Education and Training Monitor 2022

Comparative report

Chapter 1. A new indicator brings the conversation closer to the roots of inequity in education

1.1. Socio-economic status has an immense effect on educational performance

For the very first time, Figure 1 contrasts severe educational underperformance among learners from disadvantaged socio-economic backgrounds with that of learners from advantaged socio-economic backgrounds. Severe underperformance here means a low score on all three PISA scales simultaneously (reading, maths and science)⁷. Socio-economic status is captured by the OECD's index for economic, social and cultural status⁸, comparing its lowest and highest quarters. This broad measure of inequity tops no fewer than 35 percentage points in Romania (39.0) and Bulgaria (38.3)⁹. The gap is 19.3 percentage points on average across the EU, with students of low socio-economic status 5.6 times more likely to underachieve in school education than students of high socio-economic status¹⁰.

Students with disadvantaged socio-economic backgrounds are almost six times more likely to underperform in education.

Figure 1. Educational underperformance is coupled with socio-economic status

Download data Monitor Toolbox

The outsized role of socio-economic status is not limited to a minority of Member States¹¹, but there are huge disparities between countries as to the size of underachievement gaps. Firstly, a gap of under 10 percentage points only exists in Estonia and Finland (5.1 and 9.9, respectively). These two countries are generally top performers in PISA and thereby show that there is no inherent trade-off between excellence and equity. Secondly, country variation is so substantial that students from advantaged socio-economic backgrounds in some countries are still much more likely to underperform than students from disadvantaged socio-economic backgrounds in other countries.

Box 2. How does the new EU-level indicator compare with the existing EU-level targets on underachievement in reading, maths and science?

The new EU-level indicator for equity in education uses a combined measure of severe educational underachievement. Instead of looking at low achievement in reading, maths and science separately – as is the approach of the complementary EU-level target on underachievement in basic skills (Section 7.1) – it captures underperformance on all three proficiency scales simultaneously.

It is more likely for a 15-year-old to have a low score on one PISA scale than on all three PISA scales simultaneously. Across the EU, the shares of underachievement in reading (22.5%), maths (22.9%) and science (22.3%) are all higher than the share of underachievement in all three domains combined (13.0%). This illustrates how the new EU-level indicator on equity is built on a more severe definition of educational underachievement. Here, eight Member States yield shares above 15%: Bulgaria (32.0%), Romania (29.8%), Cyprus (25.7%), Malta (22.6%), Greece (19.9%), Luxembourg (17.4%), Slovakia (16.9%) and Hungary (15.4%).

Underachievement gaps by socio-economic status are also different when looking at the three PISA domains separately as opposed to combined (Figure 2). The new EU-level indicator for equity in education is more favourable to some countries and less favourable to others. Some Member States compare more positively with the EU average, such as Czechia (in particular when compared to reading and maths separately), Slovenia (maths, reading) and Denmark (science). Other Member States compare more negatively with the EU average, such as Malta (in particular when compared to reading separately) and Cyprus (maths).

Figure 2. The underachievement gap between low and high socio-economic status when combining reading, maths and science, and when looking at them separately

Source: European Commission (Joint Research Centre) calculations based on OECD’s PISA 2018 data.

Note: the indicator captures the percentage point difference between the lowest and highest quarters of socio-economic status in terms of average underachievement in reading, maths and science (seperately); countries are shown in descending order according to average underachievement gap in reading, maths and science (combined).

Download data

Even though education and training systems all across the EU try to eliminate any negative effects arising from learners’ individual circumstances, low socio-economic status may be the largest single explanatory factor obstructing equality of opportunity for many different disadvantaged groups¹². Low socio-economic status permeates the school experience in various ways. Its effects are evident for the youngest age brackets with comparative data available¹³ and even extend to 15-year-olds’ expectations of completing tertiary education in the future¹⁴. Socio-economic status is so engrained in education and training systems that learners with low socio-economic status may end up clustered in schools with a concentration of similarly disadvantaged peers¹⁵.

The passing down of educational disadvantage throughout the generations remains pervasive in all EU-level target domains covered in the 2022 Education and Training Monitor with proxies for socio-economic background available. For instance, young people whose parents have a low level of education are nine times more likely to become early school leavers (Chapter 3) and 48.6 percentage points less likely to attain a tertiary educational qualification (Chapter 5) when compared to young people whose parents have a high level of education¹⁶.

1.2. Complementary evidence on specific disadvantaged groups adds further texture

Inclusive education presupposes an assessment of the disadvantage experienced by specific population sub-groups. Some elements of this disadvantage may be due to socio-economic status, whereas others could be attributed to factors such as prejudice, discrimination, language barriers or a lack of appropriate services.

Women outperform men in virtually all EU-level education statistics. Combined underachievement in reading, maths and science is about 3 percentage points less common among girls (with the socio-economic gap in underachievement being nearly identical). The risk of early leaving from education and training is 3.5 percentage points lower among girls (Chapter 3) and tertiary educational attainment is no less than 11.1 percentage points more common among women (Chapter 5). Education and training systems do nonetheless contribute towards engraining outdated gender stereotypes¹⁷. Gender equality measures in education are particularly targeted at creating a better gender balance in certain fields of study, and at developing equality plans, particularly in higher education institutions.

Migrant young people are 12.9 percentage points more likely to become early school leavers than the overall EU average (Chapter 3)¹⁸. They are 7.1 percentage points less likely to attain a tertiary education qualification (Chapter 5). The later migrants arrive in the education trajectory, the more education and training systems struggle to integrate them. The underperformance gap between low and high socio-economic status is 20.9 percentage points when the test language is not spoken at home¹⁹, versus 17.4 percentage points when it is (Figure 3). The difference is much more pronounced in Hungary (59.1 versus 27.8), Romania (64.2 versus 37.7), Slovakia (49.2 versus 23.9) and Portugal (40.2 versus 20.2).

Figure 3. A different home language widens inequities further in some Member States

Download data Monitor Toolbox

The new EU-level indicator domain on inclusion and equity must always remain a work in progress, taking on board the latest evidence as the knowledge base is strengthened further over time²⁰. Such a flexible approach enables the monitoring exercise to acknowledge, for instance, periodical data on young people from racial and ethnic minorities who are marginalised and/or discriminated against²¹, young people with disabilities and/or special education needs²², and school-age refugees (Box 4). Furthermore, non-discrimination in education is itself a sub-dimension that may strengthen the EU-level indicator domain on inclusion and equity in the future²³.

Box 3. Investigating the share of out-of-school 15-year-olds

Another fundamentally overlooked issue may be that of out-of-school young people²⁴, who do not have a chance to underperform in the first place. Figure 4 captures the share of unenrolled 15-year-olds across the EU. These 15-year-olds may be at different stages of each country’s educational pathway (whether lower or upper secondary education), yet still at compulsory schooling age in most Member States. It is worth noting that low performing countries in Figure 1 (Romania, Bulgaria) also yield significant shares of out-of-school young people in Figure 4 (16.8% and 14.5% respectively). Further investigation of this indicator is warranted²⁵.

Figure 4. Around 3% of all 15-year-olds are not enrolled in the national education system

Download data Monitor Toolbox

1.3. Specific policies can alleviate inequities but remain underexploited

Equity and inclusion in education and training is a challenge across all Member States. Just like the scope and determinants of the problem vary from country to country, so do the various education and training systems in terms of institutional stratification²⁶ and specific policy measures to alleviate the effects of socio-economic status. This section summarises a few examples of the top-level financial and non-financial support that is provided by (and to) the education and training systems across the EU.

Most Member States provide additional financial support for schools²⁷ with disadvantaged students, either upon application or it is allocated automatically. Only Croatia, Cyprus, Greece, Hungary, Malta, Luxembourg and Romania do not report such measures. Moreover, top-level regulations or recommendations on the socio-economic composition of schools²⁸ exist in Belgium (Flemish Community), Hungary, Romania, Slovakia, Spain and Portugal, while the school catchment area can be adjusted in France, Hungary and Slovenia.

A 2022 OECD report using data from the 2018 Teaching and Learning International Survey (TALIS) highlights how experienced schoolteachers with, for instance, high levels of self-efficacy or digital skills²⁹ tend to be clustered in socio-economically advantaged schools. The same report also confirms, through a combination of TALIS and PISA data, that there is a direct link between an uneven distribution of experienced teachers and lower than average reading scores. Remedial policies could tackle such disproportionate clustering, but the latest EU comparative overview³⁰ suggests that neither financial nor non-financial incentives are commonplace.

Firstly, financial support to teachers in disadvantaged schools (increased basic statutory salaries, allowances or other financial support) is crucial³¹. However, it is not prevalent across the EU. Only nine countries report such top-level incentives (France, Greece, Hungary, Lithuania, Poland, Romania, Slovakia, Slovenia and Sweden). Secondly, non-financial incentives in terms of better working conditions³² or career benefits³³ can raise the attractiveness of teaching in disadvantaged schools. Yet only six Member States report such incentives (Belgium³⁴, France, Lithuania, Slovenia, Spain and Portugal).

Box 4. Welcoming Ukrainian refugees

Russia’s invasion of Ukraine forced many people to flee their homes. Women and children in particular are seeking protection from the war and Eurostat data suggest that over 1.4 million of children and young people are beneficiaries of international protection³⁵. Key priorities concern support to Ukrainian young people, to the education systems in Member States that are welcoming them as well as to Ukrainian teachers. Early evidence points at the difficulties of making sure school-age refugees are actually enrolled in school³⁶.

School education³⁷ in most EU education and training systems promote the integration of newly arrived children from Ukraine into regular classes, combined with intensive support for learning the language of instruction (and other subjects). Relatively fewer education systems favour the initial integration of refugee learners in separate classes. However, there is general consensus that longer-term efforts need to be focused on the integration of these learners into local schools.

This requires support to education systems in several fields:

organising reception and admission processes (including expanding capacity)
preparing educational institutions and staff to include displaced children
running targeted activities to help include displaced children in education
engaging with displaced families and communities and helping them maintain the link with Ukraine
taking long-term measures to promote inclusive education
taking specific measures for early childhood education and care³⁸.

The majority of education systems also provide top-level support for refugee learners who wish to follow distance learning according to the Ukrainian curriculum.

In higher education³⁹, Member States report a variety of large-scale measures to support the integration of refugee students, with most having had measures in place since before the Russian invasion of Ukraine⁴⁰. Only six higher education systems monitor the integration of refugees in their institutions, mostly tracking enrolment data, with no longer-term monitoring yet established. Recognition of previous educational attainment can be a particular challenge, particularly when evidence of qualifications cannot be provided. This is the reason why article 7 on the recognition of qualifications held by refugees and displaced persons was included in the Lisbon Recognition Convention.

In the field of vocational education and training (VET) and adult learning, the Commission⁴¹ has invited Member States to

ensure that people's skills and qualifications can be valued, assessed and quickly recognised, regardless of whether documentation is available
provide targeted upskilling and reskilling opportunities, VET and/or practical workplace experience
ensure quick access to initial VET, including apprenticeships, and explore ways to prolong ongoing stays of Ukrainian vocational learners
make opportunities available for adults fleeing Russia's war of aggression against Ukraine to access general education, including through second chance schooling, as well as enrolment in higher education institutions.

In higher education⁴², public funding is rarely provided on the basis of equity targets. Only in France and Italy are higher education institutions rewarded for meeting agreed targets in widening access, participation or completion. Top-level financial support in higher education is more commonly geared toward student accommodation, transport and meals. Only in Belgium⁴³, Ireland and Sweden is no such indirect support reported.

In terms of non-financial measures to support equity in higher education, only seven systems⁴⁴ require or recommend higher education institutions to offer staff training on diversity or inclusion (Figure 5). While eight systems offer higher education institutions financial support for such training⁴⁵, an additional nine offer non-financial support⁴⁶, such as administrative support or pedagogical materials and instructors.

Figure 5. A third of all Member States do not support higher education institutions in their offer of staff training on diversity or inclusion

Source: Eurydice 2022.

Download data

In a nutshell

No education and training system manages to decouple performance from socio-economic status. Young people from disadvantaged socio-economic backgrounds are almost six times more likely to underachieve at age 15 than those from advantaged socio-economic backgrounds, suggesting a strong intergenerational transmission of educational disadvantage. Such inequities do not emerge at age 15 and do not stop there. Stepping up the focus on equity in education is likely to support future progress in all existing EU-level target domains. The 2022 Education and Training Monitor demonstrates key equity challenges across other domains, with underperformance widespread among specific population sub-groups, and often clustered in the same schools or areas.

Notes

7.This is in contrast to low achievement in reading, maths and science separately, which is captured by an existing EU-level target (see Section 7.1 and also Box 2 in this section). Section 7.1 reiterates how there has been negligible progress towards reaching the existing EU-level target. An insufficient focus on equity in education may be a primary cause of this negligible progress. The new EU-level indicator domain on equity in education brings the conversation closer to the root of the problem. It is worth noting however, that the share of severe underperformance may be underestimated. This is partly because the underpinning PISA tests may have excluded newly arrived migrants, learners with language difficulties, or learners with disabilities. Student exclusions from PISA 2018, albeit small, were attributed to functional disability, intellectual disability, language or other reasons.
8.This well-established index is a measure of students’ access to family resources (financial capital, social capital, cultural capital and human capital). It comprises elements such as parental level of education, parental occupational status and various home possessions.
9.In both countries, around half of all 15-year-olds in the lowest quarter of socio-economic status underperform across reading, maths and science.
10.Comparing the two latest rounds of data collection (2015 and 2018), the gap has not changed substantially at EU level since 2015 when it stood at 18.8 percentage points and a 5.9 ratio. The addition of PISA 2022 in December 2023 will shed further light on trends over time, and may confirm whether gaps have widened since the onset of the COVID-19 crisis.
11.Even among the seven countries with overall shares of underachievement below 10%, the underachievement ratio between low and high socio-economic students ranges from 3.5 (Estonia) to 6.0 (Poland).
12.See the 2021 Council conclusions on equity and inclusion in education and training in order to promote educational success for all.
13.The 2019 Trends in International Mathematics and Science Study (TIMSS) by the International Association for the Evaluation of Educational Achievement (IEA) is partially focused on students in grade 4 (age 10 on average). Across the EU, 4.5% of fourth graders reveal a low performance in the TIMSS tests for both maths and science. This share is only 1.0% among high socio-economic groups (measured on the basis of parental education), versus 9.9% among low socio-economic groups. Monitor Toolbox
14.Strikingly, across the EU, whereas 81.8% of 15-year-olds from advantaged socio-economic backgrounds expect to complete tertiary education, only 45.4% of 15-year-olds from disadvantaged socio-economic backgrounds expect to do the same. Monitor Toolbox
15.PISA’s ‘isolation index’ can be used as a proxy for such school segregation. The index ranges from 0 (no segregation) to 1 (full segregation). Overall, the EU scores a 0.16. Segregation is, on average, lowest in Croatia (0.10), Finland (0.10) and Cyprus (0.10), and highest in Bulgaria (0.29) and Slovakia (0.29). Monitor Toolbox The European Expert Network on Economics of Education (EENEE) published a 2021 analytical report on school segregation that uses intra-class correlations on the basis of parental education and immigration status, both clearly correlating with the main indicator in Figure 1.
16.Data for early leavers from education and training (Chapter 3) and tertiary educational attainment (Chapter 5) are based on a 2021 ad hoc module of the EU Labour Force Survey (LFS).
17.Important elements that deserve further attention include, for instance: (a) the revision of textbooks and digital resources through a gender equality lens; (b) the need for gender-sensitive teaching as of ECEC; and (c) the fighting of gender-based bullying (including cyber-bullying).
18. Being born in another EU country or in a non-EU country are two vastly different concepts. Yet the difference is small in terms of early school leaving (21.4% versus 21.6%, respectively).
19. Migrant background is difficult to analyse meaningfully using the PISA data that underpins the equity main indicator (Figure 1). A first problem concern definitions, with non-immigrant students defined as ‘students whose mother or father or both was/were born in the country/economy where the student sat the PISA test, regardless of whether the student him/herself was born in that country or economy’. A second problem concerns small sample sizes, particularly for Bulgaria, Lithuania, Poland, Romania and Slovakia. However, the test language being spoken at home is an imperfect proxy for migrant background too, and may misclassify some students.
20. The Commission is examining on a wider scale the obstacles to collecting ‘equality data’ and is enabling the exchange of best practices. This is to encourage Member States, in full respect of their national contexts, to move towards collecting data disaggregated on the basis of all the relevant discrimination grounds.
21. As a prime example, the EU's Fundamental Rights Agency (FRA) reproduced early school leaving statistics for their 2019 Roma and Travellers Survey. More than half of surveyed 15-24 year-olds complete at most lower secondary education in Belgium (59% of Roma and 71% of Travellers), France (84% of Travellers), Ireland (70% of Travellers) and the Netherlands (88% of Roma and 62% of Travellers and Sinti).
22. Albeit a simplified proxy, an important addition to the EU Labour Force Survey data will be two new biennial variables on self-perceived general health and self-assessed limitations in daily activities because of on-going (physical, mental or emotional) health problems.
23. FRA captures self-reported experiences with discrimination in educational institutions (as a parent or as part of one’s own education).
24. Shares of out-of-school youth and adjusted net enrolment rates are an accepted complement in measures of educational poverty. Such indicators are calculated by the World Bank as a proxy for schooling deprivation, and by UNESCO as SDG indicator 4.1.4. The latter was reported by OECD in Education at a Glance 2021.
25. Some limitations are worth flagging. The indicator may unintentionally reflect (however negligible) non-resident populations enrolled in domestic programmes, resident populations enrolled in non-domestic programmes, or home schooling.
26. There are a number of institutional characteristics that tend to stratify an education system, which are therefore commonly associated with educational inequity. Examples are the early tracking of learners in a highly differentiated system, or an overreliance on grade repetition. A 2020 Eurydice report captures many of these institutional characteristics. Tracking already starts under the age of 13 in Germany (10), Hungary (10), Austria (10), Czechia (11), Slovakia (11), the German and Flemish Communities in Belgium (12), Ireland (12), Luxembourg (12) and the Netherlands (12). Parallel educational structures exist all the way throughout general education in Latvia, Lithuania and Spain. Grade repetition is particularly frequent in Belgium, Luxembourg, Portugal and Spain, all with at least one-fifth of 15-year-olds having repeated a grade at least once.
27. Throughout this section, top-level measures to improve equity in school education are taken from a 2020 Eurydice report. While this is the most recent comparative overview available, the Education and Training Monitor’s country reports feature more detailed and up-to-date country-specific examples.
28. These administrative measures are related to school admission policies or classroom grouping methods.
29.See Chapter 8.
30. See the 2020 Eurydice report.
31. See the 2020 Eurydice report.
32. For instance, reduced teaching time, reduced class size, job security, or access to mentoring/coaching.
33. For instance, a preferential next appointment or faster career progression.
34.Not including the German-speaking community.
35. Data for Germany, Czechia, the Netherlands and Hungary are not available. Data for France and Ireland are not included as the age breakdown is not available.
36. See a 2022 Eurydice report.
37. See a 2022 Eurydice report.
38. On 30 June 2022, the Commission published an overview of collective experience and knowledge that aims to disseminate shared expertise, information on good practice and practical insights supporting the inclusion of displaced children from Ukraine in education.
39. See a 2022 Eurydice report.
40. Higher education institutions have a substantial degree of autonomy, and may have taken measures at their own initiative to help refugee learners from Ukraine to pursue their studies in their institutions. Although such measures may be significant, they are likely to vary from one institution to another and are not considered large-scale.
41. On 14 June 2022, the Commission presented operational guidelines to support Member States in applying the Temporary Protection Directive in terms of access to the labour market, VET and adult learning. This new guidance builds on examples collected through a dedicated survey on VET-related measures. The results provide an overview of Member States actions to date, including examples of good practices that can serve as inspiration to others. These include accelerated procedures, equivalence of studies and validation procedures, individual plans, mentoring and counselling, work-based learning, and preparatory classes, including those on language and interpersonal skills.
42. Throughout this section, examples of top-level measures to improve equity in higher education are taken from a 2022 Eurydice report. While this is the most recent comparative overview available, the Education and Training Monitor’s country reports feature more detailed and up-to-date country-specific examples.
43. No top-level support in the Flemish and German-speaking Communities. Only one out of three support types in the French community.
44. Belgium’s Flemish Community, Czechia, Estonia, France, Ireland, Italy and Spain.
45. Financial support for training on diversity or inclusion is reported in Austria, Belgium’s Flemish Community, Ireland, Italy, Malta, Poland, Romania and Slovenia.
46. Non-financial support for training on diversity or inclusion is reported in Belgium’s German-speaking community, Croatia, Estonia, Finland, France, Greece, the Netherlands, Spain and Sweden.

Previous Next