The Improving Learning Outcomes for Asia (ILOA) regional project works closely with USAID’s Asia Bureau and Missions across the region to identify key questions and challenges Mission staff face in their day-to-day work. An advisory group of USAID Mission colleagues raised an important question: how best to wade through the array of education data available? What are the types, how are they used, when are different data useful, etc.? Indeed, the use of different data to effectively partner with governments to make evidence-based decisions is a top priority. As a result, ILOA produced a brief summarizing the different sources and uses of data for basic education, youth and workforce development, and higher education. The brief recognizes how data sources and uses have evolved over time, enabling ministries, their partners, and stakeholders to measure performance, inform policy and plan interventions, and manage limited resources. The brief is designed to succinctly assist USAID staff and their partners in navigating the world of education data.

On mainland Tanzania, most resource allocation decisions are centralized. The President’s Office - Regional Administration and Local Government (PO-RALG) recruits and assigns teachers, supplies teaching and learning materials and funds capital construction projects. Local Governments are provided limited funds for training support or redeployment of teachers among schools. Their main resource, therefore, is to provide management attention and support to schools. With an average of 140 schools in a District and a staff of 5 individuals, only a few schools can be supported. In 2016, The Ministry of Education and Sports developed a School Quality Assurance Framework to guide local administrators on key areas of focus and guidance for school support. The framework focuses on six areas: school inputs, teacher practice, student learning outcomes, school environment, school leadership and community engagement. To facilitate the monitoring of these areas, USAID Tusome Pamoja project piloted a data collection tool that allowed measurement of progress through indicators. Of particular interest was the use of a group administered learning assessment that established benchmarks for success for grade 2 learners across six sub-tasks for reading, writing and mathematics. Due to limited resources, this assessment was only applied to a sample of schools in each district. Districts could assess their overall performance against these indicators and as a result developed somewhat generic district level support plans.. This presentation will explore how initial challenges of vague district plans were overcome through the critical data collection process leading to the establishment of benchmarks for success. o Under a subsequent activity, USAID Jifunze Uelewe, software was developed that allowed districts to capture group administered learning data for every grade 2 student and to aggregate this information at the school level. Districts were then able to rank order all schools in the district by scores on learning sub-tasks and then select the lowest performing schools for additional management attention. At the same time, districts were able to pair high-performing and low performing schools. The result of the access to school specific data was to allow districts to direct their attention to the development of plans at the school level to address low learning performance, and the ability to track progress of these schools over time. Schools enter data on Government provided tablets and the data can be synched when headteachers have access to Government provided wifi. Decentralized administrators have long been seen as critical for translating national policy into local action. However, they are frequently hampered by a combination of distracted management attention and unclear targets or benchmarks for key inputs, which encourages a laissez-faire status quo. In Tanzania, local governments in four regions have been able to contextualize data to meet their needs and use simple technology to prioritize their attention and decision-making. Our presentation showcases the significance of data driven decision making and continuous improvement of the system. We further highlight the important of simple and meaningful change and fostering proactive decision making at the local level.

Taken together, the case studies in this Report highlight the core components of an effective performance and accountability framework – a comprehensive system to optimize individual and system-wide performance through establishing outcome goals, tracking, and incentivizing progress against them The case studies show that effectiveness is not determined by implementing a set of individual policies, tools, and structures, but rather how they are collectively used to drive a culture that empowers its people to achieve a system’s ambitious agenda. The components of a performance and accountability framework that are featured in these case studies include: • System leadership that prioritizes a culture of high-performance around targets. • Outcome focused targets and supporting indicators that are well understood and communicated throughout the system, to define and quantify a system’s aspiration. • Quantitative data which is available and regularly shared across a system to gauge system performance, and to identify areas for additional focus or support. • Targeted and tailored support for system actors to help them improve performance, including a focus on specialized support for underperformers, driven by an understanding of data and evidence. • Performance routines and reporting with key system actors focused on reviewing progress, problem solving, and decision making to unlock barriers to improvement. • System engagement and site-visits/fieldwork that surfaces key challenges and opportunities to drive improvement. • Public engagement that keeps the public up to date on progress. Some of these components may require adaptation to different parts of the system. For example, targets may feature in performance agreements for senior leadership roles, but not for frontline staff who may be engaged in a more indirect way around targets (for example, communication of the reform with a focus on enhancing practice and improved outcomes). Similarly, formal performance routines that focus on a broader strategic reform are effective at driving a performance culture at senior levels, but not so for frontline staff who are rightly focused on teaching practice in their classroom. To be effective, embedding a performance and accountability culture across a system requires a nuanced and tailored application of these components to frontline staff, middle management, and senior levels.

The Mathematics Knowledge for Teaching (MKT) is a short survey (23 items) that measures primary grade teacher knowledge by a) math domains and b) pedagogical and content knowledge. Math domain included Number Sense, Operations, Geometry, and Measurement. Pedagogical knowledge was measured by problems that measured teacher understanding of Developmental Progression, Scaffolding, and Content knowledge. In this presentation, we will discuss the process of developing the MKT survey, highlight exemplary results from the Kyrgyz Republic, and then discuss the various uses of this survey. The MKT survey build from previous work in measurement of teachers’ MKT in the United States and other countries (Ball et.al., 2008; Cole, 2012). Our goal was to create an instrument that focused on the early primary grade and was easily adaptable to multiple contexts. To do this, we created an initial instrument, conducted cognitive interviews with math and learning experts form several countries, and then conducted a pilot in the Kyrgyz Republic and Nepal. In Kyrgyz Republic, the MKT test was administered to 323 primary grade teachers in 30 pilot schools as a pre-post training survey as part of the USAID Okuu Keremet! The survey was administered online in two languages. Analysis of pre-post test showed that the survey was effective in detecting changes in teacher knowledge across all math domains and pedagogical and content knowledge areas. In Nepal, we conducted cognitive interviews with teachers, providing additional insights into how teachers were thinking about early math knowledge. Finally, we conclude with the different potential uses for this survey, such as diagnosing and measuring changes in teacher knowledge over time and using it as professional development tool to develop teacher knowledge. We will discuss implications for the use of this tool for the wider development audience.

The Sustainable Development Goals (SDGs) for education represent a major departure from the Millennium Development Goals (MDGs) - at least if educational leaders act seriously in their pursuit - in at least two important respects. First, the goals now pertain to learning outcomes. Second, there is a great deal of focus on inequality in the SDGs. Taking note of this new dual emphasis of the SDGs, this paper assembles the largest database of learning outcomes inequality data that we know of, and explores key issues related to the measurement of inequality in learning outcomes, with a view to helping countries and international agencies come to grips with the key dimensions and features of this inequality. Two issues in particular are explored. First, whether, as countries improve their average cognitive performance (as measured by international learning assessments) from the lowest to middling levels, they typically reduce cognitive skill inequality or, more importantly perhaps, whether they reduce absolute lack of skills. Second, whether most of cognitive skills inequality is between or within countries. In dealing with these measurement issues, the paper also explores the degree to which measures of cognitive skills are “proper” cardinal variables lending themselves to generalizations from the field of income and wealth distribution—the field for which many measures of inequality and its decomposition were first applied. To do this, we look into whether using the item response theory (IRT) test scores of programmes such as TIMSS influence these types of findings, relative to the use of the underlying and more intuitive classical test scores. Patterns emerging from the classical scores are far less conclusive than those of the IRT scores, in part due to the greater ability of the IRT scores to discriminate between pupils at the bottom end of the performance spectrum. An important contribution of the paper is to examine the sensitivity of standard measures of inequality to different sets of test scores. The sensitivity is high, and the conclusion is that meaningful comparisons between test score inequality and, for instance, income inequality are not possible, at least not using the currently available toolbox of inequality statistics. Finally, the paper explores the practical use of school-level statistics from the test data to inform strategies for reducing inequalities.

Big Books Type 3 (8Pg, 2Col). Rita ra Bandar. These materials were originally developed by Room to Read. Under Creative Commons Attribution, Room to Read ceded copyright for these materials to the Government of Nepal. The Government of Nepal adapted these materials with the help of RTI International. These materials were delivered to Nepali school children in grades 1-3 in 6 districts in Nepal.

[Presentation delivered at CIES 2017]. Education system data has indicated a pattern of over-enrollment in the early grades in some low-income countries. One factor that may influence the observed enrollment bulge is access to pre-primary education programs, as children who do not enter school prepared could be more likely to repeat leading to more children enrolled than are of enrollment-age. This presentation reports on a research study undertaken in Uganda to better understand pupil enrollment and repetition in Primary 1 as it relates to pre-primary access. Data was collected from schools in a district with high reported repetition, lack of or low preprimary access, and high apparent dropout between grade 1 and 2, and a district with low reported repetition, high preprimary access, and low apparent grade 1 dropout. School records were reviewed to collect ages of enrolled pupils and repeater status, in order to calculate age patterns and repetition rates in Primary 1 for each district. In addition, parents of randomly sampled pupils were interviewed about their child’s preprimary attendance, repetition of primary one or other grades, and access to preprimary education programs. Teachers of the sampled pupils were also interviewed regarding pupil age and repetition history. Overall, data was collected on 1,909 pupils in 80 schools, and 1,792 parents were interviewed. The presentation will focus on the following research questions: 1) What are the estimated repetition rates as reported by parents and schools in Primary 1 in schools in the sampled low- and high-risk districts in Uganda; 2) What are the ages of the pupils attending Primary 1 in our sample of Uganda schools and how does this relate to repetition; 3) How do the reported repetition rates relate to parent report of ECD attendance and access; and 4) What is the discrepancy between school-reported and parent-reported repetition rates. Data collection methods, measurement of repetition and ECD attendance and access, and policy conclusions will also be discussed.

The purpose of this paper is to assess Mozambique’s education data and information systems’ ability to help formulate education sector plans and policies. Emphasis is put on existing policy over the last decade or so, with some attention to more recent trends. Thus emphasis is put on basic education, because basic education was the focus of policy attention in the most recent decade or two.

The 2010 Nigeria Education Data Survey (NEDS) was a nationally representative sample survey implemented primarily by the National Population Commission (NPC) in collaboration with the Federal Ministry Of Education (FMOE) and the Universal Basic Education Commission (UBEC). The 2010 NEDS has the following specific objectives: * Provide data on the schooling status of Nigerian children of basic education age, including factors influencing whether children ever enroll in school and why students drop out of school * Quantify household expenditures on children’s schooling by examining different patterns of expenditure by various background characteristics * Measure parents’ attitudes to schooling, including the quality of schooling and provide an understanding of attitudes that shape their willingness to send their children to school * Measure the frequency of student absenteeism and reasons for missing school in order to suggest possible approaches to maximizing attendance * Provide data that allows for trend analysis and State comparisons A very high overall response rate of 98% was achieved with interviews completed in 26,934 households.