Discover how J-PAL's evidence-based approach is transforming poverty reduction through randomized evaluations and capacity building across Africa
In 2007, a remarkable gathering took place in Abuja, Nigeria that represented a quiet revolution in how we confront global poverty. Forty development practitioners from across Africa came together not for another high-level policy summit, but for something fundamentally different: a scientific training laboratory. This course, organized by MIT's Abdul Latif Jameel Poverty Action Lab (J-PAL), aimed to equip those on the frontlines of development work with sophisticated scientific tools to answer a deceptively simple question: What actually works to alleviate poverty?
At the time, this represented a radical approach to social change. Rather than relying on intuition, ideology, or inertia, J-PAL championed a method that had transformed medicine but was rarely applied to social policy: rigorous randomized evaluations. This Nigeria course was part of a broader mission to spread this scientific approach globally, ensuring that policies designed to help the world's most vulnerable would be grounded in evidence rather than assumption. 5
The significance of this approach was later recognized globally when J-PAL's co-founders, Abhijit Banerjee and Esther Duflo, along with longtime affiliate Michael Kremer, received the 2019 Nobel Prize in Economics for their experimental approach to alleviating global poverty. The Nobel Committee noted that their research had "considerably improved our ability to fight global poverty" and that their "new experiment-based approach" had transformed development economics. 5 7
At the heart of J-PAL's approach are randomized controlled trials (RCTs) - the same method used to test new pharmaceuticals. In medicine, RCTs randomly assign some patients to receive a new drug while others receive a placebo or standard treatment. This allows researchers to isolate the effect of the intervention itself, separate from other factors. 5
J-PAL applies this same rigorous methodology to social programs. When testing a new educational intervention, for instance, researchers might randomly assign some schools to implement the new program while others continue with existing approaches. By comparing outcomes between these randomly assigned groups, researchers can determine with high confidence whether the program actually caused any observed improvements.
What is the impact of a specific intervention on outcomes?
Randomly assign participants to treatment and control groups
Deliver program to treatment group only
Measure outcomes for both groups after intervention
Analyze differences between treatment and control groups
The impact of this evidence-based approach has been profound. To date, more than 600 million people worldwide have been reached by programs and policies informed by J-PAL-affiliated research. 7 The organization has grown from its founding at MIT in 2003 to a network of more than 750 researchers worldwide who have conducted over 1,600 randomized evaluations across a wide range of sectors including education, health, governance, and financial inclusion. 5 7
As J-PAL co-founder Esther Duflo explained, this approach takes "big problems and break them into manageable pieces, smaller questions that admit rigorous answers." This stands in stark contrast to "basing decisions on instinct, ideology, or inertia," notes J-PAL global executive director Iqbal Dhaliwal. 5
| Discovery/Impact | Location | Policy Influence |
|---|---|---|
| School-based deworming reduces absenteeism | Kenya | Scaled to reach 280+ million children across multiple countries 5 |
| Direct debit cards reduce poverty more effectively than food distribution | Indonesia | Improved efficiency of social assistance programs 5 |
| Consumer information campaigns protect endangered fish | Chile | Expanded nationwide environmental protection strategy 5 |
| Anonymous résumés can worsen employment chances | France | Prevented nationwide rollout of potentially harmful policy 5 |
| Mobile phones boost math and writing skills | Niger | Demonstrated technology's educational potential 5 |
In 2007, J-PAL partnered with Total and MIT International Science and Technology Initiatives (MISTI) to bring this scientific approach directly to African development practitioners through an intensive course in Abuja, Nigeria. The goal was straightforward but ambitious: to build local capacity for generating and interpreting rigorous evidence about what works in poverty reduction.
The course brought together 40 development practitioners from eight different African countries: Ghana, Kenya, Malawi, Niger, Nigeria, Rwanda, Sierra Leone, and Uganda. The participants represented a diverse cross-section of the development ecosystem, including:
This diverse mix was intentional, recognizing that evidence-informed policymaking requires collaboration across different roles and sectors.
The course was designed to provide both theoretical understanding and practical skills. Through lectures, group work, and case studies, participants engaged with a comprehensive curriculum that included:
A distinctive feature of the approach was its emphasis on integrating evaluation planning into program design from the very beginning. As one participant from Sierra Leone noted, this contrasted with the common practice of treating evaluation as an afterthought.
The immediate response to the course was overwhelmingly positive. Participants reported that the training gave them "a fresh, new outlook" on how to approach development challenges. Colina Kpayagula from Sierra Leone's Institutional Reform and Capacity Building Project noted that the workshop "opened us up to whole new experiences in conducting evaluations" and exposed participants to "different ways of analyzing issues before, during and after the evaluations, to ensure that the results are accurate."
Perhaps most importantly, participants recognized the value of "preparing for evaluations at the very start of the project and not at the end, as we are used to doing." This shift in mindset—from after-the-fact assessment to integrated learning—represented exactly the transformation J-PAL sought to achieve.
"The workshop opened us up to whole new experiences in conducting evaluations... different ways of analyzing issues before, during and after the evaluations, to ensure that the results are accurate."
"A once-in-a-lifetime experience, giving you a fresh, new outlook on addressing development challenges."
The Abuja course was not an isolated event but part of a broader strategy to build sustainable research capacity across Africa. This commitment has continued to grow through initiatives like:
Provides funding, mentorship, and training opportunities to African researchers, with 67 scholar-led projects across 15 African countries to date. 3
Based at the University of Cape Town, this regional office continues J-PAL's work across sub-Saharan Africa through research, policy partnerships, and training. 6
Targeted programs like the Girls' Education and Empowerment Portfolio in West Africa, which ensures policies affecting adolescent girls are informed by scientific evidence. 2
| Program | Focus Area | Key Features |
|---|---|---|
| African Scholars Program 3 | Supporting African researchers | Funding (up to $400,000 for full evaluations), mentorship, training |
| Girls' Education & Empowerment Portfolio 2 | Adolescent girls in West Africa | Partnerships with governments, evidence generation, policy dissemination |
| Jobs and Entrepreneurship in Africa Portfolio 6 | SME growth and job creation | Identifying high-impact strategies for enterprise development |
| Executive Education and Training 7 | Broad evidence capacity | Online courses, in-person workshops, MicroMasters program |
The ripple effects of this capacity-building approach extend beyond individual researchers to influence entire government systems. For example, J-PAL Africa has launched multi-year partnerships with governments like Benin's Ministry of Development and Coordination of Government Action to "co-identify, generate, and integrate evidence into policy across the government." 6
Conducting rigorous poverty research requires both methodological expertise and practical tools. J-PAL has developed an extensive repository of resources that make these tools accessible to researchers worldwide. 4
| Tool/Resource | Function | Application Example |
|---|---|---|
| Randomized Evaluations 5 | Establish causal impact | Determining if a new teaching method actually improves learning outcomes |
| Survey Design Guidelines 4 | Ensure accurate data collection | Designing questionnaires that minimize bias in responses |
| Phone Survey Methods 4 | Adapt data collection to challenging conditions | Conducting research during COVID-19 restrictions |
| Sample Size Calculators | Determine appropriate study scale | Ensuring studies have sufficient statistical power |
| Data Quality Checks 8 | Verify accuracy of collected data | Implementing high-frequency checks during phone surveys |
The Research Methods Initiative—a partnership between Innovations for Poverty Action and Northwestern University—further advances this methodological toolkit by funding research on three key themes: research design, questionnaire design and measurement, and data quality. 8
These resources represent a growing commitment to methodological innovation and transparency in development research, ensuring that the scientific approach to poverty reduction continues to evolve and improve.
The 2007 J-PAL course in Nigeria represented more than just a training event—it embodied a fundamental shift in how we approach the world's most pressing social challenges. By equipping local practitioners with scientific tools, it advanced a model of evidence-informed empowerment that respects both the complexity of poverty and the capacity of those closest to the problems to generate solutions.
The course's legacy continues through the growing ecosystem of African researchers, policymakers, and practitioners who are generating and using rigorous evidence to inform decisions. As J-PAL Africa's work has expanded to include specialized portfolios focused on girls' education, jobs and entrepreneurship, and agricultural development, the foundational principle remains the same: scientific evidence, when properly generated and applied, can transform well-intentioned efforts into effective solutions. 2 6
Perhaps the most powerful impact of this approach has been its demonstration that poverty fighting is not merely an act of compassion but a science that can be continuously improved. As one course participant noted, the training provided "a once-in-a-lifetime experience, giving you a fresh, new outlook" on addressing development challenges.
Nearly two decades after the Abuja course, this scientific approach to poverty reduction continues to gain momentum, proving that the most powerful weapon in the fight against poverty may not be a specific program or policy, but a method for learning what works—and ensuring that knowledge benefits those who need it most.