Local evidence changes policy: Research highlights how Senegal accomplished groundbreaking bilingual education reform

Author: Josh Stowe

Students are pictured inside a classroom in Senegal.
Notre Dame researchers have studied how Senegal embraced bilingual education. The country's experience can inform broader education policy. Photo credit: Associates in Research and Education for Development.

In Senegal, fewer than 1 percent of adults speak French as their first language. Yet this West African country, home to 39 living languages, has traditionally required students to learn in French from their first day of school.

The consequences are stark. Almost 70 percent of students can’t read a simple sentence by age 10, according to the World Bank, and by middle school, UNICEF finds that only 9 percent are reading at the minimum proficiency level.

Research has long shown that first-language-based bilingual education, which uses a student’s first language as the foundation for learning additional ones, yields better academic outcomes. But multiple government reform attempts to embrace the use of local languages had fallen short.

In recent years, that changed: A particularly effective pilot program by the Senegalese nongovernmental organization Associates in Research and Education for Development helped convince policymakers that bilingual education could be successfully implemented. The program demonstrated that a bilingual model with French and local languages not only improved math and reading scores in local languages, but also helped participants build stronger French language skills than their peers in French-only schools.

Now, policymakers are in the process of scaling bilingual education across Senegal. Researchers at the University of Notre Dame have studied how the country made that shift, and what it can teach policymakers about turning evidence into lasting educational reform. Their study traces how locally generated evidence helped overcome skepticism about bilingual education and create political momentum for change.

Published in the International Journal of Bilingual Education and Bilingualism, the research offers lessons for countries working to improve the quality of educational outcomes in low-resource, linguistically diverse contexts.

“Globally, there is growing evidence that this approach to bilingual education improves learning outcomes, but it is rarely adopted or implemented at scale,” said lead author Erina Iwasaki, a postdoctoral research associate at the Pulte Institute for Global Development, part of Notre Dame’s Keough School of Global Affairs. “This raises a crucial question: How can this policy become a political priority in more countries that would benefit from it? Our work provides answers.”

Reform that resonated

Mamadou Amadou Ly is pictured. He received global recognition for his work with Associates in Research and Education for Development. The nongovernmental organization showed that bilingual education could be successfully implemented in Senegal.
Mamadou Amadou Ly received global recognition for his bilingual education work in Senegal. Photo credit: Associates in Research and Education for Development. 

The researchers studied Senegal, which presents a compelling case study on how and why bilingual education was able to garner political support when most other similar contexts have failed to do so. For years, educators and policymakers had seen global evidence on this approach and knew, in theory, that it was a better option.

But two government-led attempts to teach students in national languages, from 1977 to 1984 and from 2002 to 2008, both struggled and were perceived as failures. The programs collapsed because of inadequate teaching materials, insufficiently trained teachers and parental skepticism, researchers found, undermining decision makers’ confidence that bilingual education could work in practice.

Then, from 2008 to 2019, Associates in Research and Education for Development partnered with the Senegalese government on a program that taught students bilingually across nine school districts, reaching more than 10,000 students. The program supported languages such as Wolof, spoken by approximately 80 percent of Senegal’s population, and Pulaar, the country’s second-most widely spoken language.

The organization deployed this program under the leadership of Executive Director Mamadou Amadou Ly, whose pioneering education work has gained global recognition. In 2025, Ly received the prestigious Yidan Prize for Education Development and Time magazine named him to its list of the 100 most influential people of 2026.

“When we began working with students, we encountered skepticism from certain government officials who had been educated in French,” Ly said. “They feared a backlash from parents, given past experiences. Some argued that neighboring countries — which had introduced national languages into their systems long before Senegal — did not, in their view, possess educational systems that outperformed Senegal’s.

“But one evaluation after another showed the same results: Students who learned in their own language performed better. The results vindicated our approach.”

Key findings from Senegal’s success

Senegal is pictured on a map.
Senegal has 39 living languages, yet traditionally required students to learn in French from their first day of school, even though most adults do not speak it as their first language.

Researchers made several findings that are important for the broader field of multilingual education:

Difficulty implementing bilingual education reforms is a key barrier that prevents them from being politically prioritized. The efficacy of first-language-based bilingual education is now widely accepted in global policy circles, but in Senegal, repeated implementation issues had raised doubts about its feasibility.

Local evidence is key to making reform happen. Researchers found that effective, locally grounded proofs of concept functioned as a “causal hinge” between longtime advocacy for bilingual education and broader political prioritization. Researchers said the pilot addressed a reluctance to change the status quo by showing that bilingual education was not only theoretically preferable, but practically feasible.

Political advocacy, on its own, is not enough to effect change. Researchers found that the collective power of supporters — including the decades-long advocacy of scholars, civil society leaders and government officials, as well as the support of international partners — was necessary but not sufficient. International partners tended to play a secondary support role in the often contentious political debates around language of instruction, they said.

“Although policymakers pay attention to the global evidence, it is not the deciding factor in changing policy,” said study co-author TJ D’Agostino, associate professor of the practice in the Keough School and a core-affiliated faculty member of the school’s Pulte Institute. “International organizations played an essential role with funding and technical assistance but, in the end, it was local leaders like Mamadou Amadou Ly and local proofs of concept that made the difference.”

Language barriers in education

More than 250 million people worldwide lack access to education in a language they speak and understand fluently, according to UNESCO.

First-language-based multilingual education addresses that gap by using a student’s first language as the foundation for learning additional ones.

Research shows the approach improves literacy and numeracy outcomes. In addition, students taught this way tend to learn a second language more effectively than peers taught through immersion alone.

Informing multilingual education policy and practice

Notre Dame researchers Irina Iwasaki and TJ D’Agostino are pictured. Photo credit: University of Notre Dame.
Notre Dame researchers Erina Iwasaki and TJ D’Agostino are pictured. Photo credit: University of Notre Dame.

Senegal’s experience can help inform broader multilingual education reform, researchers said. The study is part of a larger Notre Dame research project led by D’Agostino, Language of Instruction Transition in Education Systems, that explores how education systems can successfully facilitate students’ transition from learning in a local language or bilingually to learning primarily in an international language. The study spans six countries: Kenya, Mali, Mozambique, Rwanda, Senegal and the Philippines.

Researchers recommended that national reform prioritize quality implementation alongside efforts to seize the political momentum to drive reforms forward. They noted that poor implementation can erode support and cause policy reversal — as happened in the Philippines, where rapid scale-up outpaced the quality of implementation.

In addition, they recommended that interventions make adjustments quickly to keep implementation on track.

“Our study underscores the benefits of implementation research, realist evaluation methods and design-thinking approaches,” D’Agostino said. "This can allow reformers to learn and fail fast and small rather than failing at scale."

Researchers also underscored the need for sustained support from the donor and policy communities.

“Our work shows that advocates need to build support for new policies across one or more local pilot programs before they can have real policy traction.”

“Our work shows that advocates need to build support for new policies across one or more local pilots before they can have real policy traction,” Iwasaki said. “This suggests a longer view of development and policy impact. We can't work on the five-year funding cycles that used to be the norm. We need to think across 20 to 30 years to really have something that can make systemic change.”

The study received funding from USAID and from the Notre Dame Poverty Initiative, which supports research partnerships with communities around the world to study poverty from every angle and help find what works to build lasting opportunity.

The research aligns with the broader work of D’Agostino, a fellow of the Keough School’s Kellogg Institute and the University’s Institute for Educational Initiatives. He studies how literacy interventions can improve educational outcomes and life opportunities in contexts that include sub-Saharan Africa as well as South Bend, Indiana.

D’Agostino’s work with Ly and Associates in Research and Education for Development has continued with support from the Poverty Initiative. He recently co-authored a study of the nongovernmental organization’s summer remedial learning program. This study’s lead author, Fernanda Soares, assistant research professor with the Global Center for the Development of the Whole Child, part of Notre Dame’s Institute for Educational Initiatives, recently presented results at Oxford University’s What Works Hub for Global Education.

Additional co-authors of the Senegal education reform study included several researchers affiliated with the Keough School’s Pulte Institute and linguistics experts from Senegal: Maria Teel, the late Moussa Daff, Mbarou Gassama Mbaye, Aliou Ngoné Seck and Fary Silate Ka.

Originally published by Josh Stowe at keough.nd.edu on October 05, 2026.