Inclusion and education

CHAPTER
3
Data

---

KEY MESSAGES 
What data are collected and how they are used determine whether inclusion is served.
„
Historically, the region has focused data collection efforts on learners with special education needs and disabilities. 
But inclusion-related data collection must cover inputs, processes, outputs and outcomes on all learners and for 
uses other than just resource allocation.
„
Identifying groups makes those from disadvantaged populations visible but can reduce children to labels, which 
can be self-fulfilling. Not all children facing inclusion barriers belong to an identifiable or recognized group, while 
others belong to more than one.
Household surveys help disaggregate education outcomes at population level.
„
Household surveys, available for practically every country, disaggregate education data. In Mongolia, 92% of the 
richest youth but only 22% of the poorest complete secondary school.
„
Surveys also show intersecting characteristics: Among the poorest, girls in Turkmenistan but boys in North 
Macedonia are more likely to complete secondary school.
„
About 60% of Roma youth in the Balkans are out of school. In Montenegro, no poor Roma youth complete 
secondary school. In Georgia, internally displaced youth are seven percentage points less likely to complete 
secondary school than their non-displaced peers.
„
Formulating questions on nationality, ethnicity, religion, sexual orientation and gender identity can touch on 
sensitive personal identities. No question on ethnicity or language has been asked in the Turkish population 
census since 1965.
Statistical measurement of disability is catching up with the social model.
„
In nine education systems that applied the Child Functioning Module, the share of 5- to 17-year-olds with a 
functional difficulty in at least one domain was 7.5%, on average. In Georgia, Kyrgyzstan and Mongolia, the share 
of youth with disabilities in the out-of-school population is twice as large as their share of the in-school population.
„
Not all children with disabilities have special education needs, nor do all children with special education needs have 
disabilities. The share of students identified with special education needs ranges from 3.3% in Poland to 13% in 
Lithuania. Such variation is related to differences in country definitions, which stem from political decisions with 
historical roots.
School-level data point to persistent exclusion and segregation.
„
One in three students identified with special needs in Central and Eastern Europe is placed in special schools. 
Serbia reduced the share of children enrolled in special schools from 100% to 36% in 7 years, and the Republic of 
Moldova from 77% to 9% in 10 years.
„
In Slovakia, Roma constituted 63% of all children in special classes and 42% of those in special schools in 2018.
„
In the 2018 PISA results, schools in Bulgaria, Hungary and Slovakia were the least inclusive in the region, and 
among the least inclusive in the world, in terms of diversity of student populations by economic, social and 
cultural status.
It is necessary to monitor students’ experiences.
„
Cross-national learning achievement surveys show that about 2 in 10 children feel like outsiders in school, on 
average, with shares ranging from 1 in 10 in Albania to 3 in 10 in Bulgaria.
„
To foster inclusion, monitoring should not only serve the function of collecting data on inclusion but also be 
inclusive in methodology. The Monitoring Framework for Inclusive Education in Serbia has been integrated within 
the overall school quality assurance policy.
48
GLOBAL EDUCATION MONITORING REPORT 2021

---

Data on inclusion: The groups countries monitor vary ..................................................50
Censuses and surveys provide insights into inclusion in education ..................50
Measurement of disability has evolved along with its definition ....................... 56
Assessment criteria to identify special education needs can be arbitrary
and contentious .......................................................................................................................... 59
Labels affect those labelled and are self-confirming ..........................................60
Data for inclusion: The policies and results countries monitor vary ....................... 62
Student segregation occurs at several levels ............................................................... 62
Information on the share of students in special schools is incomplete .... 62
The concentration of vulnerable students varies by country ........................ 63 
Monitoring of inclusion in schools should be ambitious ......................................... 65
Data collection should promote inclusion......................................................................68
Conclusion ...........................................................................................................................................68
Data are critical to support inclusion in education.
First, data can highlight gaps in education opportunities 
and outcomes among learner groups. They can identify 
those at risk of being left behind and the barriers to 
inclusion. Second, with data on who is being left behind 
and why, governments can develop evidence-based 
policies and monitor their implementation (e.g. via 
resources, equipment, infrastructure, teachers and 
teaching assistants, anti-bullying strategies, parental 
involvement) and results (European Agency, 2011, 2014; 
Hollenweger, 2014b).
In defining results, few inclusion-specific outcomes can 
be distinguished from general education outcomes 
(Armstrong et al., 2010). For instance, data on where 
learners are being educated are needed. In addition, 
feelings of belonging, mutual respect and social esteem 
should be monitored (Watkins et al., 2014).
Qualitative data on such experiences can capture fine-
grained information that paints a drastically different 
picture than quantitative categorical data. Unlike 
population- or system-level indicators, such measures 
should describe learners’ individual experiences rather 
than those of learner groups or categories. One approach 
to a set of indicators involves systematically examining 
levels of authority, from schools to education ministries, 
and a range of results, including not just outputs and 
outcomes but also inputs and processes (
Table 3.1
).
Information on processes is difficult to collect and even 
more difficult to compare among schools or groups, 
let alone among countries. Frameworks for voluntary 
self-evaluation by schools or for programme evaluations 
are not necessarily suitable for official country-level 
monitoring of inclusion. Measuring inclusion is tied to how 
countries define it. While some aspects are part of most 
definitions, such as whether all students feel welcome in 
school, no single list of indicators is suitable everywhere. 
Criteria need to be locally determined and account for 
context, as vulnerabilities vary by place (Ainscow, 2005).
This chapter reviews the promise and potential obstacles 
of various approaches to collecting and analysing data 
to identify exclusion and to prompt action. It then looks 
at how countries collect data to monitor the effects of 
actions to make education systems more inclusive.
Feelings of belonging, mutual respect and 
social esteem should be monitored
49
C E N T R A L A N D E A S T E R N E U R O P E , C A U C A S U S A N D C E N T R A L A S I A

---