Preprint Annenberg EdWorkingPaper Cite
We examine how lower-performing public school teachers respond to more frequent formative evaluations. A 2011 reform in Chile required teachers rated “basic” (the second-lowest category) to be re-evaluated after two years rather than four. Using nationwide administrative data on primary school teachers and students from 2005–2015, we show that the reform increased two-year re-evaluation rates by 58.9 percentage points. In a descriptive pre-post comparison, teachers who completed re-evaluation improved their evaluation scores by about one standard deviation. Yet, difference-in-differences estimates rule out effects on student achievement above 0.04 standard deviations, and we find no meaningful changes in student-reported teaching practices or caregiver-reported teacher caring. These results suggest that increasing the frequency of formative evaluations alone may be insufficient to remediate lower-performing teachers. They also caution against interpreting large gains in evaluation scores as sufficient evidence of improvements in teacher effectiveness.
Preprint CESifo Working Paper Registration World Bank News Cite
Can governments translate evidence on “what works” into learning gains in public schools? Morocco combined targeted remediation and structured pedagogy (two approaches ranked as “great buys” by a global evidence panel) and delivered them through the existing public-school workforce in 626 schools. Using a pre-registered matched difference-in-differences design across 276 schools and 22,846 students, we estimate that the program increased scores on an independent assessment after one year by 0.90 standard deviations: 0.52 in Arabic, 1.30 in French, and 0.93 in mathematics. Administrative records also show higher grade promotion (1.7 percentage points), while enrollment, which is nearly universal, remained unchanged.
Preprint (Phase II - English) Preprint (Phase II - French) Registered report (Phase I) Cite
We investigate a government-delivered socioemotional support intervention in Morocco’s public lower-secondary schools. The intervention trained existing school social specialists to deliver small-group workshops to students in grades 7 and 8. In a cluster-randomized trial across 200 schools, all in the first year of a broader reform, 84 were assigned to the intervention and 116 to control. Assignment increased workshop participation by 43.2 percentage points and students’ knowledge of their social specialist by 16.9 percentage points. Over one school year, however, this increase in exposure did not translate into detectable average effects on the pre-specified primary outcomes (interpersonal and intrapersonal skills) or on overall learning, dropout, creativity, well-being, or study habits. If anything, students’ self-control declined by 0.097 standard deviations. Yet, we observe improvements in two sets of secondary outcomes: grade repetition fell by 1.7 percentage points, and student achievement increased in select subjects and subdomains: physics, chemistry, and biology, written expression in Arabic and French, and oral expression in Arabic. All of these results hold under a pre-specified hierarchy of hypothesis tests that controls the false discovery rate. Thus, although existing school staff can expand students’ exposure to support without additional staff or pay, a low-intensity universal protocol did not measurably improve the targeted socioemotional skills; instead, positive effects fell on academic and progression outcomes.
Preprint CEGA Working Paper Policy brief Press release Registration Cite
This study investigates whether a government-led, whole-school reform in public lower secondary schools can simultaneously reduce dropout and improve both academic and socioemotional outcomes. We employ a prospective difference-in-differences design, using machine learning on nationwide administrative data to match 200 reform schools in Morocco to 100 comparable schools. The analysis uses primary assessment data for 20,036 students and administrative enrollment records for 637,587 observations. After one year, the reform reduced end-of-year dropout by 1.6 percentage points (a 31.4 percent decline), increased learning by 0.52 standard deviations (a 3.3-fold acceleration), and improved a pre-specified index of socioemotional skills by 0.22 standard deviations. These findings demonstrate that government-led interventions can deliver multidimensional benefits during the critical lower secondary school years and highlight the potential of whole-school reform to support adolescent development.
Preprint CESifo Working Paper CEGA Working Paper Policy brief Press release Webinar recording Registration Cite
Whether targeted foundational instruction yields broad, long-term human capital gains is central to education policy but largely untested. We provide causal evidence from Zambia’s government-run foundational skills program in public primary schools. After two years, a randomized trial shows the program increases literacy by 0.10 and numeracy by 0.15 standard deviations. In mathematics, effects on targeted skills are 2.6 times larger than on comprehensive assessments, without detectable transfer to adjacent domains. Adding professional development doubles per-pupil costs without additional learning gains. Despite limited short-run transfer, event-study estimates show positive effects on grade-7 language and mathematics exam scores in early adolescence.
Preprint DOI Data Registration VoxDev Blog Cite
Home-visitation programs have improved child development in low- and middle-income countries, but they are costly to scale due to their reliance on trained workers. We evaluated an inexpensive and low-tech alternative with 2,433 caregivers of children aged 6 to 30 months served by 250 public childcare centers in Uttarakhand, India: automated phone calls offering parenting advice. The intervention was implemented largely as intended, with more than two-thirds of caregivers completing at least 10 calls. Yet, counter to expectations, it had negative but statistically insignificant effects on caregivers’ knowledge and interactions with their children, reduced their self-efficacy (by 0.11 standard deviations), and increased their anxiety (by 0.10 standard deviations). Consistent with this pattern, it had precisely estimated null effects on children’s development and language. An analysis of program materials suggests four reasons why the program may not have had the desired effects.
Preprint DOI Presentation Registration Cite
There is limited evidence on what drives teachers to change their teaching practices. Using primary qualitative data from 78 Zambian education personnel from the school to provincial level, we combine thematic analysis with unsupervised machine learning to identify drivers of pedagogical shifts. We then combine qualitative analyses with linear probability models to uncover their associations with teacher professional development. Our findings suggest that teaching practices are malleable, with change being predominantly driven by on-site continuous professional development (CPD) opportunities relating to team-based problem-solving, verbal discussions, and skills acquisition. Taken together, this study highlights the potential of school-based CPD opportunities as means to alter teaching practices, in a developing-country setting.
Preprint DOI Data Registration Cite
There is substantial emphasis on improving classroom practices, primarily through activity-based instruction, to increase the productivity of schooling. We study a large programme that seeks to promote mathematics learning in government primary schools in India. Through a cluster-randomised trial, we find that the programme increased activity-based instruction but yielded only muted impacts on learning. We provide a potential explanation: School value-added models suggest a negative relationship between activity-based instruction and test score gains. Our findings are robust to adding a community-engagement component to the intervention. These results highlight the limitations of activity-based instruction programmes for increasing school productivity.
Preprint DOI Presentation Cite
We assess the reliability and validity of cognitive and socioemotional skills measures and investigate the correlation between schooling, skills acquisition, and labor earnings. Our primary data from Pakistan incorporate two innovations related to measurement and sampling. With regard to measurement, we developed and implemented a battery of instruments intended to capture cognitive and socioemotional skills among young adults. With regard to sampling, we use a panel that follows respondents from their original rural locations in 2003 to their residences in 2018, a period over which 38% of respondents left their native villages. We first show that in terms of their validity and reliability, our skills measures compare favorably to previous measurement attempts in low- and middle-income countries. We then document that in our data (a) more years of schooling are correlated with higher cognitive and socioemotional skills, (b) labor earnings are correlated with cognitive and socioemotional skills, conditional on years of schooling, and (c) the earnings-skills correlations depend on respondents’ migration status. The magnitude of the correlations between schooling and skills on the one hand and earnings and skills on the other is consistent with a widespread concern that such skills are underproduced in the schooling system.
Preprint DOI Registration Harvard "Usable Knowledge" World Bank Blog Cite
This randomized trial investigates the long-term effects of a primary school scholarship program in rural Cambodia. We estimate impacts—9 years after program inception—on educational attainment, cognitive skills, socio-emotional outcomes, labor market outcomes, and well-being. Our results point to systematic improvements in educational attainment but no average impacts on long-term cognitive or socio-emotional outcomes. A merit-based (as opposed to poverty-based) targeting strategy did, however, increase cognitive outcomes, especially for poorer students. The results suggest positive effects on cognition for males. We find no improvements in labor market outcomes, yet positive effects on well-being, driven by recipients of merit-based scholarships. These findings shed light on the complex relationship between barriers to primary schooling and long-term outcomes, emphasizing the need for targeted approaches that consider both socioeconomic factors and individual merit, while also raising important questions about gender dynamics.
Preprint DOI One-page Summary Registration Brookings Blog SREE Blog Cite
This is one of the first studies to evaluate the impact of computer-based individualized instruction in a developing country. We randomly assigned 1,528 students in grades 6–8 in 15 “model” public schools in Rajasthan, India who were using a computer-adaptive learning software to: a control group, in which they were only able to access the activities for their enrolled grade level; or a treatment group, in which they were able to access exercises appropriate for their performance level. After 9 months, computer-based individualized instruction had a null average effect on math achievement. However, treatment students with low initial performance outperformed their control counterparts by 0.22 standard deviations. Our results suggest that computer-based individualized instruction is most beneficial for low-performing students.
We leverage data on learning for 101,084 public-school students in grades 4, 6, and 8 across 19 Indian states to diagnose their mathematic skills. These data allow us to diagnose their achievement on less frequently assessed skills. We use a novel approach to estimate the share of students who can meet fourth-grade standards. We find that the foundational skills of children are even lower than previously documented—52% mastered frequently assessed skills, but only 27% mastered typically unassessed skills. These children also make less progress than believed. Gender gaps in these skills emerge between grades 4 and 6 and persist.
Explaining the productivity paradox—the phenomenon where an introduction of information and communication technology (ICT) does not lead to improvements in labor productivity—is difficult, as changes in technology often coincide with adjustments to working hours and substitution of labor. I conduct a cluster-randomized trial in India to investigate the effects of a program that provides teachers with continuous training and materials, encouraging them to blend their instruction with high-quality videos. Teaching hours, teacher-to-student assignments, and the curriculum are held constant. Eleven months after its launch, I document negative effects on student learning in grades 9 and 10 in mathematics, and no effects in science. I also find detrimental effects on instructional quality, instructional practices, and student perceptions and attitudes towards mathematics and science. These findings suggest adjustment costs can serve as one explanation for the paradox.
Preprint DOI Registration Cite
This study is one of the first evaluations of independent (i.e., self-guided) practice in math in a developing country. We randomly assigned 4,461 students in grades 4-7 in “unaided” private schools across seven Indian cities who were using a computer-assisted learning software to: (a) a control group, in which they moved from one unit to the next upon completion; or (b) a treatment group, in which they had to complete practice exercises before progressing to the next topic. After six months, the additional practice had a precisely estimated null effect on the math achievement of the average student. However, treatment students with low initial performance outperformed their control counterparts by 0.14 standard deviations (SDs). Our results suggest that independent practice may help private-school students from relatively well-off families in need of catching up with their peers.
Preprint DOI The Economist Cite
We model learning losses due to the COVID-19 pandemic and the potential for cost-effective strategies to build back better. Data from Early Grade Reading Assessments in Ethiopia, Kenya, Liberia, Tanzania, and Uganda suggest half to over a year’s worth of learning loss. In modeling losses over time, we found that learning deficits for a child in grade 3 could lead to 2.8 years of lost learning by grade 10. While COVID-19 has stymied learning, bold, learning-focused reform consistent with the literature reviewed in this paper—specifically reform on targeted instruction and structured pedagogy—could improve learning even beyond pre-COVID-19 levels.
This study examines the relationships between school processes and literacy in public schools that serve low-income families in a midsize city in Colombia. We focus on four categories of school processes: instructional practice; school-community engagement; well-being of students, teachers, and parents; and community belonging. We present two key results. First, we find considerable variation in literacy performance across schools, despite serving students with similar family background characteristics. Second, we find community belonging to be the most salient school process for predicting literacy. Among measures of community belonging, higher levels of students’ sense of belonging and lower levels of bullying were associated with higher literacy. This study contributes to the field of international and comparative education by highlighting the importance of relational and interpersonal dimensions of school processes for children’s learning. Our study also demonstrates the value of adopting an interdisciplinary lens that combines economics, child development, and sociology to better understand the links between school processes and student learning.