By Manos Antoninis and Marcos Delprato

World leaders are preparing to put pen to paper this week to conclude years of negotiations. The most emblematic pledge is likely to be the statement in the 2030 Agenda for Sustainable Development document that “no one will be left behind”. In order to meet this pledge, across all areas of development and across all countries, we need measures to track progress in reducing inequalities. This blog looks at what these might these look like for education.

At the Global Monitoring Report we have been looking at inequalities in education for many years. This week, with the adoption of target 4.5 that aims to “eliminate gender disparities in education and ensure equal access to all levels of education and vocational training for the vulnerable”, we hope there will be renewed urgency to find better ways to report and address these inequalities.

In December, we co-organized a workshop with the UNESCO Institute for Statistics on this issue. Last week, at the 2015 UK Forum for International Education and Training (UKFIET) International Conference on Education and Development, we presented some of the challenges.

A first key challenge is determining which inequality measure to use. For the past 15 years, the parity index has been the common measure for differences in education participation between males and females. However, this is just one of a rather large possible set of measures to choose from – each with different advantages and disadvantages. And, unfortunately, different measures can yield entirely contradictory conclusions, as Patrick Montjourides of the UNESCO Institute for Statistics clearly showed at the conference, building on earlier work.

Aware of this challenge, the Technical Advisory Group on education indicators, which presented its proposal at the World Education Forum in May, left open the choice of which inequality measure should be used. To the greatest extent possible inequality should be addressed in a consistent way across the SDGs. For that reason, the final proposal of the Inter-Agency and Expert Group on Sustainable Development Goal Indicators, to be discussed in Bangkok at the end of October, is hotly awaited.

In the past, the GMR has mainly reported on inequality by country, which is easy on the eye, as shown in our World inequality Database on Education, but not easy to communicate to a global audience. We have also tried to communicate headline messages. This has included showing, for example, that in low and middle income countries, compared with the richest 20%, the poorest 20% are five times as likely not to complete primary school and six times as likely not to be literate.

At the UKFIET conference we presented preliminary results on inequality at the regional and global level based on an analysis of 78 low and middle income countries across eight different education indicators, four different inequality measures and three individual characteristics observed at two points in time, 2000 and 2010. Two issues came up strongly that need further consideration.

First, many inequality measures that are easy to communicate may not provide sufficient insights.

For example, the parity index in the primary completion rate between the richest 20% and poorest 20% tends to increase the closer we get to universal primary completion. We therefore need measures that can tell us whether a country is more unequal than what would be expected relative to its level of educational development.

Second, we need to be much clearer
about what constitutes progress. 

The following graph ranks young people in sub-Saharan Africa from the poorest to the richest and compares their record in completing lower secondary education in 2000 (blue line) and 2010 (red line). The diagonal line represents perfect equality; the furthest a line is to the right of the diagonal, the higher the inequality. For example, in both 2000 and 2010 the poorest 40% of young people represented less than 20% of lower secondary school completers. But while inequality seems to have fallen between 2000 and 2010, it is the middle classes that benefited most, while the poorest have done worse overall.

These are just two of several outstanding challenges. Some cannot be easily addressed, for example the exclusion of some of the poorest populations from the samples of many surveys. Others can be. The UIS has committed to following up on some of these, preparing for the creation of an Inter-Agency Group, which was proposed during the World Education Forum. It also announced at the UKFIET conference a proposal to establish an Observatory of Equity and Inclusion.

‘Leaving no one behind’ is a nice catchphrase, but tracking equity is easier said than done. We will need better data and closer collaboration if a transparent, easy to communicate, yet representative picture is to be produced on closing equity gaps and tracking the progress of vulnerable groups in education in the future.

At the GMR we are sometimes criticized for spending too much time trying to measure inequality that could be better spent providing solutions to how inequalities can be reduced. Yet more thinking and work towards a consensus on measuring education inequalities can and will serve a broader purpose: we need to get the issue clearly on the radar of policy makers in countries where the scale of inequality is still not sufficiently recognized and acted upon.