G LO B A L E D U C AT I O N M O N I TO R I N G R E P O R T 2 0 1 6 SUMMARY The new agenda marks an important step forward with its emphasis on completion relative to participation. Over 2008–2014, the primary completion rate was 92% in upper middle income, 84% in lower middle income and 51% in low income countries; in the last, among the poorest girls, it was 25%. The upper secondary completion rate was 84% in high income, 43% in upper middle income, 38% in lower middle income and 14% in low income countries. Not even the richest in high income countries achieve universal completion at 93%. In low income countries, just 1% of the poorest girls complete upper secondary school. COMPULSORY AND FREE EDUCATION Out of 190 countries with data on compulsory education, 23% stipulate fewer than nine years of compulsory education One of the thematic indicators is the number of years of (i) free and (ii) compulsory primary and secondary education guaranteed in legal frameworks. Out of 190 countries with data on compulsory education, 44 countries (23%) stipulate fewer than nine years of compulsory education. The median number of years is 9 for compulsory education, 11 for free education. But the concept of free education is problematic in terms of measuring progress on the target: even where fees are abolished, multiple education costs may burden families. The share of total education expenditure borne by households at each level is a stronger indication of the extent to which education is free. QUALITY The proposed monitoring framework does not focus on quality except in indicators related to learning outcomes and equity. Using an indicative framework to guide discussions of quality, two issues are selected: classroom-related inputs and processes. Textbook availability and use are critical dimensions of quality, yet school visits and classroom observations indicate official data on them are not very reliable. In Chad, about 90% of grade 2 and 6 reading and mathematics students had to share textbooks with at least 2 others. It is difficult to advocate the use of classroom observation to compare education systems. Yet broadly consistent monitoring tools bring critical issues of teaching practice and pedagogy to policy-makers’ attention. A survey of 15,000 classrooms in Brazil, Colombia, Honduras, Jamaica and Peru showed that teachers spent about 60–65% of their time on academic instruction, well below the recommended 85%. It is important to continue the search for tools that are adaptable yet reliable, valid, cost-efficient and easy to use at scale. LEARNING OUTCOMES To improve learning outcomes, a benchmark showing whether progress is taking place is vital. However, there are questions about what ‘relevant and effective learning outcomes’ are, how to measure them and how to use the findings. Measuring the proposed global indicator – reading and mathematics skills – requires consensus on the content of the learning outcomes to be assessed, the quality standards assessments need to meet and the reporting and defining benchmarks to be used. LEARNING OUTCOME MEASURES: DEFINING THE CONTENT To define a minimum proficiency level in domains such as reading and mathematics, any assessment needs basic parameters. What happens when common ground between different curricula needs to be found? What is an expected progression of learning across curricula? What questions show that a learner has reached a particular level of proficiency? How are the levels of proficiency defined? Two contested issues demonstrate the tensions. First, early-grade assessment of reading and mathematics is divisive, for political and technical reasons. Yet it draws attention to major challenges on the ground. In Malawi in 2012, 90% of grade 2 students could not read a single word in Chichewa; almost 40% still could not do so by grade 4. 39

Select target paragraph3