Tahir Andrabi, a professor of economics at Pomona College, and his research partners have spent more than a decade working to understand education at the village level in Pakistan. The result is the Learning and Education Assessment in Punjab (LEAPS) Project, the most comprehensive and systematic dataset on education in that country.
Last month, the team was awarded a three-year, $745,000 grant from the World Bank Strategic Impact Evaluation Fund (SIEF) to examine the financing constraints of private schools in Pakistan and evaluate planned interventions for these fiscally vulnerable schools, including small grants and specially designed micro-financing programs.
Also in September, the team's latest research paper "Report Card Grades: The Impact of Providing School and Child Test Scores on Educational Markets," in Pakistan, was selected one of four winners in the World Bank's first Research Academy Competition, which was designed to encourage researchers to take the extra steps necessary to share their results and insights through research papers that could be more publicly broadly shared.
The study by Andrabi's team "provided report cards with school and child test scores, prices and enrollments in 112 villages with multiple public and private providers," and then they made the results public.
"What we found, says Andrabi, "is that the information increased competition in the marketplace dramatically. In the treatment villages, test scores went up significantly. There were small increases in enrollment and private school fees decreased by 20 percent. Most of the gains in test scores came from schools that were below the median. The worst performing schools were where most of the improvements came from. Schools that came out ahead put up banners advertising their results. This really galvanized the marketplace."
While report cards are standard in the U.S., this is not true in Pakistan or much of the developing world. Distributing the information collected during their study was complicated by the fact that more than half of the households are illiterate.
"We provided very detailed information to the parents and the schools," says Andrabi, "We had tests scores for each kid related to their class, each school to others in the village and each village in the district. We had to share that information in large group meetings. You can't mail the report cards because the parents can't read. We also had smaller focus meetings with parents and schools."
Both the grant and research paper grow out of research Andrabi and his research partners Jishnu Das, senior economist of the World Bank's Development Research Group, and Asim Ijaz Khwaja, a professor of public policy at Harvard University's Kennedy School, began in 2003.
"It used to be that there was one government school and the question was who got access," says Andrabi. "But now, here and in many other developing countries, there has been an explosion of low-cost private schools, mom and pop schools, which are much more prevalent than religious schools. The teachers are local young women, and schools are the largest employers of women in the formal sector. It's a grass roots phenomenon." Between 2001 and 2005, there was a 10 percent increase in school enrollment. "What's interesting is that there is now a marketplace for education especially for the poor. We were amazed by this."
Pakistan has an estimated population of more than 180 million people with a literacy rate of 50 percent and a primary school enrollment rate of 66 percent. More than half of the population is under age 17.
The LEAPS Project
In 2003, the research team conducted a survey of 120 villages in Punjab and counted more than 800 schools, 30-40% which were private. They also interviewed more than 4,000 teachers. In 2004, they began in-depth data collection, surveying 2,000 households with children over age seven on a range of factors including number of children, their ages, the family's economic level, health and nutrition levels, the mother's time investment with the children and the learning environment in the home. They also tested approximately 12,000 third-grade children and interviewed their corresponding teachers (about 400). They tested the same children and surveyed the same teachers and households in 2005, 2006 and 2007, adding a second group of third-graders and their teachers in 2006.
In 2011, the team returned and tested all 25,000 children, interviewed the same 800 teachers and surveyed the same 2,000 households. By this time, the first group of children were 17 years old, which meant the team could begin to collect data on outcomes, such as marriages, economic status, drop out rates and causes.
"Really we have some of the best and most developed datasets on education in the developing world because we linked education to household characteristics, teacher characteristics, school details – size, budget, number of teachers, class size, resources – and village inequality," explains Andrabi. "Already the public data has been downloaded more than 600 times.
"LEAPS started as a project and has morphed into a broader agenda that includes a host of people working with the data, some on their own," says Andrabi.
"We now know a lot about the kids and the schools and the market. The next step is to figure out a mechanism that will start to improve school quality, what's cheap and what can be scaled up. Where do these schools need help?" The SEIF grant and support from the Aman Foundation, in Karachi, and Tameer Bank will take us to the next phase. Already there are about 30 micro-loans in the pilot project.
A whole generation of Pomona students has worked on this project, says Andrabi. Three of the team's research assistants, Ben Daniels '11, Nick Eubank '07 and Tristan Zajonc ‘02 have worked at the World Bank. Zajonc has co-authored a related paper, while both Daniels and Michael Stewart '13 are co-authors on forthcoming papers. Eubanks is using the dataset for his PhD at Stanford.
The LEAPS Project and related work has been supported by the World Bank, the National Science Foundation, the John Templeton Foundation, the Aman Foundation (Karachi), Tameer Bank, Dr. Imran Shah, Pomona College and Harvard University, among others.
Andrabi joined the Pomona College faculty in 1988 and teaches courses in economic principles, economic development, international economics and game theory. He has published extensively in major economics and education journals. In 2007, his work on religious education in Pakistan received the George Bereday Award for the best paper published in Comparative Education Review in 2006 from the Comparative and International Education Society.
He was also the principal investigator on a National Academy of Sciences/Higher Education Commission Pakistan grant on evaluating the recovery from the 2005 northern Pakistan earthquake. He co-founded the RISE-PAK website to help coordinate relief in the aftermath of the October 2005 Pakistan earthquake. The website was awarded the Stockholm Challenge Award (2006) for the best information and communication technology project in the public administration category.
Andrabi is currently the director of social programs for the Center for Economic Research and Policy in Pakistan (CERP) and a member of the executive committee of the Association for Analytic Learning about Islam and Muslim Societies. He is a graduate of Swarthmore College and holds a Ph.D. in economics from the Massachusetts Institute of Technology.