• Home
• Research
  • Projects
  • Publications
  • Randomization
  • Data
• People
  • Directors
  • Members
  • Staff
• Courses
• About
  • Mission
  • Background
  • News
• Regional Offices
  • Latin America
  • South Asia
  • Europe
• Get Involved
  • Jobs
  • Collaborate
  • Partners

Abdul Latif Jameel Poverty Action Lab: Summary of Mission and Activities

The Abdul Latif Jameel Poverty Action Lab (J-PAL) is a center within the Economics Department of the Massachusetts Institute of Technology whose aim is to reduce poverty by ensuring that policy is based on scientific evidence. J-PAL and its network of affiliated professors around the world runs and promotes the use of randomized evaluations, similar to clinical trials used in medicine, to rigorously test the effectiveness of social programs and policies aimed at reducing poverty.

J-PAL was founded on the belief that there would be enormous potential to improve the effectiveness of poverty programs if there was more and better evidence on the most cost effective approaches for tackling poverty. For example, should limited education budgets be spent on textbooks, teachers, or smaller classes? Is increased external auditing or more community monitoring the best way to reduce corruption in local building projects? What is the most effective way to promote safer sexual behavior in communities at risk from HIV?

Partnering with governments, private companies, and nongovernmental organizations, J-PAL researchers answer key policy questions using the most rigorous scientific methods. A range of new techniques (many developed by J-PAL researchers) allow these studies to be designed around the practical constraints faced by local partners working in some of the poorest places on earth.

But J-PAL’s work goes beyond the evaluations themselves, seeking to change the way that poverty policy is made—making it more evidence based and more effective. To achieve this objective, J-PAL undertakes three related activities:

1. Running randomized trials of specific policies and programs to generate evidence about what works;

2. Building the capacity of others to run randomized trials further expanding the evidence base on which policy makers can act;

3. Disseminating the results of high quality randomized evaluations carried out by J-PAL’s and others to policymakers and drawing out general lessons.

(1) Running randomized trials of specific policies and programs to generate evidence about what works

J-PAL affiliates are currently running over 108 randomized evaluations in 30 countries, including Malawi, Zambia, Ghana, Morocco, Peru, Colombia, Mexico, Indonesia, Philippines, and India. Over 70 evaluations have already been completed. The majority of these evaluations have been in the areas of health, education, and savings and credit. J-PAL has also been successful at pushing the envelope on how to evaluate difficult to measure outcomes – for example, in the areas of governance, accountability, participatory decision making, gender empowerment, and building trust in post conflict environments.

(2) Building the capacity of others to run randomized trials so that more evidence can be generated

J-PAL offers intensive one-week executive education courses on randomized evaluations at each of its regional offices annually, as well as one=off courses (including in Indonesia, Nigeria, and DC). Over 500 people from 28 countries have been trained through these courses. J-PAL is currently in the process of creating an online version of its course, and we anticipate eventually having full online versions in at least English, Spanish, and French.

Through these courses, we have raised awareness of the need for more rigorous evidence in policymaking and for randomized impact evaluations as a way of generating it. The governments of France, Britain, Morocco, and India, the Millennium Challenge Corporation, Care International, Save the Children, Freedom from Hunger, BRAC, and the International Rescue Committee (IRC) have all embarked on, or are about to embark, on their first randomized evaluations of social programs, nearly all with the direct encouragement and support of J-PAL, and following the lead of many smaller NGOs. Many of those running or supervising the randomized evaluations in the agencies have been trained through our executive training course. The World Bank has also dramatically scaled up its randomized evaluations. Many of the early World Bank randomized evaluations were joint with J-PAL researchers.

(3) Disseminating J-PAL’s research findings to policymakers

J-PAL does this by disseminating the results of research to policy makers, synthesizing results across a range of studies and drawing out the policy lessons, and taking these lessons to policymakers. J-PAL distributes research findings through print and web publications, and staff who focus on policy outreach. J-PAL also synthesizes results across studies, providing comparative cost effectiveness data which demonstrate for example, what is the most cost effective way to increase attendance at school or reduce diarrhea in children. This analysis across all the randomized evaluations in an area identifies programs that are candidates for scale-up. Finally, J-PAL promotes scale-up of specific approaches that it has found to be highly cost effective and replicable through targeted scale-up campaigns.

For example, J-PAL’s findings on the cost effectiveness of school-based mass deworming led to the formation of Deworm the World, an initiative of the Young Global Leaders at the World Economic Forum. As a result of these efforts a large scale school based deworming program was launched by the Government of Kenya this year through which 3 million children are being dewormed. The World Food Program has committed to add deworming to all their school feeding programs in areas that have high intestinal worm loads, and the microfinance organization SKS is due to launch a program to deworm 1 million children in Andra Pradesh this summer.

Evidence from a J-PAL study that bednets given away for free were used by pregnant mothers just as often as bednets that were sold at subsidized prices (contrary to claims made by many), and that charging for nets cut take-up dramatically has had a profound effect on the debate and policy on this issue. Population Services International changed their policy on pricing of nets to give them away for free in Kenya and Somalia. The bednet study is one of a series of J-PAL studies that have shown how even small user charges can dramatically affect access to services for the poor.

Based in part on strong evidence of effectiveness from a J-PAL, study Pratham (an Indian education NGO) raised funding for a massive scale up of their “Read India” program. Twenty million children have already benefited from this program.