William & Mary

Color It Plaid

Rob Hicks, Mike Tierney (standing) discuss adding sectors to the data base with Charlotte Jackson '07, Brad Potter '08 and Scott parks '09. Parks is the brother of Brad Parks whose honors thesis became the impetus for PLAID.Inspired by students and driven by their involvement, Project PLAID is a shining example of the power and benefits of undergraduate research.

"We can make "cookbook" research projects for lab classes, but it's not the same as being on an original research project where you keep doing it until you are satisfied and there is no known answer until you get there," said Timmons Roberts, professor of sociology.

PLAID is an acronym for Project Level Aid, an ambitious ever-expanding database of donors and recipients involved in thousands of international-aid projects. The goal is to develop the initiative into a globally recognized, authoritative guide for the international-aid community. A British non-governmental agency, for instance, which is considering lending money to a proposed Nepalese erosion mitigation project, could check the PLAID database for information on the past success and failure of such projects. And, of course, such groups also will be able to check the track records of their projects on PLAID, which is on its way to becoming accepted as a sort of "seal of approval" for aid projects.

Roberts is one of three William and Mary faculty members guiding PLAID, along with BrighamYoungUniversity political scientist, Daniel Neilson. The project grew out of an honors thesis presented by undergraduate Brad Parks to Roberts and faculty members Rob Hicks of the economics department and Mike Tierney, of government. Parks, who graduated in 2003, based his thesis on an examination of environmental assistance to developing countries.

"He wrote the best honors thesis I have ever seen in my life," noted Roberts. "It was 200 pages, 400 footnotes, 400 references or something like that--it was better than a master's thesis, close to a doctoral dissertation."

So impressed were the professors with the paper they encouraged Parks to turn the thesis into a book--Greening Aid: Understanding Environmental Assistance to Developing Countries. Parks asked the professors to participate as co-authors.

Parks' book, due out in fall of 2006, required more additional research than the foursome anticipated. Tracking development assistance can be a challenge. There is a lot of money to follow, countless donor countries and organizations involved and very few tracking entities available.

Following the Money

Annually, more than $100 billion flows to countries across the globe in the form of development assistance from bilateral donors, including the U.S., Sweden and Germany, and from traditional multilateral donors such as the United Nations and the World Bank. The primary tracking agency for this money is the Organization for Economic Cooperation and Development (OECD). The OECD gets its data from its member nations, maintaining a database on their annual reports of aid funding.

While the OECD data is useful, it doesn't tell the whole aid story. As Parks and the professors began to write the book manuscript, they realized that the OECD didn't have all the data the book needed and that they'd have to gather the information themselves-Project PLAID was born.

Missing in the OECD database was a clear guideline of how the projects were classified.

Tierney, an assistant professor in the government department, noted that a project funding the clear-cutting of a rain forest might be put in the same OECD forestry sector as a project that funds tree planting.

"For scholars interested in the environmental impact of foreign aid, such measurement errors create serious problems. The obvious, but difficult, solution, is to gather data directly and classify each project individually using a consistent coding scheme," Tierney said. "The bottom line is that sector codes are simply too tough to capture project-level variation and that different countries don't use the same criteria over time. PLAID does."

The aim of the PLAID database is to bridge this gap in understanding. To do this, the researchers have included information from more donors, more information on each project and have compiled statistics from more years.

"This is one of the most exciting research projects we have going on at the College," said Carl Strikwerda, dean of the faculty of Arts and Sciences. "It is wonderfully fitting that it grew out of a student project and that students have been so fundamental to making it succeed."

Development assistance data may be tracked on one of three levels-aggregate, sectoral or project. Aggregate is the broadest reporting level, while PLAID data is based on project-level reporting. It provides the greatest detail about the individual programs, explained Tierney, assistant professor of government.

The database has been an enormous undertaking. Thus far, more than 60 William and Mary students have participated in the project guided by the team of three professors.

In the course of the project, the student researchers have gone line by line through international development aid grant records for the last 30 years-twice. Each aid project received entries or codes on 71 independent variables-26 more than the projects in the OECD's database.

Prof Rob Hicks examines the work done by Brendan Williams '06. Williams, an economics and goverment major, studied abroad in Spain, where he interviewed government officials for PLAIDGreen and Brown

The PLAID team created the codes based on criteria and scales they established, asking not only how much money was involved and who donated it but also where the money went. In addition, they coded each project on a five-point environmental impact scale, rating the project as either "green"--globally, environmentally friendly or "brown" - providing only local environmental benefit. The researchers also recorded all project funding in terms of constant dollars (adjusted to year 2000 value) and recorded co-financers for projects where applicable.

"The PLAID database contains an immense amount from the OECD but it has been added to in a way that really makes it quite unique without really destroying the integrity of the DAC (the OECD data)," Strikwerda said.

Having the aid data evaluated based on a standard set of criteria readies the information for comparative studies by putting all the information on the same playing field. OECD's data is reported by member countries and does not include data submitted on a standardized scale. The OECD database is "historically a more passive collection system," Tierney added.

The OECD asks for annual reports on aid assistance from its member countries. If the data wasn't filed by the deadline, that information is missing from their database. PLAID filled those gaps by going back to those countries and requesting the missing information and in some cases sending researchers to individual countries to get it.

Configuring data in this way allows researchers to ask more questions of the database including where aid is needed, which projects are working and who cares about the outcome of the aid project?

"This set of data is truly one of a kind," added Miranda Hutten, an undergraduate researcher on the project. "It is also very versatile: academia as well as international organizations can benefit from incorporating this data into their research and project selection strategy."

Initial funding for the project came from grants to the College from the Mellon and Freeman foundations. Then in 2004, the National Science Foundation awarded the project $250,000 over three years. The project has also received significant funding for student researchers through the CharlesCenter.

A comprehensive retrieval of data was made possible, in part, by the generous support of a private donor. These funds have provided support for field research, foreign travel and staff support.

"We simply could not have done all this without these private gifts," said Tierney.

"The PLAID Project data set puts a lot of information in one place," noted Roberts. "It is particularly useful in understanding the impact of a particular sector of aid." The database not only includes information from more donors, but also more information on each aid program and from more years. The students have already detected a downward trend in aid funding projects destructive to the environment, Roberts added.

"This project has given a large number of students experience with real research that is potentially going to have an impact and I think it seems to be along the lines of the model that the College is trying to move in as a center for excellence in undergraduate research," Roberts said. The experience of working with PLAID has paid off.

Kaity Smoot '08 keys in donor data. A double major in international relations and economics, she hopes to work as a development economist in Sub-Saharan Africa after graduation.Research Experience for Students

"I've learned that foreign aid is not at all as simple as it appears," said Brad Potter, '08. "It's easy for one to simply say 'Throw money at a problem and it will be OK,' but after looking at development aid statistics for the last twenty years, it is clear that there is far more to the successful aid equation than just money."

Potter noted that planning, leadership and intent also play a big role in an aid project's success or failure.

Some of the lessons learned have been hard ones: "The most shocking aspect of the project was what many donor countries categorized as 'clean' environmental projects," said Hutten, a member of the class of 2007. "For many donor countries, forestry projects were considered clean projects despite the fact they aided in deforestation. Now I am more interested in the interplay between local development projects and the mitigation of large environmental trans-national issues."

The researchers and the college want to take the data to the next level. As the project moves forward, the team hopes other sectors can be added. They are already talking with other collaborators to add agriculture and health sector funding variables. Adding sectors and the capability to search the database on-line could make the data even more useful.

"We hope to have these kinds of sectors done in a couple of years," Roberts added.

Ultimately, the professors and researchers hope the project will lead to a center for international policy research on campus "so there will be an enduring structure for collaboration across disciplines and including people from outside the university producing useful research," Roberts said.

"PLAID could be a ready tool that would be net searchable and it would help with what is called donor coordination or at least strategic planning of aid," Roberts continued.

Determining the flow of aid is crucial: "On the world stage that would be the biggest contribution PLAID could make," said Hicks, an associate professor of economics.

Donor countries often duplicate efforts, the professors noted. Increasing the coordination of donors is likely to increase the effectiveness of the aid.

"Strategic planning is important because there is quite ridiculous overlap in different agencies ... and they don't even know that the other people are there," Roberts added. "They are repeating their efforts or they are stepping on each others toes and it becomes territorial and counterproductive and extremely wasteful. And for the recipients, for the poor countries, it's a nightmare."

The professors, their researchers and the College see endless possibilities for the project.

"I think it is the most comprehensive database on international development aid that is out there," Strikwerda said. It fills a real need, he added, by providing the additional data on aid that makes the information searchable.

"We have been able to produce a tremendous resource with modest funding with the continued support of the administration and with additional agency and/or foundation support, this project has the potential to be an ongoing and sustainable resource," said Roberts.

Still, whether PLAID becomes an internationally recognized resource on aid funding or not, the project has been immensely successful.

"It marks the future and the direction I think that we want to go as an institution-to continue to encourage these kinds of research projects and to continue to bring students into these kinds of research projects so that they get the most of what I think our faculty are doing," Strikwerda said.

"My favorite thing about PLAID is engaging our students in the process of discovery," added Tierney. "It's just a great feeling when a student figures something out and comes bouncing into my office to tell me all about it."

Good Housekeeping Seal of Approval

Three years into the project, the PLAID team is close to making the database public. They have almost completed coding aid projects reported to OECD from the late 1970s through 2000, nearly 500,000 cases in all. The NSF grant requires the team make the core data available to the public by 2008.

To accomplish this, the College hopes to partner with GuideStar, a nationally recognized clearing house for domestic philanthropy located in Williamsburg.

"My dream would be that PLAID/GuideStar become sort of the 'Good Housekeeping Seal of Approval' for people who are seeking an international development aid project approval," said Carl Strikwerda, dean of Arts & Sciences.

Source: Ideation magazine, Fall 2006, by Suzanne Seurattan. Photo by Cindy Baker.