Publication: How to Improve Education Outcomes Most Efficiently? A Comparison of 150 Interventions Using the New Learning-Adjusted Years of Schooling Metric
Loading...
Date
2020-10
ISSN
Published
2020-10
Author(s)
Abstract
Many low- and middle-income countries lag far behind high-income countries in educational access and student learning. Limited resources mean that policymakers must make tough choices about which investments to make to improve education. Although hundreds of education interventions have been rigorously evaluated, making comparisons between the results is challenging. Some studies report changes in years of schooling; others report changes in learning. Standard deviations, the metric typically used to report learning gains, measure gains relative to a local distribution of test scores. This metric makes it hard to judge if the gain is worth the cost in absolute terms. This paper proposes using learning-adjusted years of schooling (LAYS) -- which combines access and quality and compares gains to an absolute, cross-country standard -- as a new metric for reporting gains from education interventions. The paper applies LAYS to compare the effectiveness (and cost-effectiveness, where cost is available) of interventions from 150 impact evaluations across 46 countries. The results show that some of the most cost-effective programs deliver the equivalent of three additional years of high-quality schooling (that is, schooling at quality comparable to the highest-performing education systems) for just $100 per child -- compared with zero years for other classes of interventions.
Link to Data Set
Citation
“Angrist, Noam; Evans, David K.; Filmer, Deon; Glennerster, Rachel; Rogers, F. Halsey; Sabarwal, Shwetlena. 2020. How to Improve Education Outcomes Most Efficiently? A Comparison of 150 Interventions Using the New Learning-Adjusted Years of Schooling Metric. Policy Research Working Paper;No. 9450. © World Bank, Washington, DC. http://hdl.handle.net/10986/34658 License: CC BY 3.0 IGO.”
Other publications in this report series
Publication Buffer or Bottleneck? Employment Exposure to Generative AI and the Digital Divide in Latin America(Washington, DC: World Bank, 2024-08-01)Empirical evidence on the potential impacts of generative artificial intelligence (GenAI) is mostly focused on high-income countries. In contrast, little is known about the role of this technology on the future economic pathways of developing economies. This paper contributes to fill this gap by estimating the exposure of the Latin American labor market to GenAI. It provides detailed statistics of GenAI exposure between and within countries by leveraging a rich set of harmonized household and labor force surveys. To account for the slower pace of technology adoption in developing economies, it adjusts the measures of exposure to GenAI by using the likelihood of accessing digital technologies at work. This is then used to assess the extent to which the digital divide across and within countries will be a barrier to maximize the productivity gains among occupations that could otherwise be augmented by GenAI tools. The findings show that certain characteristics are consistently correlated with higher exposure. Specifically, urban-based jobs that require higher education, are situated in the formal sector, and are held by individuals with higher incomes are more likely to come into interaction with this technology. Moreover, there is a pronounced tilt toward younger workers facing greater exposure, including the risk of job automation, particularly in the finance, insurance, and public administration sectors. When adjusting for access to digital technologies, the findings show that the digital divide is a major barrier to realizing the positive effects of GenAI on jobs in the region. In particular, nearly half of the positions that could potentially benefit from augmentation are hampered by lack of use of digital technologies. This negative effect of the digital divide is more pronounced in poorer countries.Publication Making Agriculture Work for the Poor in Timor-Leste(Washington, DC: World Bank, 2024-07-29)With about 80 percent of poor households in Timor-Leste dependent on agriculture for their livelihood, it is widely acknowledged that growth in the agriculture sector is an important channel for poverty reduction in the country. That notwithstanding, the country’s agricultural production system is one of the least developed in the world, with the sector’s productivity being well below that of other small island developing states and below the average for other low-income states. Using data from the 2007 and 2014 waves of the Timor-Leste Surveys of Living Standards, this paper provides insight into the extent to which increases in agricultural productivity can contribute to poverty reduction, and assesses the drivers of and constraints to increasing agricultural productivity in the country. The findings show that improvements in agricultural productivity reduce the probability of being poor among agricultural households. The paper also finds that for agricultural productivity to increase, the following factors should be considered: facilitation of farm mechanization, improvement in the use of chemical (fertilizer, pesticides, and herbicides) inputs, enhancement of access to credit and extension, encouragement to farmers to join farmer groups, expansion of the commercialization of farm produce, and reduction in the gender gap in agricultural productivity.Publication Electricity Transition in MFMod(Washington, DC: World Bank, 2024-07-25)This paper describes power sector modeling methodologies for the World Bank’s macrostructural model — MFMod. Macrostructural models generally do not model sectors, such as the power sector, in detail, limiting their capacity to represent deep system transformation (for example, low-carbon energy transitions). The main constraints to adequate sector modeling are data availability and technological representation of the power system. Time-series data for specific production factors across sectors do not exist consistently for most countries in World Bank models. This paper describes two distinct methods to overcome this constraint: (i) using a more granular representation of the production function and (ii) linking the macrostructural model with the World Bank's electricity planning models. These methods provide a more nuanced technical representation of deep transformations, enabling discussions on their macroeconomic consequences. The paper provides results for Mauritania and South Africa. These methodologies serve as a blueprint for macroeconomic modeling of energy transitions in this class of models.Publication Should Farm Input Subsidy Programs Target Poor or Non-Poor Farmers? Evidence from Malawi’s Farm Input Subsidy Program(Washington, DC: World Bank, 2024-07-25)This paper addresses the question of whether farm input subsidy programs should be targeted at non-poor farmers instead of poor farmers, using a two-wave, nationally representative panel data from Malawi. The question is addressed by estimating the net gain in maize yield for targeting non-poor farmers instead of poor farmers after accounting for the difference in inorganic fertilizer use efficiency and the difference in crowding-out of commercial fertilizer by subsidized fertilizer between the poor and non-poor farmers. Consumption expenditure is used to classify households into consumption poor and non-poor households, and an asset-based wealth index is used to classify households into asset poor and non-poor households. The difference in inorganic fertilizer use efficiency is estimated with a multilevel model of maize yield, and the difference in crowding out is estimated with a double hurdle model of demand for commercial, inorganic fertilizer. The results indicate that non-poor farmers are significantly more efficient in the use of inorganic fertilizer but have significantly higher levels of crowding out, compared to poor farmers. This suggests that there is a trade-off between targeting non-poor farmers and targeting poor farmers. However, further analysis of the trade-off indicates that targeting non-poor farmers instead of poor farmers, even after accounting for the difference in crowding out, would result in an overall gain in yield of 3.14 to 4.33 kilograms of maize per kilogram of nitrogen distributed by the subsidy program. Therefore, the productivity enhancing objective of Malawi’s farm input subsidy program would be better served by targeting non-poor farmers instead of poor farmers.Publication Do Psychosocial Stimulation, Parental Distress and Early Childhood Education Enrollment Show Different Associations with Early Childhood Development Outcomes for Boys and Girls? Findings from a Phone Survey in Pakistan(Washington, DC: World Bank, 2024-07-25)This study examined whether psychosocial stimulation, parental distress, and enrollment in pre- primary education had different associations with early child development outcomes for boys and girls in Pakistan. Using data from a nationally representative phone survey in Pakistan, it assessed these relationships for two cohorts of children—those aged 0–35 months and those aged 36–72 months. The study found that among very young children (0–35 months), lower parental distress and higher psychosocial stimulation were strongly associated with better child development for both boys and girls. Girls were more sensitive to higher levels of parental distress and lower levels of psychosocial stimulation than boys. On average, girls in the sample fared worse compared to boys in their developmental outcomes in the context of low levels of stimulation and high levels of parental distress. Among the older age cohort (36–72 months), lower psychosocial stimulation and higher parental distress were each similarly associated with lower child development outcomes, regardless of child gender. Access to early childhood education was associated with better child development outcomes for both genders. The results confirm existing evidence that early learning opportunities in the first six years of life are important supports for promoting early child development for all children and suggest that girls aged 0–35 months in this sample may be uniquely sensitive to psychosocial stimulation and parental distress.