Inclusion and education

FIGURE 3.3: 
Among the poorest, there are considerable gender gaps in secondary completion
Difference in upper secondary school completion rate between poorest male and female youth, selected countries in Central and 
Eastern Europe, the Caucasus and Central Asia, 2011–18
-25
-20
-15
-10
-5
0
5
10
15
20
25
Percentage points
Caucasus and Central Asia
Mongolia
Turkmenistan
Kyrgyzstan
Georgia
Armenia
Kazakhstan
Tajikistan
Central and Eastern Europe
Estonia
Poland
Slovakia
Belarus
Ukraine
Rep. Moldova
Hungary
Czechia
South-eastern Europe
Montenegro
Slovenia
Bosnia/Herzeg.
Romania
Croatia
Serbia
Bulgaria
N. Macedonia
Females have higher attendance rate than males
Males have higher attendance rate than females
Source: 
World Inequality Database on Education.
Multiple characteristics can intersect 
to push people deeper into education 
disadvantage
52
GLOBAL EDUCATION MONITORING REPORT 2021

---

remote populations, migrant and displaced people, and 
the poor are also commonly recognized. Few countries 
link recognition of specific groups with a mandate to 
collect data on their inclusion in education, however.
Who is recognized in a census or survey may reflect 
political power and representation. Data that highlight 
inequality among groups are not always welcome for 
political reasons; groups in power may question their 
reliability and worry that drawing attention to such gaps 
will fuel resentment among the disadvantaged. A global 
analysis of 138 censuses in the 2000 round showed that 
more than one-third included no ethnic classification 
(Morning, 2008). Political changes can have a major 
impact on how groups are captured.
Household surveys have served an important function 
in highlighting the relative education progress of various 
ethnic groups. The UNICEF Multiple Indicators Cluster 
Surveys (MICS), for example, have cast light on Roma 
populations’ exclusion from education in south-eastern 
Europe, especially at the upper secondary school level. 
In Kosovo,
1
almost 60% of youth among the Ashkali, 
Egyptian and Roma communities were not attending 
upper secondary school in 2019/20, with practically no 
change observed since 2014. In Montenegro, 75% of Roma 
youth living in Roma settlements did not attend upper 
secondary school in 2018, while the national average 
was 6% (
Figure 3.4
). Just 3% of Roma youth completed 
secondary school, compared with the national average of 
86% (Montenegro Statistical Office and UNICEF, 2019).
Questions on nationality, ethnicity or religion touch on 
sensitive points of personal identity and can be intrusive. 
They may also trigger fear of persecution. No question on 
ethnicity or language has been asked in Turkey since the 
1965 population census. However, a question on Kurdish 
ethnicity has been asked in successive rounds of the 
Demographic and Health Surveys (Koc et al., 2008), while 
learning assessment surveys, such as the Programme for 
International Student Assessment (PISA), identify students 
who do not speak the language of instruction at home. 
Various factors hamper identification of immigrant and 
refugee populations for policy purposes, especially when 
those populations are transient. Statistical offices use 
techniques to adjust overall census results, but these 
cannot replace fine-grained mapping of such populations,