The monitoring framework supporting SDG 4 on education is ambitious, even if some of the hardest questions in education development remain sidelined. It has a major formative role, serving to signal issues that deserve attention and for which countries should invest in monitoring. At the same time, major efforts are under way to develop indicators, standards and tools to strengthen data comparability across countries, a process requiring close collaboration among international agencies, countries, funders and experts.

There are 11 global indicators for SDG 4. The UNESCO Institute for Statistics (UIS) is the sole custodian agency for eight indicators and collaborates with the International Telecommunication Union (ITU) for the indicator on information and communications technology (ICT). The custodian agency for one indicator on early childhood development is UNICEF and for aid to scholarships is the OECD. With an additional 32 thematic indicators, that brings the SDG 4 monitoring framework to 43 indicators.

The UIS coordinates developments in global and thematic indicators with Member States and agencies through the Technical Cooperation Group on the Indicators for SDG 4 – Education 2030 (TCG), which it convenes with UNESCO. As of 2018, four indicators will be reported on for the first time (participation in adult literacy programmes, comprehensive sexuality education, school-based violence and attacks in schools), bringing the total to 33 out of the 43 indicators. Work continues or is about to begin on language of instruction, distribution of resources and teacher professional development.

The Global Alliance to Monitor Learning, also convened by the UIS, coordinates work on more sophisticated learning outcome indicators, primarily minimum proficiency in reading and mathematics, adult literacy and digital literacy. The UIS is pursuing three alternative strategies towards linking and benchmarking in reading and mathematics proficiency. First, it supports an initiative that will see students in selected Latin American and west African countries take both a regional and an international assessment to enable more robust comparisons across surveys. Second, it has carried out a mapping of the content of various assessments, through which experts will assign a level of difficulty to items across surveys to place them on a reporting scale. Third, it continues efforts to link proficiency scales using statistical techniques.

Taking stock of the Education for All era, 2000–2015

The 2015 Education for All Global Monitoring Report (EFA GMR) assessed the EFA record but was based largely on 2012 data. An updated assessment has been made, based on 2015 data. It does not change the conclusion – progress during the EFA era, while significant, ultimately fell well short of reaching the targets – but is an important stocktaking exercise, looking ahead to 2030.

Two key findings stand out. With respect to EFA goal 2 on primary completion, after stagnant participation and completion rates until about 1997, the primary gross enrolment ratio, net enrolment rate and gross intake ratio into the last grade of primary picked up until about 2008, then stalled (Figure 6a). By contrast, with respect to goal 5 on gender equality, although the target of parity in enrolment by 2005 was not achieved, progress continued throughout the 1990s and 2000s, reaching parity in 2009 in primary and secondary education and near parity in youth literacy in 2015. Disparity remains in adult literacy, with 63% of illiterates being female, and has reversed in tertiary education, with males now less likely to participate (Figure 6b).

Monitoring the education status of migrants and displaced populations presents numerous challenges

The SDG monitoring framework explicitly focuses on disaggregating indicators by various characteristics associated with disadvantage. However, systematic data on the education status of immigrants and refugees are patchy. In the World Bank’s Microdata Catalogue, over 2,000 out of almost 2,500 household surveys contain information on education, but only around one out of seven of those includes migration, and a smaller fraction include displacement.

Migrant households are mobile and less likely to be present for household survey enumerator visits or to be interviewed due to language barriers or legal concerns. And since migration flows can change rapidly, sampling frames may not keep up. This particularly affects displaced people: Data tend to be collected more systematically in refugee camps, but less than 40% of refugees and even fewer IDPs live in camps.

Ensuring inclusion of immigrants and refugees in standard, general-purpose surveys may not always solve sampling and data collection issues. Standard surveys do not capture the dynamism of the migration phenomenon and may be too infrequent to generate timely information. Flexible approaches, including research-focused surveys linking sending and receiving communities, as well as rapid data collection using non-random sampling, can be more effective. Finally, surveys may not capture education–migration dynamics or origin country qualifications.

In March 2016, the United Nations Statistical Commission began hosting an Expert Group on Refugee and IDP Statistics. Comprising 40 member states and at least 15 regional and international organizations, it has made recommendations on improving data collection. UNESCO and UNHCR are developing the Refugee Education Management Information System, a free, open-source, web-based tool to help countries collect, compile, analyse and report refugee education data.