Deprivation and Marginalization in Education (DME)

About the data set

”Marginalization in education is a form of acute and persistent disadvantage rooted in underlying social inequalities. It represents a stark example of ‘clearly remediable injustice’.” UNESCO, Education for All, Global Monitoring Report 2010

Education is a basic human right. It is also a catalyst for poverty reduction, economic growth, and social mobility. Ensuring that all citizens receive a good quality education should be one of the central priorities of all governments. Most countries endorse the principle of equal opportunity in education. At the heart of that principle is the idea that what children achieve should reflect their efforts and their talents, not their social circumstances. Yet deep and persistent inequalities based on wealth, gender, ethnicity, language, and location point to marked disparities in life chances.

The Deprivation and Marginalisation in Education (DME) data set documents the different dimensions of marginalization using a number of measures. These include:

  • ‘Education poverty’ defined as the share of the population aged 17-22 with less than 4 years in school (the minimum needed to gain basic literacy skills), and less than 2 years in school
  • National disparities in school attendance and years of education associated with social characteristics, such as parental wealth, gender and other markers for disadvantage.
  • The composition of the ‘bottom 20%’ of the national distribution in education by years in school

This interactive site allows users to explore national data in all of the above areas. It can be used to identify social groups and regions experiencing marginalization in education, capture national inequalities, and make comparisons across countries. Instructions in each section provide guidance on how to construct figures.

It should be emphasized that the DME data provides a partial picture. It does not cover important dimensions of education, including learning achievement. Moreover, caution has to be exercised in cross-country comparisons because of variance in survey years. Even so, school attendance and accumulated years in education are important indicators in their own right.

The development of the DME database was a collaborative exercise between UNESCO’s Education for All Global Monitoring Report team and the Department of Economics at the University of Göttingen, Germany, which computed the data. Statistics are drawn from Demographic and Health Surveys as well as Multiple Indicator Cluster Surveys.


  1. Groups for which the total sample was smaller than 150, or for which those in the relevant age group (usually age 17-22) was smaller than 30, have been excluded. The data date from between 1999 and 2007, depending on the country. The source and year in which each survey was conducted can be found by downloading the full data set.