Filmer, Deon

Development Research Group
Profile Picture
Author Name Variants
Fields of Specialization
Education, Evidence-based public policy, Inequality and shared prosperity, Jobs and poverty, Social protection and labor, Social Protection and Growth
External Links
Development Research Group
Externally Hosted Work
Contact Information
Last updated August 31, 2023
Deon Filmer is a Lead Economist in the Research Group at the World Bank and Co-Director of the World Development Report 2018 Learning to Realize Education’s Promise. He has also previously served as Lead Economist in the Human Development department of the Africa Region of the World Bank. He works on issues of human capital and skills, service delivery, and the impact of policies and programs to improve human development outcomes—with research spanning the areas of education, health, social protection, and poverty and inequality. He has published widely in refereed journals, including studies of the impact of demand-side programs on schooling and learning; the roles of poverty, gender, orphanhood, and disability in explaining education inequalities; and the determinants of effective service delivery. He has recently co-authored the following books: Making Schools Work: New Evidence from Accountability Reforms, Youth Employment in Sub-Saharan Africa, and From Mines and Wells to Well-Built Minds: Turning Sub-Saharan Africa's Natural Resource Wealth into Human Capital. He was a core team member of the World Bank's World Development Reports in 1995 Workers in an Integrating World and 2004 Making Services Work for Poor People, and a contributor to 2007’s report Development and the Next Generation. He holds a PhD and MA from Brown University and a BA from Tufts University.
Citations 365 Scopus

Publication Search Results

Now showing 1 - 4 of 4
  • Thumbnail Image
    Teacher Performance-Based Incentives and Learning Inequality
    (World Bank, Washington, DC, 2020-09) Filmer, Deon ; Habyarimana, James ; Sabarwal, Shwetlena
    This study evaluates the impacts of low-cost, performance-based incentives in Tanzanian secondary schools. Results from a two-phase randomized trial show that incentives for teachers led to modest average improvements in student achievement across different subjects. Further, withdrawing incentives did not lead to a "discouragement effect" (once incentives were withdrawn, student performance did not fall below pre-baseline levels). Rather, impacts on learning were sustained beyond the intervention period. However, these incentives may have exacerbated learning inequality within and across schools. Increases in learning were concentrated among initially better-performing schools and students. At the same time, learning outcomes may have decreased for schools and students that were lower performing at baseline. Finally, the study finds that incentivizing students without simultaneously incentivizing teachers did not produce observable learning gains.
  • Thumbnail Image
    The Lost Human Capital: Teacher Knowledge and Student Achievement in Africa
    (World Bank, Washington, DC, 2019-05) Bold, Tessa ; Filmer, Deon ; Molina, Ezequiel ; Svensson, Jakob
    In many low-income countries, teachers do not master the subject they are teaching, and children learn little while attending school. Using unique data from nationally representative surveys of schools in seven Sub-Saharan African countries, this paper proposes a methodology to assess the effect of teacher subject content knowledge on student learning when panel data on students are not available. The paper shows that data on test scores of the student's current and the previous year's teachers, and knowledge of the correlation structure of teacher knowledge across time and grades, allow estimating two structural parameters of interest: the contemporaneous effect of teacher content knowledge, and the extent of fade out of teacher impacts in earlier grades. The paper uses these structural estimates to understand the magnitude of teacher effects and simulate the impacts of various policy reforms. Shortfalls in teachers' content knowledge account for 30 percent of the shortfall in learning relative to the curriculum, and about 20 percent of the cross-country difference in learning in the sample. Assigning more students to better teachers would potentially lead to substantial cost-savings, even if there are negative class-size effects. Ensuring that all incoming teachers have the officially mandated effective years of education, along with increasing the time spent on teaching to the officially mandated schedule, could almost double student learning within the next 30 years.
  • Thumbnail Image
    Identifying Effective Teachers: Lessons from Four Classroom Observation Tools
    (World Bank, Washington, DC, 2020-08) Filmer, Deon ; Molina, Ezequiel ; Wane, Waly
    Four different classroom observation instruments -- from the Service Delivery Indicators, the Stallings Observation System, the Classroom Assessment Scoring System, and the Teach classroom observation instrument -- were implemented in about 100 schools across four regions of Tanzania. The research design is such that various combinations of tools were administered to various combinations of teachers, so these data can be used to explore the commonalities and differences in the behaviors and practices captured by each tool, the internal properties of the tools (for example, how stable they are across enumerators, or how various indicators relate to one another), and how variables collected by the various tools compare to each other. Analysis shows that inter-rater reliability can be low, especially for some of the subjective ratings; principal components analysis suggests that lower-level constructs do not map neatly to predetermined higher-level ones and suggest that the data have only few dimensions. Measures collected during teacher observations are associated with student test scores, but patterns differ for teachers with lower versus higher subject content knowledge.
  • Thumbnail Image
    Preparation, Practice, and Beliefs: A Machine Learning Approach to Understanding Teacher Effectiveness
    (World Bank, Washington, DC, 2021-11) Filmer, Deon ; Nahata, Vatsal ; Sabarwal, Shwetlena
    This paper uses machine learning methods to identify key predictors of teacher effectiveness, proxied by student learning gains linked to a teacher over an academic year. Conditional inference forests and the least absolute shrinkage and selection operator are applied to matched student-teacher data for math and Kiswahili from grades 2 and 3 in 392 schools across Tanzania. These two machine learning methods produce consistent results and outperform standard ordinary least squares in out-of-sample prediction by 14–24 percent. As in previous research, commonly used teacher covariates like teacher gender, education, experience, and so forth are not good predictors of teacher effectiveness. Instead, teacher practice (what teachers do, measured through classroom observations and student surveys) and teacher beliefs (measured through teacher surveys) emerge as much more important. Overall, teacher covariates are stronger predictors of teacher effectiveness in math than in Kiswahili. Teacher beliefs that they can help disadvantaged and struggling students learn (for math) and they have good relationships within schools (for Kiswahili), teacher practice of providing written feedback and reviewing key concepts at the end of class (for math), and spending extra time with struggling students (for Kiswahili) are highly predictive of teacher effectiveness. As is teacher preparation on how to teach foundational topics (for both Math and Kiswahili). These results demonstrate the need to pay more systematic attention to teacher preparation, practice, and beliefs in teacher research and policy.