entity borrowers. Many of the projects being funded focus on building  public infrastructure or natural resource extraction across developing countries, especially the Global South. 

The geocoding process usually requires deciphering many of these documents by translating texts from a multitude of world languages into English.

“It gives a kind of holistic picture where you can see where all the tracking and these documentations occur,” Nguyen said. “You can begin to see patterns and correlations that exist.”

The role of students has evolved over the years, especially as the capacity and resources of the research lab has grown. Their responsibilities evolved from low discretion work of simply compiling information to even helping create classification schema for how to define and find new sources of data.

AidData also cultivates a great atmosphere for student researchers, according to current and former students. 

“The GRI has a lot of intelligible, sophisticated and eloquent individuals who you can network, create long-lasting friendships and build future opportunities with,” Nguyen said. “These conversations have spurred me into a better direction of what I want to do with my future career.”

Nguyen wants to work in public policy, authoring policy related to education, and cited how working with the research lab has helped him evaluate his career path. 

“AidData allows me to have a lot of understanding about how institutions think, about how policy should be data-driven and utilized in the most leveraging manner,” he said.

Many former students have also returned to AidData and now work as staff members. Elston returned to the research lab after graduating and now works with undergraduate students directly, who has also expressed an interest in creating evidence-driven policy in the future.

“I love mentoring students and training them, teaching them the methodology, because I have been in their shoes before,” Elston said.

She applauded the immense benefits of student researchers in helping AidData meet deadlines to deliver data to collaborators. 

“It would be possible, but it would probably take decades, or something like that for us to compile this dataset,” Elston said. “We need at least 50 research assistants at any given time to be productive.”

Elston also discussed what is so rewarding about working for the research lab. 

“The best part about AidData is that you are actually having an impact on policy.”

“The best part about AidData is that you are actually having an impact on policy,” she said.

In terms of the future for AidData, it appears the sky’s the limit — especially as strategic competition between the US and China seems poised to continue into the 21st century according to Malik.

“I think the demand for AidData’s work, particularly [regarding] China, is just going to continue to grow,” Malik said. “The bottom line is that the competition is going to stay and the demand for analytics will be there… the volume of data that’s coming out that we can capture is also increasing.”

“I think in the next few years AidData will be seen not just as a group that aggregates and collects data, but also as a group that provides important insight that helps governments make big decisions around this issue of competition with China,” she added.

Parks elaborated about expanding the full-time AidData staff in the next few years to pursue a wider set of goals. That starts with what Parks described as “building a deeper bench of intellectual leaders.”

The research lab currently has nine principal investigators, each leading an entirely different line of empirical inquiry. Parks highlighted AidData’s purposeful selection of a diverse set of intellectual leaders to broaden and strengthen the research lab’s impact in years to come.

“We have kind of reached a point where it’s time to decentralize authority to a broader set of principal investigators…to take forward the next generation of research at AidData,” Parks said. “This is an ongoing process.”


Please enter your comment!
Please enter your name here