How can impact evaluation improve student learning outcomes and enhance education quality?
Covering early childhood to secondary education, this publication compiles a decade's worth of education-focused impact evaluations supported by the World Bank. The selection of studies featured in this compendium includes the latest rigorous evidence on education issues affecting learning in low- and middle-income countries. Each study provides policymakers and practitioners with guidance to enhance evidence-based decision-making.
Topics covered include performance-based incentives for schools and teachers, approaches to parental training, and strategies for improving student retention in secondary education. Find studies related to early childhood, schools & teachers, students & parents, or browse country-specific studies under each tab above.
Eligibility criteria:
The studies featured in must meet the following criteria: they must be rigorous (experimental or quasi-experimental), focus on education from early childhood development to grades K-12, and include results related to student learning outcomes. Additionally, studies are available online (preferably as publications or working papers) and were published from 2015 - 2024. Studies that are follow-ups of an original experimental impact evaluation are not included and for interventions covered by multiple studies, only the most recent publication is included.
By impact evaluation, we mean rigorous evaluation methods designed to measure the causal effect of a program by comparing outcomes between a group that received an intervention (or treatment group) and a group that did not receive the intervention (or the comparison group).
Submit papers for future editions of the compendium:
As we look to the future, we see an opportunity to translate this research into evidence-based policy and program design. Your contributions to the compendium can help us expand this vital research, enabling us to conduct more rigorous impact evaluations in diverse contexts, explore new and innovative approaches to improving learning outcomes, investigate long-term impacts of educational interventions, and develop and test scalable solutions that can be implemented across multiple countries and regions.
If you have a paper that meets the eligibility criteria referenced above that you¡¯d like featured in future iterations of this compendium, write to us at eduimpacteval@worldbank.org.
This compendium was produced by the World Bank¡¯s Education team, in close collaboration with Development Research Group, Development Impact Group, and Strategic Impact Evaluation Fund.
Early Childhood
Interventions that improve access and/or quality of childcare centers (i.e., preschools, playgrounds, community-based centers)
There can be significant benefits to child development from group-based parenting support in the context of an informal preschool setting, but in this context, the early benefits faded over time. Policymakers designing similar programs should consider and experiment with strategies to improve the sustainability of impacts over time.
Brief and relatively affordable in-service teacher training, built into existing governmental systems, can improve key dimensions of classroom quality and early childhood development in both private and public preprimary institutions. Engaging parents on new educational approaches does not necessarily lead to positive effects on child development, especially if parents do not buy-into new educational programs. Parent meetings should be carefully designed and implemented to avoid negative effects that counteract school-based interventions.
It is hard to implement large-scale preschool programs that consistently result in improved child development outcomes. To maintain positive impacts, ongoing investments in preschool quality and subsequent educational stages (e.g., primary school) are necessary. Moreover, simply building preschools is insufficient; improvements in infrastructure need to be accompanied by enhancements in educational processes and teacher-child interactions. Specifically, the quality of education and teacher training are crucial for sustained benefits.
The impact of preschool programs can be highly context-specific and determined in large part by the sometimes unexpected, behavioral responses to an intervention. The design of preschool interventions should start with a good understanding of parental and teacher decision-making. A better understanding of the behavioral underpinnings, and the more general determinants of the counterfactual, could potentially inform better design of preschool programs.
Community-led preschools have the potential to be a cost-effective policy option for helping children meet their development potential even in the most resource deprived parts of the world. While this study¡¯s findings are encouraging, it is important for policymakers to consider that the program was delivered by an NGO, and participation was demand-driven ¨C both of which may have contributed to positive effects, and which may not be present if implementing community-led preschools at wide scale.
Pre-primary education can have positive effects on children¡¯s development. However, policymakers should consider children¡¯s maturity and timing of pre-primary education services when trying to ensure the benefits of pre-primary education programs are delivered consistently over time.
This preschool program illustrates the promise of pre-primary education and the power of supplyside components (e.g., short distances between family homes and school). Policymakers supporting the development of pre-primary education in low-income contexts should carefully consider school location, instructor support and training and the adequacy of infrastructure and materials for learning.
Interventions that financially support parents and/or train parents on parenting practices
Combining community-based behavioral change parenting programs with cash transfers can support changes in a range of parenting practices, and lead to some improvements in children¡¯s socio-emotional development. Programs that only offer cash transfers may improve household welfare but do not necessarily improve parenting knowledge and practice or early childhood development.
Combining community-based behavioral change parenting programs with cash transfers can support changes in a range of parenting practices, and lead to some improvements in children¡¯s socio-emotional development. Programs that only offer cash transfers may improve household welfare but do not necessarily improve parenting knowledge and practice or early childhood development.
With an ongoing focus on quality and fidelity, interventions that directly support parents to improve interactions with their children can support improved early childhood development. Such approaches could be expanded across different settings, offering Rwanda and other governments a vital tool to help break intergenerational cycles of poverty and violence.
Delivering the same ECD intervention through home visiting or group sessions had similar impacts on children¡¯s cognitive and language development. In this specific Indian context, group sessions cost only 28 percent of home visiting, which makes the former a more cost-effective and scalable method of delivery.
Distributing storybooks can encourage parents to read with their children and increase children¡¯s knowledge of book content. Follow-up training on dialogic reading is a promising way to improve the quality of parentchildren reading interactions and children¡¯s knowledge of storybook vocabulary.
Investing in early childhood interventions through scalable public health programs is cost-effective and can yield long-term benefits for children's cognitive and socioemotional development, as well as for parenting practices. Large-scale parenting programs can be integrated into national health systems and achieve substantial positive outcomes when adequately supported by skilled and motivated facilitators. With this said, design is crucial: tailoring programs to the needs of participants enhances their impact. The combination of structured sessions and interactive components is more effective than standard group sessions alone.
Combining community-based behavioral change parenting programs with cash transfers can support changes in a range of parenting practices, and lead to some improvements in children¡¯s socio-emotional development. Programs that only offer cash transfers may improve household welfare but do not necessarily improve parenting knowledge and practice or early childhood development.
Delivering the same ECD intervention through home visiting or group sessions had similar impacts on children¡¯s cognitive and language development. In this specific Indian context, group sessions cost only 28 percent of home visiting, which makes the former a more cost-effective and scalable method of delivery.
There can be significant benefits to child development from group-based parenting support in the context of an informal preschool setting, but in this context, the early benefits faded over time. Policymakers designing similar programs should consider and experiment with strategies to improve the sustainability of impacts over time.
Brief and relatively affordable in-service teacher training, built into existing governmental systems, can improve key dimensions of classroom quality and early childhood development in both private and public preprimary institutions. Engaging parents on new educational approaches does not necessarily lead to positive effects on child development, especially if parents do not buy-into new educational programs. Parent meetings should be carefully designed and implemented to avoid negative effects that counteract school-based interventions.
Community-led preschools have the potential to be a cost-effective policy option for helping children meet their development potential even in the most resource deprived parts of the world. While this study¡¯s findings are encouraging, it is important for policymakers to consider that the program was delivered by an NGO, and participation was demand-driven ¨C both of which may have contributed to positive effects, and which may not be present if implementing community-led preschools at wide scale.
This preschool program illustrates the promise of pre-primary education and the power of supplyside components (e.g., short distances between family homes and school). Policymakers supporting the development of pre-primary education in low-income contexts should carefully consider school location, instructor support and training and the adequacy of infrastructure and materials for learning
Distributing storybooks can encourage parents to read with their children and increase children¡¯s knowledge of book content. Follow-up training on dialogic reading is a promising way to improve the quality of parentchildren reading interactions and children¡¯s knowledge of storybook vocabulary.
The findings call for caution before implementing large-scale text message parenting interventions in high-poverty settings, as there are risks of unintended effects. Program designers should carefully consider: (i) how to work with local leaders to support program impact, for example by seeking local leaders¡¯ buy-in and input, and/or empowering them with specific roles during the intervention implementation, and (ii) ways to complement textmessage parenting interventions with other evidence-informed ECD approaches.
There can be significant benefits to child development from group-based parenting support in the context of an informal preschool setting, but in this context, the early benefits faded over time. Policymakers designing similar programs should consider and experiment with strategies to improve the sustainability of impacts over time.
Community-led preschools have the potential to be a cost-effective policy option for helping children meet their development potential even in the most resource deprived parts of the world. While this study¡¯s findings are encouraging, it is important for policymakers to consider that the program was delivered by an NGO, and participation was demand-driven ¨C both of which may have contributed to positive effects, and which may not be present if implementing community-led preschools at wide scale.
Schools and Teachers
Interventions that target schools, school leaders and/or teachers
Increasing school resources alone is not enough to impact school learning outcomes. Policymakers may also consider all the other binding constraints on education systemsthat need to be addressed alongside financial interventions. Relatedly, spending on certain categories, like information and communication technologies, might not yield the desired outcomes unless accompanied by complementary investments and training. Policymakers should target spending on areas with proven impact on learning outcomes.
When combined with diagnostic feedback, performance management tools and training can significantly enhance school management. These tools help in strengthening principals¡¯ capacity to develop and implement effective school improvement plans. However, simply providing assessment data to school principals does not seem to be enough to drive improvement and may even lead to negative outcomes. Schools need more than just data; they need the capacity to act on it effectively.
Improvements in school leadership can be achieved in a low-income context through governmentled training program at scale. Such training can help school leaders improve how schools use their existing resources (e.g., teachers¡¯ time and learning materials) to help the lowest-performing students, which in turn can help improve student learning. Policymakers aiming to influence teaching practices and school cultures, however, may need to combine leadership training with other interventions such as school-based in-service training for teachers.
With regards to training design, there may be potential to boost the impact of school managerial training programs by ensuring school leaders participate in the training¡¯s full duration. Moreover, direct training from professional trainers may be more effective than cascade train-the-trainer training model approaches. Policymakers need to weigh the costs against the potential impacts of direct managerial training approaches, as the cost of direct training for school managers can be high (in this study it was ~$470 USD per school), with low or no impact on student learning outcomes. Policymakers may opt for other more cost-effective education interventions to improve student learning.
Providing clear, practical tools for classroom observation and supporting differentiated instruction has the potential to significantly improve management quality and student outcomes. Providing managers with training on instructional practices is likely a key component to improve student learning.
When students, teachers, and parents are informed that their school test scores are low, it can trigger a process of self-evaluation and analysis; this process itself may lead to an improvement in learning outcomes ¨C especially if there are no shaming or punitive measures for poor performance. Knowledge of performance can be powerful when well disseminated, understood, and framed as an opportunity for improvement, but the study shows this process is not guaranteed just by making student learning outcomes public. Policymakers considering low-stakes accountability programs should consider ways to support teachers and school leaders to interpret and use diagnostic learning data as a catalyst for improvement.
School staff may be better equipped than central-level officials to identify solutions to school-specific problems using local knowledge. Granting autonomy with the development of school-specific interventions can motivate teachers, engage students, and improve student-teacher interaction. Moreover, combining teacher autonomy with technical assistance (e.g., mentors) can improve learning outcomes, particularly in low-capacity environments.
Increasing school resources alone is not enough to impact school learning outcomes. Policymakers may also consider all the other binding constraints on education systemsthat need to be addressed alongside financial interventions. Relatedly, spending on certain categories, like information and communication technologies, might not yield the desired outcomes unless accompanied by complementary investments and training. Policymakers should target spending on areas with proven impact on learning outcomes.
The study shows, where class sizes are extremely large (above 90 students per class), hiring additional teachers and constructing new classrooms can reduce pupil-teacher ratios, leading to reductions in student repetition rates and improvements in test scores. Moreover, where there are inter-district disparities in learning outcomes, targeted and coordinated packages of interconnected interventions can reduce learning gaps.
With regards to training design, there may be potential to boost the impact of school managerial training programs by ensuring school leaders participate in the training¡¯s full duration. Moreover, direct training from professional trainers may be more effective than cascade train-the-trainer training model approaches. Policymakers need to weigh the costs against the potential impacts of direct managerial training approaches, as the cost of direct training for school managers can be high (in this study it was ~$470 USD per school), with low or no impact on student learning outcomes. Policymakers may opt for other more cost-effective education interventions to improve student learning.
Pay-for-performance school grants can differentially affect learning improvements and inequalities in primary and junior secondary schools, depending on how schools allocate the funds. Policymakers may encourage schools to experiment with various funding allocation approaches to determine which strategies improve student performance and learning inequalities in the short- and long-term. The authors encourage policymakers to consider other metrics beyond learning outcomes, such as quality assurance measures or dimensions, to assess schools¡¯ performance and determine pay-for-performance grant allocation.
Although salary increase policies can improve teacher welfare, they do not necessarily produce corresponding improvement in student learning. Unconditional pay increases do not appear to be an effective policy option to improve teachers' effort and productivity in the short- to medium-term. Policymakers should consider unconditional salary increase policies with caution since such policies are difficult to reverse. It may be beneficial to combine salary increase policies with other interventions shown to improve student learning outcomes to improve both teacher welfare, effort, and student learning.
The study shows, where class sizes are extremely large (above 90 students per class), hiring additional teachers and constructing new classrooms can reduce pupil-teacher ratios, leading to reductions in student repetition rates and improvements in test scores. Moreover, where there are inter-district disparities in learning outcomes, targeted and coordinated packages of interconnected interventions can reduce learning gaps.
Providing schools with lights and videos does not appear to be enough to elicit significant achievement gains in student learning. Policymakers may wish to combine educational access initiatives (e.g., providing electricity and videos) with educational quality initiatives that support teachers and students to engage in higher quality learning activities and processes together.
School staff may be better equipped than central-level officials to identify solutions to school-specific problems using local knowledge. Granting autonomy with the development of school-specific interventions can motivate teachers, engage students, and improve student-teacher interaction. Moreover, combining teacher autonomy with technical assistance (e.g., mentors) can improve learning outcomes, particularly in low-capacity environments.
Programs that provide hands-on instructional support to under-trained instructors and training to parents can support student learning. Parents play a significant role in improving student learning, and by acknowledging the importance of engaging parents, policymakers and practitioners can enhance the design and implementation of educational interventions in underprivileged settings.
Providing clear, practical tools for classroom observation and supporting differentiated instruction has the potential to significantly improve management quality and student outcomes. Providing managers with training on instructional practices is likely a key component to improve student learning.
This study shows that the combination of teacher feedback and expert coaching is a promising strategy for raising school quality and student learning.
The study shows, where class sizes are extremely large (above 90 students per class), hiring additional teachers and constructing new classrooms can reduce pupil-teacher ratios, leading to reductions in student repetition rates and improvements in test scores. Moreover, where there are inter-district disparities in learning outcomes, targeted and coordinated packages of interconnected interventions can reduce learning gaps.
Low-cost incentive programs can effectively improve learning outcomes, but their success depends on careful design and implementation. Ensuring incentives are sustained over time and incorporating strategies to prevent the exacerbation of pre-existing inequalities are crucial. By addressing these factors, such programs have the potential to offer equitable benefits across diverse student groups.
Some teachers characteristics like motivation and gender can play a role in the effectiveness of incentive interventions. This study found that teachers respond to performance incentives and can do so in ways that improve student learning outcomes, even in a low-capacity environment. However, non-financial rewards¡ªwhich are often preferred by policymakers¡ªmight only be effective for some types of teachers. As such, policymakers may want to consider understanding the influence of such characteristics prior to implementation.
These results support the view that pay-for-performance (P4P) schemes can enhance teacher effort while also alleviating concerns about potential negative impacts on teacher selection.
Pay-for-performance school grants can differentially affect learning improvements and inequalities in primary and junior secondary schools, depending on how schools allocate the funds. Policymakers may encourage schools to experiment with various funding allocation approaches to determine which strategies improve student performance and learning inequalities in the short- and long-term. The authors encourage policymakers to consider other metrics beyond learning outcomes, such as quality assurance measures or dimensions, to assess schools¡¯ performance and determine pay-for-performance grant allocation.
When combined with diagnostic feedback, performance management tools and training can significantly enhance school management. These tools help in strengthening principals¡¯ capacity to develop and implement effective school improvement plans. However, simply providing assessment data to school principals does not seem to be enough to drive improvement and may even lead to negative outcomes. Schools need more than just data; they need the capacity to act on it effectively.
The impact of diagnostic feedback highlights the significant potential of large-scale assessments to enhance school management and classroom instruction. By providing detailed data on student performance, these assessments enable principals and teachers to make informed decisions, improve teaching strategies, and address specific learning gaps, ultimately leading to better educational outcomes.
This study shows that the combination of teacher feedback and expert coaching is a promising strategy for raising school quality and student learning.
When students, teachers, and parents are informed that their school test scores are low, it can trigger a process of self-evaluation and analysis; this process itself may lead to an improvement in learning outcomes ¨C especially if there are no shaming or punitive measures for poor performance. Knowledge of performance can be powerful when well disseminated, understood, and framed as an opportunity for improvement, but the study shows this process is not guaranteed just by making student learning outcomes public. Policymakers considering low-stakes accountability programs should consider ways to support teachers and school leaders to interpret and use diagnostic learning data as a catalyst for improvement.
The results of this study indicate that government support for local private providers may be a viable alternative to purely public provision: it seems possible for governments to set contracts with private, local entrepreneurs to provide high quality, low-cost educational solutions.
Schools delivering highly standardized education can support substantial learning gains at scale in lower-middle-income settings. Policymakers may consider adding more structure and standardization to their education systems. However, standardization efforts should be pursued with caution as the effects of standardization could vary between public and private schools, and according to local circumstances and implementation. Moreover, since this study was conducted, Bridge has closed most of its schools as controversies related to teacher pay and conditions, safety of the school grounds, and the effect of school fees on missed classes have come to light. Policymakers should consider the potential pitfalls arising from some design decisions in the pursuit of standardization and to evaluate efforts to introduce more structure into public education systems.
The study underscores the potential of philanthropically funded private schools to improve learning outcomes in Mexico and other developing countries. Given the high demand for private schooling and the presence of supportive networks, further research is needed to understand the long-term benefits, prevalence, and scalability of such schools to enhance educational and economic outcomes for the region's most vulnerable children
Entering into PPP contractual agreement can lead to positive effects in terms of enrollment, or student performance. However, these effects may come from a composition effect, whereby PPS attract more advantaged students. Policymakers should consider how to enable equitable access to PPP schools and review the impact of PPP schools with more diverse composition of students.
Designing interventions that take advantage of the underlying competitive structure of markets, such as financing that induces competitive actions, can be an important element of education policy. Although a private financier might prefer to finance a single school in each village, financing all private schools within a village may be preferable for society. The usual ¡°priority sector¡± lending policies could be augmented with a ¡°geographical targeting¡± subsidy that rewards the market for increasing financial saturation in an area: the density of coverage matters.
Providing schools with lights and videos does not appear to be enough to elicit significant achievement gains in student learning. Policymakers may wish to combine educational access initiatives (e.g., providing electricity and videos) with educational quality initiatives that support teachers and students to engage in higher quality learning activities and processes together.
Incorporating financial education into the school curriculum can improve financial literacy and some behavioral outcomes in the short-term.
The program shows financial education programs can support valuable outcomes like financial literacy and improved graduation rates, but the mixed effect on financial behaviors warrants caution especially since findings are only reported on the short-term. The authors suggest policymakers design and implement financial literacy programs with care, potentially working with researchers to investigate whether curriculum content and/or delivery (e.g., simple versus complex messaging) influences students¡¯ financial behavior in the short- and long-term.
Students and Parents
Interventions that target students
Providing accurate and timely information about graduation probabilities and the economic benefits of education has the potential to significantly improve graduation rates. Specifically, brief, well-designed information sessions can have a substantial impact on educational outcomes by fostering informed decision-making among students without significant costs. In contexts where providing information on the returns to education is not feasible due to data scarcity, it becomes imperative to seek alternative information sources that can effectively motivate students to pursue education. This paper contributes to this effort by presenting empirical evidence of a piece of information (PF) constructed using data available within the educational system that can be easily adopted in other settings. Lastly, tailoring interventions to students with lower academic standings has the potential to yield higher graduation rates.
Information campaigns are cost-effective and may be enough to help girls envision educational and workforce futures beyond existing norms and stereotypes in some contexts. However, information campaigns appear unlikely to reduce upper secondary dropout rates, may not be able to improve learning outcomes among students from disadvantaged backgrounds, and risk exacerbating existing inequalities. Policymakers will want to consider how to complement information campaigns with measures that reduce the effects of differences in initial academic skill or socioeconomic backgrounds.
Findings are consistent with prior research showing scholarships are effective when well targeted to financial needs and academic achievement. This study found that scholarship programs should target children and families with a clear need of financial support, and for whom financial constraints pose a barrier to pursuing education. Moreover, interventions aimed at improving learning and graduation rates in secondary school should be complemented and preceded by interventions aimed at improving foundational skills in primary school. In other words, it is important to address the quality of schooling at all levels as well as access to schooling at upper levels.
This study highlights the crucial role of implementation in determining the potential impact of an intervention, demonstrating that an intervention may not be adopted even if it appears beneficial. Accurately measuring program take-up and piloting interventions beforehand is essential to understand their implementation and adoption. Also, progress must be made to understand how to tackle secondary school dropouts to improve schooling and professional trajectories.
Providing accurate and timely information about graduation probabilities and the economic benefits of education has the potential to significantly improve graduation rates. Specifically, brief, well-designed information sessions can have a substantial impact on educational outcomes by fostering informed decision-making among students without significant costs. In contexts where providing information on the returns to education is not feasible due to data scarcity, it becomes imperative to seek alternative information sources that can effectively motivate students to pursue education. This paper contributes to this effort by presenting empirical evidence of a piece of information (PF) constructed using data available within the educational system that can be easily adopted in other settings. Lastly, tailoring interventions to students with lower academic standings has the potential to yield higher graduation rates.
Information campaigns are cost-effective and may be enough to help girls envision educational and workforce futures beyond existing norms and stereotypes in some contexts. However, information campaigns appear unlikely to reduce upper secondary dropout rates, may not be able to improve learning outcomes among students from disadvantaged backgrounds, and risk exacerbating existing inequalities. Policymakers will want to consider how to complement information campaigns with measures that reduce the effects of differences in initial academic skill or socioeconomic backgrounds.
Entertainment education and EdTech innovations delivered outside schools can be important complementary investments in stretched school systems, especially in rural, illiterate, and poor communities. Relative to other education investments evaluated in LMICs, the "Movies and Mobiles" intervention ranked in the top decile in terms of learning gains and the top quartile in terms of cost-effectiveness.
Digital solutions that deliver otherwise unavailable instructional ¡®content¡¯ may boost learning. However, simply having a new gadget with the same instructional content as available in textbooks is unlikely to improve learning outcomes. Policymakers looking to use educational technology to provide access to educational content should consider the relevance of and need for content provided via technology. Policymakers should also weigh the costs and impacts of educational technology, especially against strategies that may be more cost-effective and scalable (e.g., improving student secondary school retention through returns to education campaigns).
Entertainment education and EdTech innovations delivered outside schools can be important complementary investments in stretched school systems, especially in rural, illiterate, and poor communities. Relative to other education investments evaluated in LMICs, the "Movies and Mobiles" intervention ranked in the top decile in terms of learning gains and the top quartile in terms of cost-effectiveness.
Programs that provide hands-on instructional support to under-trained instructors and training to parents can support student learning. Parents play a significant role in improving student learning, and by acknowledging the importance of engaging parents, policymakers and practitioners can enhance the design and implementation of educational interventions in underprivileged settings.
This study highlights the crucial role of implementation in determining the potential impact of an intervention, demonstrating that an intervention may not be adopted even if it appears beneficial. Accurately measuring program take-up and piloting interventions beforehand is essential to understand their implementation and adoption. Also, progress must be made to understand how to tackle secondary school dropouts to improve schooling and professional trajectories.
Findings are consistent with prior research showing scholarships are effective when well targeted to financial needs and academic achievement. This study found that scholarship programs should target children and families with a clear need of financial support, and for whom financial constraints pose a barrier to pursuing education. Moreover, interventions aimed at improving learning and graduation rates in secondary school should be complemented and preceded by interventions aimed at improving foundational skills in primary school. In other words, it is important to address the quality of schooling at all levels as well as access to schooling at upper levels.
Providing high-frequency information to parents about their children¡¯s school attendance in lowincome settings can improve children¡¯s attendance rates, even in the absence of CCTs, where asymmetries in information exist. Initiatives to improve school attendance may be enhanced by working directly with and supporting children¡¯s own agency and educational decisions. Further research could explore whether supporting children¡¯s educational agency has long-term effects on their education and life trajectories.
Low-cost incentive programs can effectively improve learning outcomes, but their success depends on careful design and implementation. Ensuring incentives are sustained over time and incorporating strategies to prevent the exacerbation of pre-existing inequalities are crucial. By addressing these factors, such programs have the potential to offer equitable benefits across diverse student groups.
While goal setting can positively affect effort, time management, and self-discipline, these behavioral improvements are not always insufficient to significantly enhance academic performance. Supporting students to develop realistic goals is important and policymakers should carefully consider the use of extrinsic incentives alongside goal setting as these may sometimes weaken overall impact of goal-setting interventions.
While goal setting can positively affect effort, time management, and self-discipline, these behavioral improvements are not always insufficient to significantly enhance academic performance. Supporting students to develop realistic goals is important and policymakers should carefully consider the use of extrinsic incentives alongside goal setting as these may sometimes weaken overall impact of goal-setting interventions.
This study highlights the crucial role of implementation in determining the potential impact of an intervention, demonstrating that an intervention may not be adopted even if it appears beneficial. Accurately measuring program take-up and piloting interventions beforehand is essential to understand their implementation and adoption. Also, progress must be made to understand how to tackle secondary school dropouts to improve schooling and professional trajectories.
Providing schools with lights and videos does not appear to be enough to elicit significant achievement gains in student learning. Policymakers may wish to combine educational access initiatives (e.g., providing electricity and videos) with educational quality initiatives that support teachers and students to engage in higher quality learning activities and processes together.
Digital solutions that deliver otherwise unavailable instructional ¡®content¡¯ may boost learning. However, simply having a new gadget with the same instructional content as available in textbooks is unlikely to improve learning outcomes. Policymakers looking to use educational technology to provide access to educational content should consider the relevance of and need for content provided via technology. Policymakers should also weigh the costs and impacts of educational technology, especially against strategies that may be more cost-effective and scalable (e.g., improving student secondary school retention through returns to education campaigns).
Interventions that target parents, financially support parents and/or train parents on parenting practices
Combining community-based behavioral change parenting programs with cash transfers can support changes in a range of parenting practices, and lead to some improvements in children¡¯s socio-emotional development. Programs that only offer cash transfers may improve household welfare but do not necessarily improve parenting knowledge and practice or early childhood development.
Providing high-frequency information to parents about their children¡¯s school attendance in lowincome settings can improve children¡¯s attendance rates, even in the absence of CCTs, where asymmetries in information exist. Initiatives to improve school attendance may be enhanced by working directly with and supporting children¡¯s own agency and educational decisions. Further research could explore whether supporting children¡¯s educational agency has long-term effects on their education and life trajectories.
Measures aimed at creating community-wide, inclusive platforms to promote dialogue can lead to greater interest of the community in education with impacts on access and staffing. However, the attempt to reform existing SMC institutions had limited effect, mostly leading to undermining the effects of dialogue interventions.
Providing high-frequency information to parents about their children¡¯s school attendance in lowincome settings can improve children¡¯s attendance rates, even in the absence of CCTs, where asymmetries in information exist. Initiatives to improve school attendance may be enhanced by working directly with and supporting children¡¯s own agency and educational decisions. Further research could explore whether supporting children¡¯s educational agency has long-term effects on their education and life trajectories.
Programs that provide hands-on instructional support to under-trained instructors and training to parents can support student learning. Parents play a significant role in improving student learning, and by acknowledging the importance of engaging parents, policymakers and practitioners can enhance the design and implementation of educational interventions in underprivileged settings.
Entertainment education and EdTech innovations delivered outside schools can be important complementary investments in stretched school systems, especially in rural, illiterate, and poor communities. Relative to other education investments evaluated in LMICs, the "Movies and Mobiles" intervention ranked in the top decile in terms of learning gains and the top quartile in terms of cost-effectiveness.
Interactive voice response (IVR) calls are a promising, low-cost method of educational outreach that may increase the extent to which parents read with their young children. However, the call design matters significantly for uptake.
Combining community-based behavioral change parenting programs with cash transfers can support changes in a range of parenting practices, and lead to some improvements in children¡¯s socio-emotional development. Programs that only offer cash transfers may improve household welfare but do not necessarily improve parenting knowledge and practice or early childhood development.
With an ongoing focus on quality and fidelity, interventions that directly support parents to improve interactions with their children can support improved early childhood development. Such approaches could be expanded across different settings, offering Rwanda and other governments a vital tool to help break intergenerational cycles of poverty and violence.
Delivering the same ECD intervention through home visiting or group sessions had similar impacts on children¡¯s cognitive and language development. In this specific Indian context, group sessions cost only 28 percent of home visiting, which makes the former a more cost-effective and scalable method of delivery.
Distributing storybooks can encourage parents to read with their children and increase children¡¯s knowledge of book content. Follow-up training on dialogic reading is a promising way to improve the quality of parentchildren reading interactions and children¡¯s knowledge of storybook vocabulary.
Investing in early childhood interventions through scalable public health programs is cost-effective and can yield long-term benefits for children's cognitive and socioemotional development, as well as for parenting practices. Large-scale parenting programs can be integrated into national health systems and achieve substantial positive outcomes when adequately supported by skilled and motivated facilitators. With this said, design is crucial: tailoring programs to the needs of participants enhances their impact. The combination of structured sessions and interactive components is more effective than standard group sessions alone.
Combining community-based behavioral change parenting programs with cash transfers can support changes in a range of parenting practices, and lead to some improvements in children¡¯s socio-emotional development. Programs that only offer cash transfers may improve household welfare but do not necessarily improve parenting knowledge and practice or early childhood development.
Delivering the same ECD intervention through home visiting or group sessions had similar impacts on children¡¯s cognitive and language development. In this specific Indian context, group sessions cost only 28 percent of home visiting, which makes the former a more cost-effective and scalable method of delivery.
There can be significant benefits to child development from group-based parenting support in the context of an informal preschool setting, but in this context, the early benefits faded over time. Policymakers designing similar programs should consider and experiment with strategies to improve the sustainability of impacts over time.
Brief and relatively affordable in-service teacher training, built into existing governmental systems, can improve key dimensions of classroom quality and early childhood development in both private and public preprimary institutions. Engaging parents on new educational approaches does not necessarily lead to positive effects on child development, especially if parents do not buy-into new educational programs. Parent meetings should be carefully designed and implemented to avoid negative effects that counteract school-based interventions.
Community-led preschools have the potential to be a cost-effective policy option for helping children meet their development potential even in the most resource deprived parts of the world. While this study¡¯s findings are encouraging, it is important for policymakers to consider that the program was delivered by an NGO, and participation was demand-driven ¨C both of which may have contributed to positive effects, and which may not be present if implementing community-led preschools at wide scale.
This preschool program illustrates the promise of pre-primary education and the power of supplyside components (e.g., short distances between family homes and school). Policymakers supporting the development of pre-primary education in low-income contexts should carefully consider school location, instructor support and training and the adequacy of infrastructure and materials for learning
Distributing storybooks can encourage parents to read with their children and increase children¡¯s knowledge of book content. Follow-up training on dialogic reading is a promising way to improve the quality of parentchildren reading interactions and children¡¯s knowledge of storybook vocabulary.
The findings call for caution before implementing large-scale text message parenting interventions in high-poverty settings, as there are risks of unintended effects. Program designers should carefully consider: (i) how to work with local leaders to support program impact, for example by seeking local leaders¡¯ buy-in and input, and/or empowering them with specific roles during the intervention implementation, and (ii) ways to complement textmessage parenting interventions with other evidence-informed ECD approaches.
There can be significant benefits to child development from group-based parenting support in the context of an informal preschool setting, but in this context, the early benefits faded over time. Policymakers designing similar programs should consider and experiment with strategies to improve the sustainability of impacts over time.
Community-led preschools have the potential to be a cost-effective policy option for helping children meet their development potential even in the most resource deprived parts of the world. While this study¡¯s findings are encouraging, it is important for policymakers to consider that the program was delivered by an NGO, and participation was demand-driven ¨C both of which may have contributed to positive effects, and which may not be present if implementing community-led preschools at wide scale.
Ghana
Brief and relatively affordable in-service teacher training, built into existing governmental systems, can improve key dimensions of classroom quality and early childhood development in both private and public preprimary institutions. Engaging parents on new educational approaches does not necessarily lead to positive effects on child development, especially if parents do not buy-into new educational programs. Parent meetings should be carefully designed and implemented to avoid negative effects that counteract school-based interventions.
Kenya
Distributing storybooks can encourage parents to read with their children and increase children¡¯s knowledge of book content. Follow-up training on dialogic reading is a promising way to improve the quality of parentchildren reading interactions and children¡¯s knowledge of storybook vocabulary.
Schools delivering highly standardized education can support substantial learning gains at scale in lower-middle-income settings. Policymakers may consider adding more structure and standardization to their education systems. However, standardization efforts should be pursued with caution as the effects of standardization could vary between public and private schools, and according to local circumstances and implementation. Moreover, since this study was conducted, Bridge has closed most of its schools as controversies related to teacher pay and conditions, safety of the school grounds, and the effect of school fees on missed classes have come to light. Policymakers should consider the potential pitfalls arising from some design decisions in the pursuit of standardization and to evaluate efforts to introduce more structure into public education systems.
Malawi
Improvements in school leadership can be achieved in a low-income context through governmentled training program at scale. Such training can help school leaders improve how schools use their existing resources (e.g., teachers¡¯ time and learning materials) to help the lowest-performing students, which in turn can help improve student learning. Policymakers aiming to influence teaching practices and school cultures, however, may need to combine leadership training with other interventions such as school-based in-service training for teachers.
The study shows, where class sizes are extremely large (above 90 students per class), hiring additional teachers and constructing new classrooms can reduce pupil-teacher ratios, leading to reductions in student repetition rates and improvements in test scores. Moreover, where there are inter-district disparities in learning outcomes, targeted and coordinated packages of interconnected interventions can reduce learning gaps.
Mozambique
Community-led preschools have the potential to be a cost-effective policy option for helping children meet their development potential even in the most resource deprived parts of the world. While this study¡¯s findings are encouraging, it is important for policymakers to consider that the program was delivered by an NGO, and participation was demand-driven ¨C both of which may have contributed to positive effects, and which may not be present if implementing community-led preschools at wide scale.
Nigeria
Entertainment education and EdTech innovations delivered outside schools can be important complementary investments in stretched school systems, especially in rural, illiterate, and poor communities. Relative to other education investments evaluated in LMICs, the "Movies and Mobiles" intervention ranked in the top decile in terms of learning gains and the top quartile in terms of cost-effectiveness.
Digital solutions that deliver otherwise unavailable instructional ¡®content¡¯ may boost learning. However, simply having a new gadget with the same instructional content as available in textbooks is unlikely to improve learning outcomes. Policymakers looking to use educational technology to provide access to educational content should consider the relevance of and need for content provided via technology. Policymakers should also weigh the costs and impacts of educational technology, especially against strategies that may be more cost-effective and scalable (e.g., improving student secondary school retention through returns to education campaigns).
Rwanda
With an ongoing focus on quality and fidelity, interventions that directly support parents to improve interactions with their children can support improved early childhood development. Such approaches could be expanded across different settings, offering Rwanda and other governments a vital tool to help break intergenerational cycles of poverty and violence.
These results support the view that pay-for-performance (P4P) schemes can enhance teacher effort while also alleviating concerns about potential negative impacts on teacher selection.
Tanzania
Providing schools with lights and videos does not appear to be enough to elicit significant achievement gains in student learning. Policymakers may wish to combine educational access initiatives (e.g., providing electricity and videos) with educational quality initiatives that support teachers and students to engage in higher quality learning activities and processes together.
Uganda
Entering into PPP contractual agreement can lead to positive effects in terms of enrollment, or student performance. However, these effects may come from a composition effect, whereby PPS attract more advantaged students. Policymakers should consider how to enable equitable access to PPP schools and review the impact of PPP schools with more diverse composition of students.
Indonesia
Pre-primary education can have positive effects on children¡¯s development. However, policymakers should consider children¡¯s maturity and timing of pre-primary education services when trying to ensure the benefits of pre-primary education programs are delivered consistently over time.
Pay-for-performance school grants can differentially affect learning improvements and inequalities in primary and junior secondary schools, depending on how schools allocate the funds. Policymakers may encourage schools to experiment with various funding allocation approaches to determine which strategies improve student performance and learning inequalities in the short- and long-term. The authors encourage policymakers to consider other metrics beyond learning outcomes, such as quality assurance measures or dimensions, to assess schools¡¯ performance and determine pay-for-performance grant allocation.
Argentina
When combined with diagnostic feedback, performance management tools and training can significantly enhance school management. These tools help in strengthening principals¡¯ capacity to develop and implement effective school improvement plans. However, simply providing assessment data to school principals does not seem to be enough to drive improvement and may even lead to negative outcomes. Schools need more than just data; they need the capacity to act on it effectively.
The impact of diagnostic feedback highlights the significant potential of large-scale assessments to enhance school management and classroom instruction. By providing detailed data on student performance, these assessments enable principals and teachers to make informed decisions, improve teaching strategies, and address specific learning gaps, ultimately leading to better educational outcomes.
Brazil
This study shows that the combination of teacher feedback and expert coaching is a promising strategy for raising school quality and student learning.
School staff may be better equipped than central-level officials to identify solutions to school-specific problems using local knowledge. Granting autonomy with the development of school-specific interventions can motivate teachers, engage students, and improve student-teacher interaction. Moreover, combining teacher autonomy with technical assistance (e.g., mentors) can improve learning outcomes, particularly in low-capacity environments.
The program shows financial education programs can support valuable outcomes like financial literacy and improved graduation rates, but the mixed effect on financial behaviors warrants caution especially since findings are only reported on the short-term. The authors suggest policymakers design and implement financial literacy programs with care, potentially working with researchers to investigate whether curriculum content and/or delivery (e.g., simple versus complex messaging) influences students¡¯ financial behavior in the short- and long-term.
Mexico
Information campaigns are cost-effective and may be enough to help girls envision educational and workforce futures beyond existing norms and stereotypes in some contexts. However, information campaigns appear unlikely to reduce upper secondary dropout rates, may not be able to improve learning outcomes among students from disadvantaged backgrounds, and risk exacerbating existing inequalities. Policymakers will want to consider how to complement information campaigns with measures that reduce the effects of differences in initial academic skill or socioeconomic backgrounds.
Findings are consistent with prior research showing scholarships are effective when well targeted to financial needs and academic achievement. This study found that scholarship programs should target children and families with a clear need of financial support, and for whom financial constraints pose a barrier to pursuing education. Moreover, interventions aimed at improving learning and graduation rates in secondary school should be complemented and preceded by interventions aimed at improving foundational skills in primary school. In other words, it is important to address the quality of schooling at all levels as well as access to schooling at upper levels.
With regards to training design, there may be potential to boost the impact of school managerial training programs by ensuring school leaders participate in the training¡¯s full duration. Moreover, direct training from professional trainers may be more effective than cascade train-the-trainer training model approaches. Policymakers need to weigh the costs against the potential impacts of direct managerial training approaches, as the cost of direct training for school managers can be high (in this study it was ~$470 USD per school), with low or no impact on student learning outcomes. Policymakers may opt for other more cost-effective education interventions to improve student learning.
When students, teachers, and parents are informed that their school test scores are low, it can trigger a process of self-evaluation and analysis; this process itself may lead to an improvement in learning outcomes ¨C especially if there are no shaming or punitive measures for poor performance. Knowledge of performance can be powerful when well disseminated, understood, and framed as an opportunity for improvement, but the study shows this process is not guaranteed just by making student learning outcomes public. Policymakers considering low-stakes accountability programs should consider ways to support teachers and school leaders to interpret and use diagnostic learning data as a catalyst for improvement.
Increasing school resources alone is not enough to impact school learning outcomes. Policymakers may also consider all the other binding constraints on education systemsthat need to be addressed alongside financial interventions. Relatedly, spending on certain categories, like information and communication technologies, might not yield the desired outcomes unless accompanied by complementary investments and training. Policymakers should target spending on areas with proven impact on learning outcomes.
Increasing school resources alone is not enough to impact school learning outcomes. Policymakers may also consider all the other binding constraints on education systemsthat need to be addressed alongside financial interventions. Relatedly, spending on certain categories, like information and communication technologies, might not yield the desired outcomes unless accompanied by complementary investments and training. Policymakers should target spending on areas with proven impact on learning outcomes.
The study underscores the potential of philanthropically funded private schools to improve learning outcomes in Mexico and other developing countries. Given the high demand for private schooling and the presence of supportive networks, further research is needed to understand the long-term benefits, prevalence, and scalability of such schools to enhance educational and economic outcomes for the region's most vulnerable children.
India
Delivering the same ECD intervention through home visiting or group sessions had similar impacts on children¡¯s cognitive and language development. In this specific Indian context, group sessions cost only 28 percent of home visiting, which makes the former a more cost-effective and scalable method of delivery.
Pakistan
The results of this study indicate that government support for local private providers may be a viable alternative to purely public provision: it seems possible for governments to set contracts with private, local entrepreneurs to provide high quality, low-cost educational solutions.