Parental Involvement Project: Group photo at the end of surveyors' training.
I grew up in a rural area without hospitals, electricity, or running water, and I was the first in my family to attend formal schooling. This background shapes my research approach. I study development challenges through close attention to institutions, incentives, and behavior, informed by lived experience. My work focuses on expanding people’s agency in education, fertility, and child marriage. Each project is grounded in real-world contexts and motivated by concrete mechanisms observed in practice.
Activating Parental Managerial Capital: How Role Clarity Improves Education Outcomes--(Under Review)
Why do some households fail to translate educational investments into learning? This paper introduces parental managerial capital—the ability to organize children’s time, monitor effort, and sustain routines that support learning—as a productive input in human-capital accumulation. In Benin, I first document a large intention–implementation gap: most parents say that monitoring attendance, enforcing lesson review, and tracking progress are their responsibility, yet many do not implement the corresponding routines. I then test whether relaxing managerial-capacity constraints improves outcomes using a randomized phone-based coaching intervention that provides a simple, literacy-independent managerial technology. The intervention substantially narrows the intention–implementation gap and raises achievement: GPA increases by 0.11 standard deviations, and grade completion rises by 12 percentage points among students initially at risk of repetition. Although the program provides no grades or performance records, treated parents become more accurate about their child’s academic standing, suggesting that stronger household management generates information endogenously through monitoring and school contact. The findings identify household managerial capital as an important and an overlooked source of educational inequality.
Crisis-Induced Learning Losses and Recovery: Evidence from School Closures in Nigeria During COVID-19 (Joint with Adeniran, A., Okoye, D. and Wantchekon, L.) RISE Working Paper Series. 22/120.--(Under Review)
Millions of children worldwide are affected by school closures due to conflicts, disease outbreaks, and natural disasters, leading to learning losses and precarious future path for students in developing countries. There is an urgent need to provide possible policy options to get children back on track. This paper provides evidence from Nigerian schools during the COVID-19 pandemic, suggesting that a full recovery from crisis-induced learning losses is possible. Using data from a random sample of schools, we find significant learning losses of about .6 standard deviations in English and Math. However, a program designed to slow down the curriculum and cover what was missed during school closures led to a full recovery of all learning losses within 2 months. Having an educated mother and access to learning at-home helps to minimize learning losses and speed up recovery. Students who participated in the program do not lag behind one year later and remain in school.
Close to Home: Cognitive Distance and the Productivity of Learning (with Leonard Wantchekon)--Draft coming soon
Why does the same information generate different levels of understanding across individuals, even when exposure is held constant? A central challenge in learning and information transmission is that new material often varies in how interpretable it is to the receiver. This paper studies whether the effectiveness of instruction depends on what we call cognitive distance—the extent to which new information can be anchored in the learner’s existing experiences, language, and mental frameworks. When such anchors are available, information may be easier to process, more engaging, and more likely to be retained. We test this idea in a randomized field experiment with primary-school students in Benin. Schools were assigned to either continue with the standard curriculum or to adopt a history curriculum designed to reduce cognitive distance by grounding content in students’ local environments. In one treatment arm, the material is delivered in French; in a second arm, it is reinforced through discussion in local languages. The intervention leaves instructional time unchanged, allowing us to isolate the role of content interpretability. Reducing cognitive distance increases students’ engagement with reading and improves learning outcomes. Treated students spend more time reading outside the classroom and exhibit higher levels of reading fluency and comprehension. These gains extend beyond the subject in which the intervention is implemented, indicating that making content more interpretable can increase the overall productivity of learning. Additional evidence shows that effects are stronger for students whose linguistic environments are more closely aligned with the intervention. The results highlight a general mechanism: the returns to instruction depend not only on exposure, but also on how easily new information can be connected to what learners already know. While locally grounded content is one way to provide such anchors, the broader implication is that improving the interpretability of information may enhance learning across a wide range of contexts.
Other Ongoing Projects
Female Teachers and Girls-Only Classes (with Leonard Wantchekon): Data collection complete
Improving School Financing: The Role of Digitizing Tuition Fees Payment (with Leora Klapper and Owen Ozier)—Pilot Ongoing
Balancing the Scales: Gender Norms and Girls’ Educational Outcomes and Aspirations--Pilot Ongoing
A New Perspective on Child Marriage: Evidence from Benin