Recent years have witnessed a remarkable increase in educational investment in Ireland and DEIS has followed this trend. The most salient characteristics of DEIS are its focus on concentrations of disadvantage, and the earmarked allocation of additional resources, which may explain its efficiency given its relatively modest share in the overall budget for education. Potential areas for improvement include the further fine-tuning of the indicator(s) used to measure disadvantage and the revision of the algorithm for the classification of schools into different DEIS categories. The possibilities for additional support for the most disadvantaged schools as well as disadvantaged students outside DEIS schools are explored, and the question of the responsiveness of the scheme to evolving needs is discussed.
OECD Review of Resourcing Schools to Address Educational Disadvantage in Ireland
3. Resourcing schools to address educational disadvantage
Abstract
Context and features
This section examines the volume as well as the allocation mechanisms of the Delivering Equality of Opportunity In Schools (DEIS) Programme, including both financial contributions and non-monetary supports. It starts with an overview of the baseline funding1 of primary and post-primary education in Ireland, as the background against which DEIS has developed. Then, it analyses the different components of DEIS and the rules for allocation of additional resources to schools.
Overall picture of education funding in Ireland
The Irish education budget has experienced some fluctuations in the past decades. Figure 3.1 shows the fluctuation of overall expenditure per student for primary, secondary (post-primary) and post-secondary non-tertiary education in particular, in Ireland compared to the Organisation for Economic Co-operation and Development (OECD) average over the past 30 years. After being below average for more than a decade, expenditure per student peaked in 2007 at more than 20% above the OECD average. This peak was short‑lived due to cutbacks following the financial crisis of 2008-2010. Since 2016, Ireland is catching up again, with a remaining gap of six percentage points in 2020 (the latest available year for comparison).
Figure 3.1. Trends in total expenditure per student in primary, secondary (post-primary) and post-secondary non-tertiary education
Index in PPP (purchasing power parity) dollars – OECD 2010 = 100
Source: Calculations based on OECD (2005[1]), Education at a Glance 2005: OECD Indicators, Table B1.4, https://doi.org/10.1787/eag-2005-en; OECD (2008[2]), Education at a Glance 2008: OECD Indicators, Table B1.5, https://doi.org/10.1787/eag-2008-en; OECD (2012[3]), Education at a Glance 2012: OECD Indicators, Table B1.5, https://doi.org/10.1787/eag-2012-en; OECD (2016[4]), Education at a Glance 2016: OECD Indicators, Table B1.5, https://doi.org/10.1787/eag-2016-en; OECD (2017[5]), Education at a Glance 2017: OECD Indicators, Table B1.3 and Table C1.3; and OECD (2023[6]), Education at a Glance 2023: OECD Indicators, Table C2.2, https://doi.org/10.1787/e13bef63-en.
Table 3.1 compares the spending on education in Ireland in 2020 as a percentage of gross domestic product (GDP) and expressed in absolute terms per student. Considering only government funding, the proportion of GDP spent on education in Ireland was 2.7%, against 4.3% on average across OECD countries. However, considering the expense as a share of GDP distorts the comparison because, in the Irish case, GDP figures are inflated by the presence of multinational firms (OECD, 2022[7]). As explained in Chapter 1, the Modified Gross National Income (GNI*)2 provides a measure of the size of the Irish economy specifically adjusted to lessen the impact of globalisation activities that disproportionately affect Irish economic aggregates. Education expenditure as a percentage of GNI* stood at 5.3% in 2019 and 5.8% in 2015 (Department of Education, n.d.[8]).
The absolute level of Irish expenditure per student in 2020 lied approximately 10% below the OECD average for primary education in 2020, and the corresponding ratios for secondary and tertiary were approximately 5%. Given that there are strong economies of scale in schools and that the average size of primary schools in particular is very small in Ireland (see Chapter 1), one can expect that schools still experience some financial stress despite the budgetary efforts since 2016.
Table 3.1. Expenditure on education
USD (United States Dollar) equivalent PPP, 2020
Source: OECD (2023[6]), Education at a Glance 2023: OECD Indicators, Table C1.1., Table C2.3, Table C2.3 and Table C4.1., https://doi.org/10.1787/e13bef63-en.
Between 2020 and 2024, the Irish government has enhanced the education budget by 20.2%, mainly as a response to the energy crisis and increased inflation (Department of Education, 2023[9]). However, a large part of this increase was announced as a one-off support (e.g. for COVID-19 pandemic supports and increased cost of living). It is yet unclear to what extent the “permanent” increase has outweighed rising costs and how it will impact the real spending per student at primary and secondary (post-primary) levels.
The relatively economical public funding in the past may explain why, until today, schools have recurred to private fundraising activities. The quest for voluntary contributions from parents is still a common practice in Irish schools for certain items and services. Despite recent legal and budgetary measures to reduce the pressure that is sometimes exerted on parents, the debate on free access to primary and post-primary education remains open (see Chapter 5).
Patterns of “baseline” school funding in Ireland
Before diving into the specificities of DEIS, it is worth summarising the main baseline funding mechanisms that are used to transfer resources to schools, irrespective of the social composition of their student population. A rough distinction can be made between capital expenditure, personnel and operating costs.
Capital investments (in school buildings) are carried out in programmes that are based on discretionary administrative decisions: this means that, among the schools applying for such funding, priorities are set according to their needs but also to available budgets and policy priorities rather than based on fixed formulas. In 2023, over 300 building projects were under construction, including new buildings as well as extensions or renovations of existing buildings (Department of Education, 2024[10]). Altogether, these capital investments account for approximately 9% of the overall budget devoted to primary, post-primary and post-secondary non-tertiary education (Department of Public Expenditure, NDP Delivery and Reform and Department of Finance, 2022[11]; 2023[12]).
Personnel costs (including pensions) make up 80% of the education budget (Department of Public Expenditure, NDP Delivery and Reform and Department of Finance, 2022[11]; 2023[12]). This category comprises teaching and non-teaching staff. Teaching staff are allocated to schools in terms of “teacher-hours” based on formulas that take into account the level (primary, post-primary), type (mainstream, special) and track (at post-primary level) of education as well as the number of students enrolled in each segment. Salaries are paid directly by the Department of Education (DoE) in order to iron out differences in pay related to qualifications and seniority across teachers and schools. Non-teaching staff include administrative personnel, special needs assistants for students with special educational needs and other non-teaching staff. More than 40 000 special needs teachers and special needs assistants are engaged to cater for students with special educational needs – as far as possible, in inclusive settings (Department of Education, 2024[10]).
Finally, operating expenses of schools are covered by a package of (cash) grants, of which the capitation grant is the most important one. As the term suggests, a fixed amount “per capita” (per student) is transferred to each school3. Enhanced capitation rates apply for students with special educational needs, and Traveller and Roma students. In addition, “earmarked” grants are provided for specific ancillary services, textbooks and other specific programmes (Department of Education, 2024[10]). The share of operating expenses in the overall budget for education is approximately 11% (Department of Public Expenditure, NDP Delivery and Reform and Department of Finance, 2022[11]; Department of Public Expenditure, NDP Delivery and Reform and Department of Finance, 2023[12]).
DEIS programme
The Delivering Equality of Opportunity In Schools (DEIS) programme is the Irish version of an equity funding strategy, i.e. a programme providing additional resourcing to schools that cater for a large proportion of socially disadvantaged students, with the aim of overcoming economic, social and cultural obstacles in their school careers, and equalising educational opportunities. Two-thirds of all European countries or regions have equity funding schemes in education (European Executive Agency for Education and Culture, Eurydice and Parveva, 2017[13]). In this section, the main features of the DEIS programme will be analysed through the lens of a theoretical-normative framework derived from the literature on equity funding systems (Demeuse et al., 2008[14]; Nicaise, Vandevoort and Verelst, 2024[15]). For the sake of simplicity, the analysis will focus on the present version of the DEIS programme (see Chapter 1 for an overview).
Objectives
As was mentioned in Chapter 1, the general ambition of the programme is to overcome “the impediments to education arising from social or economic disadvantage which prevent students from deriving appropriate benefit from education in schools” (Government of Ireland, 1998[16]). This corresponds to the mainstream definition of educational equity funding, pointing at socio-economic background as the main undesired determinant of inequalities in educational outcomes. In present-day OECD societies, the notion of “social disadvantage” is often extended to include socio-cultural obstacles that can be related to ethnic minority or immigrant background, which in the Irish case refers to Traveller and Roma communities as well as the growing (yet diverse) immigrant population.
The declared objectives in the first official documents – including the statement that “all students should have the opportunity to reach their potential” (Department of Education, 2024, p. 39[10]) – remained rather vague, as the term “potential” was not specified and is controversial in the scientific literature.4 However, the DEIS Plan 2017 adopted a comprehensive and ambitious set of quantitative targets in terms of (1) raising the proportion of students attaining minimum standards of literacy and numeracy, (2) narrowing the gaps between disadvantaged and other schools in terms of cognitive achievement, school career transitions (retention rates, progression into further and higher education) and well-being, and (3) teacher education, parental engagement and community links (Department of Education, 2017[17]).
Targeting educationally disadvantaged students
While it may seem intuitive to delineate the target group for equity funding as schools facing challenges in educational outcomes, such as low academic performance or high rates of repetition or dropout, providing supplementary resources to schools based solely on these criteria could yield unintended consequences. These consequences might include discouraging or penalising initiatives aimed at mitigating failure. Furthermore, additional funding directed towards improving outcomes may inadvertently lead to selective enrolment practices or incentivise schools to exclude their most academically challenged students (Hanushek, 1981[18]). To avoid such adverse effects, it is important to link the additional support to exogenous variables, i.e. student characteristics beyond the control of the school that are, nevertheless, strong predictors of poor educational outcomes. Socio-economic, ethnic and immigrant backgrounds are such exogenous characteristics (Box 3.1), and, as such, target schools in the DEIS programme have been identified on the basis of the social background of students. Social background characteristics are not only good predictors of educational disadvantage but also relatively stable across the school career and observable from a very early age, as social background mainly refers to the social position of parents. This has two further implications: firstly, preventive strategies can be implemented for socially disadvantaged children from early childhood onwards. Early intervention may avoid the accumulation of learning deficits and is, therefore, deemed more effective than remediation at later ages. Secondly, interventions targeting disparities in opportunities associated with social background, irrespective of actual achievement, will also benefit advantaged children. Indeed, from an “equal opportunities” perspective, these students also deserve additional support because it may further boost their chances to excel.
Box 3.1. Indicators of socio-economic background at individual and school level: lessons from other countries
Until 2017, England (United Kingdom) used eligibility for free school meals, along with a few other individual characteristics, as a proxy for disadvantage. The free meals indicator was abandoned when the National Audit Office found a significant proportion of non-take up, due to a range of motives, including stigma. The indicator was then replaced with eligibility for Universal Credit, a social protection scheme that assigns rights based on more reliable national register data.
In the Netherlands, individual student characteristics (parents’ level of education, mother’s country of birth, duration of residence in the country, postcode and entitlement to school grants) are collected into a central database administered by the Central Bureau of Statistics. Weights are assigned to all characteristics. Moreover, the overall weight at school level is multiplied with a coefficient for schools operating in concentration areas.
The Flemish Community of Belgium adopts a similar method. Most information is available from administrative register data, such as the mother’s level of education (registered by the Ministry of Education for all mothers who completed their education in Belgium) and the school grants (automatically assigned based on tax declarations). For recent immigrants, this is combined with self‑declared information on the mother’s education level and home language (registered at first enrolment). For secondary schools with a concentration of disadvantaged students (higher than 55%), the overall weight is multiplied by 1.5. An additional concentration multiplier of 1.11 is applied to schools in the Brussels Capital Region.
Source: Nicaise, Vandevoort and Verelst (2024[15]), “The effectiveness of equity funding of schools: a comparative analysis of Flanders, The Netherlands, France, Ireland and England” in The Routledge International Handbook of Equity and Inclusion in Education, https://doi.org/10.4324/9781003282921.
In the Irish case, the key indicator used is the Pobal Haase Pratschke Index (HP Index), described in Chapter 1. The HP Index is measured at individual level for all students in a school, based on their home address, but actually reflects the “socio‑economic capital” of their neighbourhood rather than their own social background. More precisely, an average value is measured for all households in a Small Area5 in which they live, based on a principal component analysis of data drawn from the population census, resulting in three components: demographic profile, social class composition and labour market situation (see Chapter 1 for more details). Using the Small Area averages from the census for each characteristic has the advantage that there are no gaps in the data, no administrative burden is imposed on schools for data collection purposes, and no intrusive questionnaires need to be filled upon registration in schools.
As was explained in Chapter 1, HP Index scores are standardised with a mean value of zero and a threshold of -10 points, equivalent to one standard deviation below the mean. Students within each school are identified as disadvantaged if their HP Index score is in the “bottom tail of the distribution”, more precisely at least 10 points below the population average. They receive a double weight if their index value is more than 20 points (or two standard deviations) below that average. “Borderline disadvantaged” students, with HP Index values between -7.5 and -10, are given a weight of 0.5; while all students above the -7.5 threshold get zero weight. Further corrections are made to include students self-identified as belonging to Traveller and Roma communities, and those residing in State-funded emergency homeless accommodation, in an International Protection Accommodation Service (IPAS) setting and Emergency Reception and Orientation Centre (EROC).
Next, the average weight at the school level is calculated. Schools are eligible for DEIS support either if their average student weight exceeds a given threshold, or if the proportion of students with an HP Index score at or below -10 exceeds a given threshold.
In the case of primary schools, the location of a school also plays a role: a distinction is made between urban and rural settings; within urban settings the most disadvantaged schools are labelled Urban Band 1 while other eligible schools are labelled Urban Band 2. Altogether (including the non-eligible schools), this makes four levels of support. At post-primary level, no distinction is made between urban bands and rural areas, resulting in a binary distinction between DEIS and non-DEIS.6
Concentration and coverage
The potential impact of equity funding on the target group depends on a combination of parameters such as (a) the share of the education budget spent on (additional) equity funding, (b) the relative size of the target group (expressed as a percentage of the corresponding school population), and (c) the absolute volume of the education budget. For any given volume of investment in equity funding, the potential impact of the additional resourcing at student level is inversely proportional with the size of the target population. Box 3.2 illustrates the argument with some comparative figures for five European education systems.
Box 3.2. Measures of financial impact of equity funding in five European education systems
Figure 3.2 shows that the quantitative coverage of equity resourcing varies between countries (e.g. 10% of the overall student population in the Netherlands versus 49% in the Flemish Community of Belgium).7 The desirability of a wide or narrow targeting remains a topic for debate. A narrower definition of the target population allows for more concentrated funding, with potentially more powerful effects on students’ outcomes. In contrast, proponents of a wide coverage argue that stigmatisation of target schools should be avoided, that equal educational opportunities are a remit for every school, and that even limited extra resources may encourage more “mixed schools” to put equity higher on their agenda.
Figure 3.2. Key parameters of financial impact of equity funding in selected systems
Note: Parameters in the figure relate to 2023 for Ireland (see Table 3.2) versus 2020 for other countries.
Sorted in descending order of the proportion of budget (panel A), proportion of the overall school population identified as target group (panel B) and average amount of additional equity funding per student (panel C).
Source: Adapted from Nicaise, Vandevoort and Verelst (2024[15]), “The effectiveness of equity funding of schools: a comparative analysis of Flanders, The Netherlands, France, Ireland and England” in The Routledge International Handbook of Equity and Inclusion in Education, https://doi.org/10.4324/9781003282921.
Ireland currently provides additional resources based on the DEIS status to 29.9% of its primary schools and 32.3% of its post-primary schools, catering for 28.0% of its students at primary and 25.5% at post‑primary level (Chapter 1). Note, however, that “students in DEIS schools” include a publicly unknown number of non-disadvantaged students, while, on the other hand, a significant proportion of disadvantaged students attend non-DEIS schools. In Annex 3.A, the OECD review team proposes own rough estimates of the latter proportion, based on the overall proportion of students in DEIS schools (cited above) on the one hand, and the relative proportions of disadvantaged students within DEIS and non-DEIS schools reported in recent studies (Delaney et al., 2023[19]; Nelis et al., 2021[20]) on the other hand. According to these calculations, in 2021 primary DEIS schools catered for 42.3% of all disadvantaged students, while the corresponding coverage rate at post-primary level in 2018 amounted to 39.3%.8 However, these estimates are based on samples and, more importantly, the DEIS scheme was extended in 2022.
The reason why DEIS does not cover all disadvantaged students is that only schools with a concentration of students identified as deprived by the HP Index receive DEIS support. DEIS aims to compensate for social disadvantage at school and geographical area level, not at individual level9. In a publication explaining the DEIS refined identification model since 2021, the Department of Education justifies this decision as follows (Department of Education, 2022, p. 13[21]):
“Disadvantage occurs throughout our communities and schools, and resources are provided to all schools to support all children to have the opportunity to reach their potential. The very nature of our school system is predicated on it being publicly funded. Universal provision, that all children can attend school for free, means that no child is excluded from accessing school due to financial disadvantage. The Department provides a wide range of supports to all schools, DEIS and non-DEIS, to support the inclusion of all students and address barriers to students achieving their potential.”
The DoE refers to examples of “universal” measures such as the lowered class size (23:1) in primary education, supporting staff and tools for students with special educational needs, schoolbook grants, psychological services, examination fee waivers for low-income students taking Junior Cycle examinations, and services provided by other departments, such as early childhood education and free school meals in primary schools. Furthermore, the Country Background Report, prepared by the DoE (2024[10]) for this Review, reports student support teams in post-primary schools, summer programmes (including a home‑based programme), Transition Year (which acts as a bridge between Junior and Senior Cycles), local community-based programmes in deprived areas, such as the North East Inner City (NEIC) multidisciplinary teams and “City Connects” projects, the Ireland Traveller and Roma Strategy (focusing mainly on school attendance and transition to third-level education), and specific resources (mainly additional teaching hours) for students whose first language is not English. A detailed discussion of all these programmes can be found in Chapter 5.
Nevertheless, the DEIS programme remains specific in targeting exclusively schools with high concentrations of educational disadvantage, not individual disadvantaged students. From an internationally comparative perspective, the general trend since the turn of the millennium has been to shift towards more individualised student-based algorithms, often still combined with school- and/or area-based criteria in equity funding schemes (Bernardo and Nicaise, 2000[22]; Demeuse et al., 2008[14]).10 See for example Box 3.3.
Box 3.3. Student-, school- and area-based characteristics used in the allocation of equity funding
In 2020, the OECD conducted a survey on equity funding schemes in 31 countries. Out of 28 countries that provided information on equity funding criteria at national level, 25 use student-based characteristics, 23 use school characteristics and 14 use indicators relating to the population by local areas (Figure 3.3). This means that most countries combine school- and student-based criteria in their allocation of equity funding, while half of them also use area-based criteria.
Figure 3.3. Student-, school- and area/population-based characteristics used in allocation of equity funding
Source: OECD (2021[23]), Education at a Glance 2021: OECD Indicators, Figure D6.4, https://doi.org/10.1787/b35a14e5-en.
Ireland predominantly uses school-based criteria: individual disadvantaged students do not generate additional funding for their school, unless there is a concentration of disadvantaged peers in the school. The “small area characteristics” linked to students’ place of residence do not play the same role as area-based features linked to the location of schools – except for the urban-rural distinction at primary level.
Source: OECD (2021[23]), Education at a Glance 2021: OECD Indicators, https://doi.org/10.1787/b35a14e5-en.
In response to advancements in modern e-government, equity funding schemes are progressively transitioning towards student-centred algorithms. Under this approach, each student is assigned a weight based on various factors, including their home language, neighbourhood characteristics, parents' educational attainment, income level, and other relevant variables. Student-based funding allows for a more fine-grained allocation of equity funding across schools as well as, in principle, a full coverage of the target population. The feasibility of this approach crucially depends on the possibility to integrate information on households across different government departments, in order to avoid the burden of data collection for schools. In some countries the government administration is even able (within the limits of privacy protection) to help schools identify the students that are socially disadvantaged, which may allow schools to target their support at those students even before challenges emerge. The main concern with this approach is the balance between the efficient targeting of support and the privacy protection of the students’ families.
The school-based algorithm (the cornerstone of the DEIS scheme) is based on the overall composition of the school population rather than the social background of individual students. Empirical research suggests that the concentration of disadvantage within schools is an even stronger obstacle to educational success than the socio-economic disadvantage of individual students (Burger, 2019[24]; Flannery, Gilleece and Clavel, 2023[25]; Gustafsson, Nilsen and Hansen, 2018[26]; OECD, 2018[27]). This finding justifies a system of equity resourcing that rises more than proportionately with the number of disadvantaged students in any given school. As such, the DEIS programme prioritises schools with high concentrations of disadvantage and, at primary level, differentiates between three levels of additional support.
The categorisation remains, nevertheless, subject to debate: section Challenges will examine some issues relating to the thresholds used and their unintended implications.
The volume of additional investments in DEIS
Describing the precise financial effects of the DEIS programme on schools, especially tracking trends, is difficult due to its expansion over time. Additionally, specific supports within DEIS receive co-financing from different government departments, and funding allocations have been shifted between the Department of Education and the Department of Children, Equality, Disability, Integration, and Youth (DCEDIY). Moreover, the Department of Social Protection provides funding to the School Meals Programme11 which was confined to DEIS schools and extended to non-DEIS primary schools in 2024. Besides, there is support from the Department of Further and Higher Education, Research, Innovation and Science (DFHERIS), from regional regeneration funds and local authorities for which the volume is unknown. Estimates must also be handled with care because the categories of government expenditure were aggregated differently between the reference years in Table 3.2.12
Given these warnings, Table 3.2 provides some estimates of key parameters relating to the budget allocation to DEIS in two reference years: 2016 (the year before the first extension of DEIS) and 2023 (the first year of full implementation of the second extension).
Table 3.2. Budget investment in DEIS (in current prices), 2016 and 2023
Source: Department of Public Expenditure, NDP Delivery and Reform and Department of Finance (2021[28]), Expenditure Report 2016: Part II Expenditure Allocations 2016 – 2018, https://www.gov.ie/pdf/?file=https://assets.gov.ie/181137/694a49bd-e585-4f24-9452-9b67c327e081.pdf#page=null (accessed 4 March 2024); Department of Public Expenditure, NDP Delivery and Reform and Department of Finance (2022[11]), Budget 2023 Expenditure Report: Part III - Estimates for Public Services, https://www.gov.ie/pdf/?file=https://assets.gov.ie/235731/6615fabb-2e04-4249-8ff6-e43bd7342db2.pdf#page=null (accessed 4 March 2024).
The overall (measurable) investment in DEIS is quite modest compared with the overall budget for pre‑primary, primary, post-primary and post-secondary non-tertiary education (3.0%). The share in the overall budget has increased slightly (0.4 percentage points) while the extension in terms of students covered was 50%. In terms of extra support per student, the estimated amount has increased 12.2% in nominal terms, but declined by 4.8% in real terms.13
Earmarking and conditioning of equity resourcing
A prominent feature of the DEIS programme is that most of the additional support to schools is provided in kind (and is thus earmarked). Earmarking might limit the freedom and flexibility of beneficiaries of support in using the funding. Yet, many decision makers at local level, including principals and teachers, may not always be sufficiently equipped to select the most effective and efficient instruments and strategies to combat educational disadvantage (OECD, 2023[29]). Teachers and principals tend to address educational disadvantage mainly from a pedagogical-didactical angle (such as through remediation classes, differentiation, specific language teaching tools) because that happens to be their strongest professional competence, while a more multidisciplinary approach is needed.14 Many would also opt for straightforward solutions such as reduced class size, a strategy that is probably effective but not the only option, nor a sufficient condition by itself for better educational outcomes (Mathis, 2016[30]). According to the Education Endowment Foundation (2021[31]) the main argument against reduced class size is its high cost. Recent research also suggests that enhancing the qualifications of teachers has a stronger impact on the performance of disadvantaged students than a reduction of the student/teacher ratio (Education Endowment Foundation, 2021[31]; OECD, 2018[27]). Therefore, earmarking part of the resources for continuing professional learning of the school staff – as is the case with DEIS support – appears to be justified. According to the European Executive Agency for Education and Culture, Eurydice and Parveva (2017[13]), the majority of equity funding systems in Europe (24 out of 40 examined cases) finance professional learning activities for teachers – mandatory or not (Box 3.4).
Box 3.4. Types of activities supported by equity funding schemes in Europe
Table 3.3 provides an overview of the types of activities supported by equity funding schemes in Europe as of 2015/2016. The relative importance of each type is reflected by the frequency of use rather than budget volumes allocated. Additional staff and continuing professional learning support appeared used most often, followed by special allowances for teachers or other staff and career advice services for students.
Table 3.3. Types of activities supported by equity funding in European countries (2015/16)
Note: In England (United Kingdom), Finland, the Flemish Community of Belgium, the Netherlands, Northern Ireland (United Kingdom) and Wales (United Kingdom), there is no information available due to school autonomy.
Source: Adapted from European Executive Agency for Education and Culture, Eurydice and Parveva (2017[13]), Structural indicators on achievement in basic skills in Europe – 2016, Figure 5.3, https://data.europa.eu/doi/10.2797/092314.
In the DEIS programme, the additional resources consist of two earmarked cash grants and a set of (in‑kind) services, co‑financed by various government departments, depending on the DEIS category (primary Urban Band 1 or 2 or Rural; or post-primary):
DEIS grant (to be used for the implementation of a school-based DEIS action plan);
Enhanced School Books Grant (now only for Senior Cycle at post-primary level);
Additional teaching hours to reduce class size;
Administrative Principal/Deputy Principal allocated at lower enrolment threshold;
Access to School Completion Programme;
Learning supports for literacy and numeracy;
Support for the professional continuous learning of teachers;
Planning supports (training for principals);
School Meals Programme (free provision of meals to children); and
Home School Community Liaison (HSCL) Coordinator.
The School Meals Programme is partly organised by local authorities (and voluntary organisations)15 and financed by the Department of Social Protection, while the HSCL Coordinators are funded by the DoE. Note also that the responsibility for early childhood education is shared between the Department of Children and the Department of Education. A more extensive description of all supports can be found in Chapter 5.
Figure 3.4 displays the distribution of the DEIS budget and related support (SCP and SMP) by type of support.
Figure 3.4. Distribution of DEIS-related support by type (2023)
Source: Calculations based on data provided by the Department of Education.
Approximately four-tenths (43%) of all additional resources granted to DEIS schools consist of additional staff (20.9% additional teachers in primary schools, 12.2% HSCL Coordinators, 2.8% vice-principals and administrative staff and 7.0% guidance staff in post-primary schools). Nearly half of the support (47.6%) consists of external services: mainly the School Completion Programme and the School Meals Programme (which by itself accounts for one third of the investment in DEIS (33.7%)), but also supporting didactical materials and training. The cash transfers to schools include the DEIS grant (7.3%) and the schoolbooks grant for post-primary schools16 (0.5%). The “other support” (1.6%) category relates to continuing professional learning, funding for the Early Start programme, the cost of increased National Educational Psychological Service (NEPS) supports, etc.
All in all, this means that schools have little discretion on how to use the additional supports. During interviews and school visits, some principals and academics advocated greater flexibility and choice in the allocation of resources at the local level. In contrast, the DoE claimed that earmarking is a deliberate policy option informed by what works best to boost equal opportunities.
The international experience also shows that the use of earmarking practices depends on the degree of centralisation versus school autonomy that characterises national education systems (Nicaise, Vandevoort and Verelst, 2024[15]; OECD, 2017[32]). In more liberal systems, the government will rather recur to alternative accountability rules based on outcomes such as the reduction of performance gaps between disadvantaged and other students. The challenge with this approach might be that equity strategies take a lot of time to generate visible effects, which makes it harder to intervene timely when schools fail to produce satisfactory effects.
Nevertheless, many countries that implemented equity funding have also set conditions relating to the use of the additional resources. Schools are obliged to draw up specific action plans for the achievement of more equal outcomes and to evaluate the effectiveness of their own actions. In England (United Kingdom), schools are even obliged to publish their targets and progress online, not just in average performance but also in equity of outcomes (Department for Education, 2023[33]). Within DEIS, schools are also expected to develop their own DEIS action plans, including SMART (Specific, Measurable, Achievable, Relevant and Time-bound) targets, and to evaluate their own achievements in terms of social equity of outcomes (Chapter 6). For this purpose, principals receive specific training, and they can get support from the DoE Inspectorate.
Irrespective of the monitoring and feedback by the Inspectorate, such planning and self-evaluation cycles are meant to boost the critical reflection of school teams on the effectiveness of their efforts; indirectly, this should lead to continuous self-monitoring, adaptation to changing needs and improvement of strategies for the schools.
Whereas a SMART operationalisation of the objectives was not available in earlier policy documents on DEIS, the DEIS Plan 2017 includes a table of clear and ambitious targets in relation to literacy and numeracy, school careers, teacher education, well‑being and parental engagement. Schools are expected to set SMART targets for themselves in their local DEIS plans, with regard to eight key themes (Department of Education, 2024[10]):
Attendance;
Retention;
Literacy;
Numeracy;
Well‑being;
Supporting educational transitions;
Partnership with parents and others; and
Academic outcomes (at the post-primary level only).
The next two sections will assess the strengths and challenges of the resourcing mechanisms of the DEIS programme as they have applied since 2022. Lessons from past arrangements will be referred to only if they help understand the latest policy reforms.
Strengths
The baseline resourcing of Irish schools is a relatively stable level playing field
It is useful to examine the DEIS resourcing mechanisms in conjunction with the baseline resourcing (i.e. resourcing that abstracts from the additional DEIS support) because experience in other countries shows that there are sometimes structural inequalities between schools that partly offset the effect of equity funding. Disadvantaged schools are often operating in older buildings, lack modern equipment, and have greater difficulties in attracting and retaining well-qualified teachers and principals, resulting in a so‑called “Matthew effect” (OECD, 2023[29]; Poesen-Vandeputte and Nicaise, 2014[34]).17 In the Irish case, this does not seem to be a major issue, at least at the primary level. Gilleece and Nelis (2023[35]) compared a set of indicators of school resourcing between DEIS Urban Band 1, Urban Band 2 and non‑DEIS primary schools, based on the school survey linked to the National Assessment of Mathematics and English Reading (NAMER) 2021 (Table 3.4). In all three categories of schools, the teachers had on average 10-11 years of experience, while the distribution by gender and temporary/permanent employment status was very similar. Whereas nearly all teachers completed the initial teacher education, those working in non-DEIS schools had more often acquired an additional qualification (44‑45%) than those teaching in DEIS Urban Band 1 (38-42%) and Urban Band 2 (30-34%) schools. However, these differences in additional qualifications of teachers do not seem to be correlated with their students’ performance on national tests. The percentage of teachers participating in continuing professional learning or teacher professional learning (TPL) did not differ significantly by DEIS status (Gilleece and Nelis, 2023[35]). Nor did the problems in teacher recruitment or retention differ between the three categories of schools (ibid.).
Table 3.4. Difficulties in attracting and retaining teachers by DEIS status
Percentage of students in schools whose principals reported experiencing difficulties in attracting and retaining teachers
Note: Schools are weighted by the number of sixth-class students.
Source: Gilleece and Nelis (2023[35]), Ireland’s 2021 National Assessments of Mathematics and English Reading: Exploring the home backgrounds, classrooms and schools of pupils in Urban DEIS schools, Table 4.5.
The authors did report some inequalities in relation to ICT materials and information books: whereas one in four students participating in NAMER 2021 did not have access to computers on average in primary schools, the problem was more common in DEIS schools, with some variation by type of DEIS school and class. A similar pattern emerges in regard to printed information books for students: two-fifths of the students in DEIS Urban Band 1 schools had no access to such books, against one in four students in non‑DEIS schools (Gilleece and Nelis, 2023[35]). Moreover, DEIS school buildings were much less open to the community outside class hours than those of non-DEIS schools (ibid.).
As mentioned in Chapter 4, PISA 2022 data suggest that disadvantaged post-primary schools faced greater teacher shortages, including inadequate qualifications and challenges related to absenteeism (which was also reported anecdotally in interviews).18
All in all, abstracting from the DEIS support, the designated primary schools do not seem to be systematically less well-resourced through government funding, but post-primary DEIS schools do suffer disproportionately from a lack of human resources. DEIS schools also get less support from parents and the local community: this is precisely one of the reasons for the additional DEIS investments. The survey carried out for the Society of St Vincent de Paul among parents and principals about voluntary contributions in post-primary schools showed that parents contribute less in DEIS schools (GrantThornton and SVP, 2023[36]). In regard to primary schools, Gilleece and Nelis (2023[35]) found that DEIS schools request fewer parental contributions (Table 3.5).
Table 3.5. Parental contributions requests by DEIS status
Percentage of students in schools whose principals reported requesting parental contributions
Note: Schools are weighted by the number of sixth-class students
Source: Gilleece and Nelis (2023[35]), Ireland’s 2021 National Assessments of Mathematics and English Reading: Exploring the home backgrounds, classrooms and schools of pupils in Urban DEIS schools, Table 4.2.
The use of the HP Index for the identification of needs enhances trust and support for DEIS
The DEIS identification mechanism has undergone significant improvements regarding the use of more overarching and objective measures. During the review visit and interviews with stakeholders, the OECD review team gained the impression that there was a lot of discontent about the allocation of DEIS resources in the past. While some respondents admitted that they were not aware of the criteria for the inclusion of schools in the DEIS programme, others thought that the criteria were arbitrary or a product of case‑by‑case bargaining. The adoption of the HP Index as a scientifically underpinned – though still imperfect – instrument was a significant step forward in identifying and prioritising the needs of schools for additional resources. The main advantages of the HP Index are the following (Department of Education, 2022[21]):
It is based on census data and, hence, covers the entire Irish population and territory. Nation‑wide statistics are reliable and stable for the derivation of indicators;
The Index reflects the various dimensions of social disadvantage: economic, social, cultural and human capital;
It is used by several Irish Government departments and services (Department of Health (DOH), Department of Rural and Community Development (DRCD), Department of Children, Equality, Disability, Integration and Youth (DCEDIY), Pobal and the Higher Education Authority (HEA)). It is applied in several other areas of social policy (local development, mobility, health, the residential property price index), which makes policies more transparent and fosters opportunities for synergies;
The updating of the Index is cheap as no specific data collection is required for its implementation. The burden of data collection is removed from schools and parents, and removes any incentive to manipulate data.
An additional feature of the algorithm used within DEIS is its focus on social disadvantage at the school level rather than the individual level. As was pointed out earlier, several studies have shown that the concentration of social disadvantage within residential or educationally localised pockets of disadvantage is far more harmful than a student’s low individual socio-economic background. This justifies the priority of a school‑level over an individual-level index of deprivation.
Earmarking and conditionality of DEIS resourcing ensures an adequate multidimensional tackling of disadvantage
A striking feature of the DEIS programme is that the bulk of additional resources granted to schools is earmarked and provided in kind. In many other countries, schools (or their intermediaries) get a supplement to their subsidies for operational costs and an additional block grant of teacher hours, which can be allocated to various teaching or non-teaching tasks (OECD, 2023[29]). In some cases, the additional teacher hours are also convertible into other types of professional services (European Commission, Directorate-General for Education, Youth, Sport and Culture, et al., 2020[37]). All these services are provided separately, though coordinated in packages, and schools cannot shift resources between them. As a result of this multi-pronged approach, there is a guarantee that schools tackle educational disadvantage using the whole range of instruments, rather than just reducing class size or engaging additional teaching staff. Moreover, schools are obliged to design their own DEIS plans with school-specific targets, and to evaluate their own progress (see Chapter 6). As indicated above, schools receive specific support for this planning and self-evaluation exercise, thus building capacity for strategic action. Overall, the strong steering by the DoE and the earmarking of DEIS supports may explain why a relatively modest budgetary effort proves to be effective in reducing social inequality in outcomes.
Challenges
The validity of the deprivation indicator could be further improved
Despite the progress achieved with the use of the HP Index as an indicator of social disadvantage, as argued in the previous section, there is room for further improvement of the algorithm for the allocation of DEIS resources.
First of all, the DoE could explore possibilities for balancing the content of the HP Index in the future. While stakeholders agree that it currently reflects the multidimensional reality of socio-economic disadvantage accurately, the limited inclusion of variables reflecting immigrant or ethnic minority status is surprising. It is partly remedied by the separate “weighting” of Traveller and Roma students, recently arrived refugees and students experiencing homelessness. In regard to immigrant newcomers in particular, there is a separate support programme for English as an additional language (EAL) as well as specific psychological advice and supports for teachers of newcomer students and those from culturally and linguistically diverse backgrounds offered by NEPS. In its publication on the refined DEIS identification model, the DoE (2022[21]) explicitly mentions that the inclusion of (other) immigration-related variables has been examined and discarded after consideration of evaluation research that suggests that the native‑immigrant performance gap becomes insignificant after controlling for socio-economic background and language spoken at home (see also OECD (2023[38])). Yet, the language variable may also capture the effects of “omitted third variables”, such as other cultural and psychological barriers that are typically experienced by immigrant students (and the schools receiving them).
Ireland remains an outlier in this regard. Among the 40 equity funding schemes examined by the European Executive Agency for Education and Culture, Eurydice and Parveva (2017[13]), only 12 do not use immigrant status as a criterion for extra support to schools and a similar picture emerges from the OECD survey of 31 equity funding schemes (OECD, 2023[29]). A potential argument for Ireland’s exceptional position is that its high-tech industries have also attracted highly qualified immigrants. However, this argument cannot be generalised to all immigrants. Therefore, further analysis of the possibilities for the development of a more fine-grained set of immigration-related indicators might be needed. Shifting immigrant patterns and the growing heterogeneity within the immigrant population should also be taken into account. Nelis et al. (2021[20]) observed persistent gaps linked to immigrant status and home language. It remains an open question whether language support and specific services for some refugee children are sufficient to cover all the needs of students with an immigrant background. Furthermore, in statistical analyses, home language may also capture other hidden effects such as the cultural distance between home and host country and ethnic segregation. Some countries, such as the Netherlands, have, therefore, replaced the language spoken at home with other criteria such as geographical region of origin and duration of stay in the host country (see also Eurydice (2023[39])). If such data are unavailable in the Irish census, they could be collected together with data on language spoken at home and plugged into the algorithm for identification of eligible students, as is already the case for Traveller and Roma students.
Moreover, the rising concern about mental health issues among young people may justify a search for indicators of youth mental health at the Small Area level as an additional input into the HP Index. All in all, a variant of the HP Index may capture the specific needs of socially disadvantaged students in a more accurate way, without losing coherence with other policies using the same basic tool. The OECD review team acknowledges, however, that mental health indicators, at Small Area or individual level, are currently unavailable at the population level, even in the international context.
The room for improvement of the HP Index is illustrated by the validation study carried out by Gilleece and McHugh (2022[40]). They found high correlations (0.7 to 0.8) between the HP Index scores and two other potential indicators of social disadvantage at the post-primary level, one of which was the percentage of students entitled to an examination fee waiver. 19 Yet, the latter appeared to be a marginally better predictor of school‑average reading performance than the more sophisticated HP Index.20
Apart from the content of the HP Index per se, the design of the algorithm transforming the Index into a classification tool is complex, with several stages of transformation, cutting-off and addition of data (see Chapter 1 for a detailed description of the algorithm). Although the focus on concentrations of disadvantage is beyond dispute, it remains unclear to what extent the resulting classification into four categories at the primary level (Urban Band 1, Urban Band 2, Rural and non-DEIS) and two at the post-primary level (DEIS and non-DEIS) is valid, what proportion of the theoretical target groups actually receives support, and to what extent the most disadvantaged students are receiving the strongest support.
Several interviewees during the OECD review mentioned that a relatively large number of “first-generation” beneficiary schools may no longer have the same high levels of concentrated disadvantage to merit inclusion in the DEIS programme. Although revising the eligibility of current beneficiaries is a sensitive political issue, continuing to provide levels of support beyond the identified need for such schools may result in horizontal equity issues that undermine the credibility of the selection criteria. Another consideration to be kept in mind is that the DEIS budget cannot be stretched indefinitely, which means that at some point, budget constraints may prevent positive decisions to cover needs (e.g. the claim for DEIS plus funding in areas of extreme deprivation (DCU Educational Disadvantage Centre, 2020[41]; DCU Educational Disadvantage Centre, 2022[42])) that are more pressing than those of historically included schools (vertical equity issues).21
The internal validity of the resulting classification deserves a closer inspection. The key question is not just to what extent the inequalities in outcomes between DEIS and non-DEIS schools are shrinking across time, but more importantly, to what extent social inequalities in outcomes between students are diminishing. To clarify this argument, it suffices to look at the research findings of Duggan et al. (2023[43]), who examined trends in inequalities in TIMSS science and mathematics performance in the fourth class over the period 2011-2019. The authors found a significant reduction of inequalities between DEIS and non-DEIS schools, but a significant increase in inequalities by socio‑economic background at the student level. Karakolidis et al. (2021[44]) had already found similar results when comparing NAMER and TIMSS trends for mathematics between 2009 and 2015. There are theoretically three potential (and possibly complementary) explanations for this phenomenon: (a) part of the DEIS schools and students were not really disadvantaged (misallocation of DEIS resources among schools through “false positive” selection into DEIS); (b) the additional resources allocated to DEIS schools benefitted the less disadvantaged students within these schools more (Matthew effect within DEIS schools); (c) the achievement gap between socially advantaged and disadvantaged students in non-DEIS schools continued to increase and this outweighed the beneficial effects of DEIS in the selected schools (coverage gap of the DEIS scheme or “false negative” exclusion from DEIS).
Note that these results do not cover the latest period22 and do not yet allow to assess the validity of the HP Index. However, they suggest that, for the observed periods, the alignment between some of the selected schools and the target group of socio-economically disadvantaged students might not be entirely precise, possibly due to some misclassification issues or incomplete coverage of the target population. Similar research on more recent data, including immigrant background along with socio-economic indicators, as well as a wider set of outcomes, could help clarify the extent of some of the highlighted issues.
Thresholds in the DEIS classification result in large differences in levels of support
As explained in the section Context and features as well as in Chapter 1, primary schools are classified into four categories (Urban Bands 1 and 2, Rural and non-DEIS) while at the post-primary level, a dichotomy is used between DEIS and non-DEIS schools. The categorisation of schools is based on a combination of thresholds for the degree of disadvantage at the school and area level. Such thresholds in the allocation of additional resources (a minimum level of average disadvantage or a minimum share of disadvantaged students at school level) can have a range of adverse effects. To begin with, thresholds draw a dividing line between schools that benefit from additional resources and schools that do not, with a risk of fuelling stigma among the former and/or envy among the latter. They may also reinforce segregation, with non‑targeted schools referring weaker students to those receiving extra support, or better-off parents opting out of the “stigmatised” schools.
From a mere resourcing point of view, dropping out of the selection, particularly at the post-primary level, can make a large difference. Vertical and horizontal equity issues may also arise between schools. For example:
A large school may not receive support while serving a greater absolute number of socially disadvantaged students if it does not exceed a given concentration threshold; and
Within a given DEIS category, differences in the severity of needs do not result in different levels of support, except for the DEIS grant and, partly, the allocation of HSCL Coordinators.23
There is an on-going debate about tapering of the DEIS support, i.e. using sliding scales of support instead of “in or out” thresholds. Schools that fall just outside the threshold for inclusion in the DEIS programme may also need additional support to foster social inclusion, despite the existence of mainstream instruments for that purpose as mentioned in the Context and features section. The demand of non‑DEIS and Rural DEIS schools for the HSCL Scheme, mentioned during the OECD review team’s interviews and school visits is just one example of such a much-advocated need. The relatively high proportion of socio‑economically disadvantaged students who are currently not covered adds to the volume of needs. The OECD estimates that in 2021, the proportion of disadvantaged students going to non-DEIS schools stood at 49% at the primary level and 58% at the post-primary level (see Annex 3.A). The programme has since been extended to 322 additional schools, therefore, this proportion has likely decreased.
At the other end of the distribution, the most disadvantaged schools – backed by the Educational Disadvantage Centre of Dublin City University (2020[41]; 2022[42]) – advocate for the introduction of a DEIS plus category, claiming that their needs are overwhelming and the current level of support appears to be insufficient. Principals report severe cases of intergenerational poverty, family breakups, trauma linked to state care placement, homelessness, mental health issues, violence, (parental) substance abuse and bullying, all symptoms of pockets of extreme marginalisation that are not included in the HP Index. They flag the need for different kinds of additional support (additional infrastructure, administrative support, in‑service training for teachers, further reduction of class size) but mainly psychologists for trauma treatment, as well as preventive mental health care (see also Fleming and Harford (2021[45])).
The risk of misclassification is inherent in any typology using thresholds. This risk is illustrated in the study of the HP Index by Gilleece and McHugh (2022[40]). The authors cross‑tabulated the quintile distributions of schools ranked by the HP Index scores with other indicators of disadvantage. For example, only 17 out of 31 schools (55%) in the bottom quintile of their (PISA) sample were classified in the same quintile using the HP Index and the PISA indicator of socio‑economic status. Taking the two bottom quintiles together, the degree of overlap between both criteria increased to 67%. This might suggest that “in‑or‑out” cut-off lines, such as those used in DEIS may involve a non-negligible risk of misclassification, though further analysis is needed.24
Voluntary contributions continue to present a challenge to some parents
The issue of voluntary contributions (as well as other school-related costs charged to parents) has not been extensively researched yet and remains a subject of controversy. Based on the first wave of the cohort study Growing Up in Ireland, collected in 2017/18 for a representative sample of 7 563 9-year-old children, the ESRI published some key findings related to parental contributions (Trinity College Dublin, 2018[46]):
“59% of families of 9-year-olds paid a voluntary contribution to their child’s primary school; 7% were asked for a contribution but did not pay; 34% were not asked for a contribution. Overall, 23% of families paid less than €50; 23% paid between €50 and €99 and 12% paid €100 or more.”
Yet, given the voluntary nature of the contributions, there were differences by level of income (Trinity College Dublin, 2018[46]):
“24% of families in the highest income group paid a voluntary contribution of €100 or more, compared to 5% of families in the lowest income group.”
More recent figures are available from the school principal questionnaire of NAMER 2021, which asked, among other things, whether the school asked parents for a school contribution (Gilleece and Nelis, 2023[35]). The findings are in line with the Growing Up in Ireland survey. They also show that DEIS schools tend to be far more cautious in this matter. Whereas 66% of students in the urban non-DEIS had principals who replied that they asked for voluntary contributions, only 21% of DEIS Urban Band 1 and 30% of Urban Band 2 students’ principals did so. This suggests that the DEIS support (in particular, the grants for operating costs and the School Books Grant Scheme) helps these schools reduce the financial pressure on parents. The flip side of this advantage is that it may reinforce segregation, by attracting more socio‑economically disadvantaged students into DEIS schools. Note also that not asking for voluntary contributions does not necessarily mean that there are no paying services offered by the schools. Most Irish schools, even at the primary level, continue charging parents for various kinds of items other than the usual parental expenses relating to school uniforms, sports outfits, meals (at the post-primary level) or after-school supervised study (for example, rent of lockers, learning materials, excursions, maintenance, and secretarial costs) (GrantThornton and SVP, 2023[36]). Chapter 5 discusses the findings of these surveys in greater detail.
The Irish law (Section 64 of the Admissions to School Act 2018) already prohibits requests for voluntary contributions, or any kind of other contribution, when parents apply for enrolment or re-enrolment in a school (Government of Ireland, 2018[47]). The DoE has also sought to legislate for increased consultation between school communities in regard to the planning, policies and activities in schools. The Students and Parents Charter Bill was published in September 2019 and has been passed by the Seanad (the upper house of the Irish parliament). The Bill passed Second Stage in Dáil Éireann (lower house of the Irish parliament) in July 2021 and is awaiting an order for Committee Stage (Houses of the Oireachtas, 2019[48]). This Bill provides for the Minister to issue guidelines and for all schools to prepare a Charter, following consultation with parents and students. The overall aim of the Bill is to improve the level of engagement between the school and community by inviting feedback, comments and observations from the community on a range of issues, including voluntary contributions and how they are spent, as well as complaint procedures.
There is also a separate private members bill, the Education (Voluntary Contributions) Bill, which was introduced in parliament in 2021. This bill aims to mandate the Minister of Education to regulate (not abolish) voluntary contributions (Houses of the Oireachtas, 2022[49]). This bill would mean schools would be obliged to inform parents that contributions are not mandatory, and that the enrolment of their children cannot be made dependent on the payment of voluntary contributions.
In the meantime, the (universal) school capitation grant for operating costs of primary and post-primary schools has been increased, first to compensate for the peak in energy prices in 2021‑22, and next to roll out the free books scheme into all primary schools and Junior Cycle at the post‑primary level, including other books, audio and digital materials. Parents/Guardians will no longer be required to make any contribution towards the cost of schoolbooks (except for Senior Cycle), including the cost of core classroom resources, in the school year 2024/25. Schools will be required to communicate with parents/guardians in relation to the scope of the scheme in their school. The extension of the School Meals Programme, financed by the Department of Social Protection, and prioritised towards schools in the DEIS programme, is also worth mentioning. Until recently, it was targeted at DEIS primary schools only, but as of April 2024 it was made available to primary students outside of DEIS schools (Department of Education, 2024[10]). At the post-primary level, exam fees have not been requested for 2023 and 2024. There is an income-based exemption in place, which means children from lower-income families do not need to pay. Additional funding has been channelled to the School Transport Scheme, which has allowed for reduced school transport charges for families who use the scheme. Eligible children with a medical card and children with special educational needs are exempt from the annual ticket charge. Civil society organisations continue to demand further measures. For example, in May 2023 St Vincent de Paul published a report on voluntary contributions in post‑primary schools, based on a (non-representative) online survey of 1 447 parents, along with a survey of 19 schools and six interviews with principals (GrantThornton and SVP, 2023[36]).25 Almost all (86%) of the parents declared that their school had asked for such contributions: they ranged between EUR 30 and EUR 550 (euros), with an average of EUR 140. Moreover, 80% claimed that the school had not clearly communicated that the charges were optional. Many parents also complained about other school-related costs that were not tagged as voluntary contributions, such as classroom resources, school clubs, fees for Transition Year and Junior Cycle and Leaving Certificate examinations. The main recommendation of the report is a re‑evaluation of the capitation grant to schools, which was eroded due to austerity measures in the 2010-2020 period. According to the authors of the online survey report, a 33% uprating would be needed to compensate for the loss of value in that decade only (ibid.).26
Policy recommendations
Further strengthen access to free education
The principle of progressive universalism in social policies implies that targeted support for disadvantaged groups should always build on a strong basis of universal rights. The right to free education (at least up to the lower secondary level) is enshrined in the Universal Declaration of Human Rights as well as the United Nations’ Convention on the Rights of the Child. At the European Union (EU) level, it is included in the European Pillar of Social Rights and the Child Guarantee agenda. As mentioned in Chapter 2, Ireland has adhered to all these conventions and joint commitments. Free access to post-primary education (e.g. Transition Year in Ireland) remains a matter for debate, depending on the legal duration of compulsory education.
The increased baseline funding for the operation expenses of schools set out in the previous section can be seen as part of the “universal” counterpart of the selective DEIS funding (which includes the DEIS grant27) (Houses of the Oireachtas, 2022[49]). Under a regime of free school choice, a sound balance between both pillars (universal and targeted) is essential to prevent social segregation among schools, with a concentration of students from low-income families in strongly subsidised disadvantaged schools, while other schools would remain accessible only to middle- and upper-class children due to low public resourcing supplemented with high voluntary contributions. While recognising the importance of consultation, information and negotiation of financial contributions with parents and students as well as all steps implemented to reduce costs for parents in recent years, the OECD review team suggests taking further financial and monitoring measures to guarantee free basic education for all:
1. Reviewing, by level of education, the categories of costs that should not be charged to parents in accordance with international conventions, such as examination fees (currently temporarily suspended), administrative costs, and rent of lockers (more generally, all goods and services that are directly linked to the achievement of educational attainment targets);
2. Monitoring periodically the actual school-related expenses by parents through surveys and, if needed, adjusting the capitation grant to a level that allows schools to fully cover at least the costs referred to in point 1.; and
3. Advising schools on how to minimise charges and how to manage voluntary contributions correctly; and, if needed, sanctioning unlawful pressure exercised by schools.
Continue refining and validating the indicator(s) of social disadvantage underpinning the targeting of DEIS resources
The analysis of the HP Index in the previous sections revealed both strengths and shortcomings of this indicator. Strengths include the multidimensionality and relatively cheap derivation of the HP Index, its privacy‑proofness and the absence of administrative burden on schools, as well as its use in several policy areas. More recently, other individual characteristics were combined with the HP Index for Traveller and Roma students, homeless and refugee children. Language barriers are also being addressed through specific language support for non-native speakers.
However, the HP Index remains mainly socio-economic and only indirectly and partly captures the disadvantages linked to (a) psychological and socio-emotional well‑being, and (b) cultural barriers and immigrant background. Yet, there is evidence in the international literature about the obstacles linked to mental health issues and students with an immigrant background.28 Thus, the target effectiveness and efficiency of the DEIS programme could be improved by including additional dimensions of social disadvantage.
Further research would also be helpful to assess (and possibly improve) the scientific validity of the HP Index as a key indicator. The finding by Gilleece and McHugh (2022[40]) that a single variable (the percentage of students entitled to an exam fee waiver for the Junior Cycle examination) predicted average reading performance at school level better than HP Index values underscores the need for on-going and detailed consideration of the most appropriate variables to use in identifying DEIS schools. Such validation studies should be repeated at primary as well as post-primary levels, with a range of relevant dependent variables (e.g. cognitive and non-cognitive outcomes, early leaving from education and training, transition to tertiary education29) to sketch a comprehensive picture of the predictive power of the (present or amended) HP Index and alternative indicators. Chapter 6 discusses some options relating to alternative data sources that may supplement (or, in the future, replace) the HP Index with more direct information on the social profile of students.
Examine scenarios to attenuate the adverse effects of key thresholds in the DEIS classification algorithm
The present algorithm that results in the classification of schools uses mainly two key characteristics of the student body at the school level, derived from the HP Index: a weighted “severity” of disadvantage and its “extent” (the proportion of students considered disadvantaged). Cut-off thresholds are used at the individual level to measure degrees of disadvantage (with low-scoring students getting a zero weight and the most severely disadvantaged getting a double weight), and subsequently at the school level and area‑level to classify schools. The combination of thresholds at individual, school and area levels makes the algorithm complex and less predictable, which results in frustration among some schools that are not part of the programme. Above all, the “all-or-nothing” threshold at the post-primary level makes it more difficult to cut back on schools that no longer qualify for support because their scores of disadvantage are insufficient (or have fallen). The addition of the “borderline threshold”30 to include students with scores just below the initial threshold was a way to accommodate for the pressure from schools at the margins of the DEIS programme. However, this extension did not alter the “in‑or‑out” thresholds at school and area levels. As was argued in the section on Challenges, such thresholds involve a non‑negligible risk of misclassification and may unintendedly create inequalities in the treatment of similar schools.
It is, therefore, worth examining if smoother algorithms could be designed. Several options can be considered based on the same principle. Using the complete distributions of the HP Index at the individual level and the concentration at the school level:
At the individual level, a continuous value of the degree of disadvantage or an interval variable using more than four values (0, 0.5, 1 and 2) measuring the severity of disadvantage;
At the school level, a continuous value (ranging from zero to one, for example) reflecting the degree of concentration of disadvantage, based on the proportion of students exceeding a given threshold of disadvantage in the school, or an interval variable using more than three thresholds at smaller intervals (also at the post-primary level).
The school weights could be calculated using some kind of multiplicative combination of the two criteria, with every student’s score increasing as the proportion of disadvantaged peers increases. For schools with a low concentration, a limited DEIS supplement could apply so that fewer schools would feel excluded, whereas schools with a high concentration of very severe disadvantage (the “DEIS plus candidates”) would be more heavily supported.
The resulting algorithm would be “smoother” and more “logical”. It would avoid the “stacking” of cut-off lines in successive steps at individual and school levels. The classification into Urban Band 1, Urban Band 2 and Rural could even become redundant if a wider variation of scores is used at the school level. The “penalties” of misclassification would also be reduced because of the smoother distribution of resources.
The relatively rigid division between earmarked sub-packages could make up a potential obstacle to the implementation of a “thresholdless” resourcing scheme under the DEIS programme. In principle, different categories of staff could be allocated on a part-time basis (expressed in hours per month, for example). However, this would increase the costs of administration and transport between schools. An intermediate solution could, therefore, consist of a formula with multiple smaller thresholds derived from the “smooth” algorithm. The city of Hamburg (Germany) offers an instructive example of a school-level social index with six levels (Box 3.5).
Of course, all these suggestions would necessitate an ex-ante study, including simulations of the redistributive effects between schools in terms of support received. The impact on “winners” and “losers” should be carefully examined before the implementation of any alternative algorithm, taking into account that redistribution of resources is less painful under extending budgets than under “zero-sum games”.
Note also that the “tapering” of the DEIS classification and the enrichment of the HP indicator (and its add‑ons) can be implemented separately, in subsequent stages.
Box 3.5. Index-based equity funding for schools in Hamburg (Germany)
In 1996, the German city state of Hamburg introduced a Sozialindex (Social Index) for all public schools to distribute additional staff and funding to schools. The Social Index is calculated using eight indicators based on which schools are assigned to one of six Levels (Level 1 indicating disadvantaged student populations and Level 6 student populations from an advantaged socio-economic background):
The proportion of students with non-German family languages;
The proportion of students with special education needs;
The proportion of students receiving educational assistance (Bildungs- und Teilhabepaket);
The proportion of school leavers with general higher education entrance qualifications in students’ areas of residence;
The proportion of under-15-year-olds receiving social benefits in the students’ area of residence;
The proportion of eligible people receiving educational assistance in the students’ areas of residence;
The proportion of 15-65-year-olds who are unemployed in students’ areas of residence; and
Voter turnout in students’ areas of residence.
The eight indicators are merged from different data collections, and data from the last three years is collated to mitigate the effects of annual fluctuations. The Social Index is updated every five years.
Schools at Level 1 and 2 receive more staff to form smaller classes. Primary schools at Levels 1 and 2 receive more funding and staff for special needs education. In lower secondary schools, funding and staff are allocated on a per-student basis and schools at Levels 1 and 2 receive more funding per student than those at Levels 3-6. The lower the Social Index of a school, the more staff hours they receive for language support and all-day care.
The Social Index is also used to draw comparisons between schools in comparative assessments and to form comparison groups in the context of educational reporting. This serves to prevent schools with more difficult circumstances from being compared with more advantaged schools without considering the social context in which they operate.
Source: Schulte, Hartig and Pietsch (2014[50]), “Der Sozialindex für Hamburger Schulen [The social index for schools in Hamburg]” in Grundlagen für eine daten- und theoriegestützte Schulentwicklung [Foundations for data- and theory-supported school development] and OECD (2022[51]), Quality and Equity of Schooling in the German-speaking Community of Belgium, https://doi.org/10.1787/9a6b6f3a-en.
Extend partial additional support to all students defined as disadvantaged
In the introductory section, it was acknowledged that all schools have access to a limited array of specific resources and instruments to cater to socially disadvantaged students. It was also acknowledged that priority resourcing of schools dealing with a concentration of disadvantage is justified, as concentration per se has a negative effect on students’ learning opportunities. Nevertheless, there are several arguments to invest more in all disadvantaged students, irrespective of their school’s degree of concentration:
Individual socio-economic and cultural minority background also affects the opportunities of students, irrespective of concentration effects (Burger, 2019[24]; Gustafsson, Nilsen and Hansen, 2018[26]; OECD, 2018[27]).
From an ethical point of view, all disadvantaged students have a right to additional support if the policy objective is to achieve equal opportunities. This individual entitlement would also mean that students making transitions between schools or levels of education would always continue to benefit from some extra support, irrespective of the location or composition of their new school.
In the Irish case, in particular, the present focus on concentration schools appears to cover less than half of the disadvantaged student population – although the precise impact of the extension of DEIS in 2022 is not yet known (see Annex 3.A). Moreover, two studies suggest that the reduced gap between DEIS and non-DEIS schools does not (yet) seem to coincide with a reduced gap between disadvantaged and non-disadvantaged students (Duggan et al., 2023[43]; Karakolidis et al., 2021[44]). This could suggest that, during the observed periods, the alignment between some of the selected schools and the target group of socio-economically disadvantaged students might not be entirely precise. Nonetheless, as mentioned earlier, further work is needed to draw any strong conclusions.
Covering all disadvantaged students reduces the harmful effects of potential misclassification of schools, compared with a scheme where (rather artificial) cut-off lines determine eligibility for a full package of additional resources.
The current combination of measures and resources, besides DEIS, could be integrated into a more coherent overall framework to address social disadvantage.
Full coverage of the target group(s), irrespective of their geographical environment, also means that all schools are accountable for the achievement of equal opportunities. The shared responsibility of all schools is important, not only vis-à-vis the disadvantaged students themselves, but also from a political point of view. If all schools are (at least potentially) eligible for some DEIS funding, however limited the extra funding may be, the support for DEIS at the grassroots level will grow. Even schools with relatively low concentrations of disadvantaged students may be more inclined to accept applicants from disadvantaged backgrounds and maybe reach out to them as a way to take up social responsibility. Stigmatisation of selected DEIS schools, envy effects and strategic behaviour that could result in enhanced segregation would be avoided.
The combination of full coverage of the target groups and enhanced funding of concentration schools would consistently reflect the principle of progressive universalism, which is widely advocated in social policies. Progressive universalism is an ethical principle that combines quality services for all, based on fundamental rights, with priority investment in vulnerable groups due to their higher level of needs.
Prepare the periodic updating of the indicators of social disadvantage to develop a more dynamic resource allocation model
The revision of the criteria for inclusion of schools in the DEIS programme since 2017 has led to additional schools receiving support. During the review process, the OECD review team learned that dozens of schools that were already included earlier but do not currently have the same level of disadvantage based on the HP Index have continued to receive full support. A revision of the support level spread over several years would be recommended to allow schools to adjust their internal resource allocation.
Similarly, the newly available data from the 2022 census has led to an update of the HP Index. The social profile of schools and neighbourhoods may have changed due to the mobility of residents, recent immigration waves, urban regeneration programmes, changing labour market conditions, etc., which might necessitate a re‑targeting of support. The allocation of DEIS resources might need to be adjusted on a periodic basis, taking shifts in the social profile of schools into account, to prevent new inequities from arising. As such, this raises the sensitive issue of cutbacks, and in some cases, phasing-out, of support for schools whose needs have diminished. Cutbacks should not be seen as a “penalty on good performance” because the allocation of DEIS support occurs exclusively on the basis of exogenous measures of need. Nor should phasing out some schools result in layoffs if services are gradually reallocated to other schools. The combination of a “smoother” algorithm, advocated in the previous recommendation, combined with transition periods spread over several years could make the reallocation of resources more acceptable and prepare the ground for future adjustments.
References
[22] Bernardo, J. and I. Nicaise (2000), Educational priority policies, Bristol University Press.
[24] Burger, K. (2019), “The socio-spatial dimension of educational inequality: A comparative European analysis”, Studies in Educational Evaluation, Vol. 62, pp. 171-186, https://doi.org/10.1016/j.stueduc.2019.03.009.
[54] Connor, H. and S. Guerin (2019), Key Insights into the Socio-Economic Background of FET Learners 2019, https://www.solas.ie/f/70398/x/45b5072653/solas-key-insights-report.pdf (accessed on 6 May 2024).
[42] DCU Educational Disadvantage Centre (2022), Opening Statement to the Oireachtas Joint Committee on Education, Further and Higher Education, Research, Innovation and Science: Mental Health Supports in Schools and Tertiary Education, https://data.oireachtas.ie/ie/oireachtas/committee/dail/33/joint_committee_on_education_further_and_higher_education_research_innovation_and_science/submissions/2022/2022-11-08_opening-statement-dr-paul-downes-head-of-educational-disadvantage-centre-dcu_e (accessed on 29 February 2024).
[41] DCU Educational Disadvantage Centre (2020), Opening Statement Submission to the Oireachtas Joint Committee on Education, Further and Higher Education, Research, Innovation and Science, https://data.oireachtas.ie/ie/oireachtas/committee/dail/33/joint_committee_on_education_further_and_higher_education_research_innovation_and_science/submissions/2022/2022-11-08_opening-statement-dr-paul-downes-head-of-educational-disadvantage-centre-dcu_e (accessed on 29 February 2024).
[19] Delaney, E. et al. (2023), PIRLS 2021: Reading results for Ireland, Educational Research Centre.
[14] Demeuse, M. et al. (eds.) (2008), Les politiques d’éducation prioritaire en Europe [Priority education policies in Europe], INRP - Institut national de recherche pédagogique.
[33] Department for Education (2023), What maintained schools must publish online, https://www.gov.uk/guidance/what-maintained-schools-must-publish-online#pupil-premium-and-recovery-premium (accessed on 4 March 2024).
[10] Department of Education (2024), OECD Review of resourcing schools to address educational disadvantage: Country Background Report Ireland, Department for Education, Dublin, https://s3-eu-west-1.amazonaws.com/govieassets/296017/4d1ac422-5475-470e-b910-4c80a83c43bc.pdf.
[9] Department of Education (2023), Main Features of Budget 2024: Department of Education, https://assets.gov.ie/273435/c77f2daa-d3f5-468c-a894-bb8931dd513b.pdf (accessed on 29 February 2024).
[21] Department of Education (2022), The Refined DEIS identification model, https://assets.gov.ie/220043/d6b98002-a904-427f-b48a-0fa0af756ea7.pdf (accessed on 29 February 2024).
[17] Department of Education (2017), DEIS Plan 2017, https://www.gov.ie/pdf/?file=https://assets.gov.ie/24451/ba1553e873864a559266d344b4c78660.pdf#page=null (accessed on 30 November 2023).
[8] Department of Education (n.d.), Note on use of GDP and modified GNI as a measurement of public expenditure on education in Ireland, https://www.gov.ie/pdf/?file=https://assets.gov.ie/236374/ec8c69cd-eaf0-40a2-aac7-704baf743a49.pdf#page=null (accessed on 22 April 2024).
[12] Department of Public Expenditure, NDP Delivery and Reform and Department of Finance (2023), Budget 2024 Expenditure Report: Part III - Estimates for Public Services, https://www.gov.ie/pdf/?file=https://assets.gov.ie/273325/9fddc8bd-0114-46db-9514-cf5ddd919690.pdf#page=null (accessed on 4 March 2024).
[11] Department of Public Expenditure, NDP Delivery and Reform and Department of Finance (2022), Budget 2023 Expenditure Report: Part III - Estimates for Public Services, https://www.gov.ie/pdf/?file=https://assets.gov.ie/235731/6615fabb-2e04-4249-8ff6-e43bd7342db2.pdf#page=null (accessed on 4 March 2024).
[28] Department of Public Expenditure, NDP Delivery and Reform and Department of Finance (2021), Expenditure Report 2016: Part II Expenditure Allocations 2016 - 2018, https://www.gov.ie/pdf/?file=https://assets.gov.ie/181137/694a49bd-e585-4f24-9452-9b67c327e081.pdf#page=null (accessed on 4 March 2024).
[52] Donohue, B. et al. (2023), Education in a Dynamic World: the performance of students in Ireland in PISA 2022, Educational Research Centre.
[15] Downes, P. et al. (eds.) (2024), The effectiveness of equity funding of schools: a comparative analysis of Flanders, The Netherlands, France, Ireland and England, Routledge, https://doi.org/10.4324/9781003282921.
[43] Duggan, A. et al. (2023), “Trends in educational inequalities in Ireland’s primary schools: an analysis based on TIMSS data (2011–2019)”, Large-scale Assessments in Education, Vol. 11/1, https://doi.org/10.1186/s40536-023-00188-2.
[31] Education Endowment Foundation (2021), Reducing class size, https://educationendowmentfoundation.org.uk/education-evidence/teaching-learning-toolkit/reducing-class-size (accessed on 29 February 2024).
[53] Elford, G. (2015), “Social Class, Merit and Equality of Opportunity in Education”, Res Publica, Vol. 22/3, pp. 267-284, https://doi.org/10.1007/s11158-015-9280-3.
[37] European Commission, Directorate-General for Education, Youth, Sport and Culture, et al. (2020), The governance of equity funding schemes for disadvantaged schools – Lessons from national case studies – Analytical report, Publications Office, https://doi.org/10.2766/989607.
[13] European Executive Agency for Education and Culture, Eurydice and Parveva (2017), Structural indicators on achievement in basic skills in Europe – 2016, Education, Audiovisual and Culture Executive Agency, https://doi.org/10.2797/092314.
[39] Eurydice (2023), Netherlands, https://eurydice.eacea.ec.europa.eu/national-education-systems/netherlands/early-childhood-and-school-education-funding (accessed on 29 April 2024).
[50] Fickermann, D. and N. Maritzen (eds.) (2014), Der Sozialindex für Hamburger Schulen [The social index for schools in Hamburg], HANSE – Hamburger Schriften zur Qualität im Bildungswesen.
[57] Finch, H., M. Hernández Finch and B. Avery (2021), “The Impact of National and School Contextual Factors on the Academic Performance of Immigrant Students”, Frontiers in Education, Vol. 6, https://doi.org/10.3389/feduc.2021.793790.
[25] Flannery, D., L. Gilleece and J. Clavel (2023), “School socio-economic context and student achievement in Ireland: an unconditional quantile regression analysis using PISA 2018 data”, Large-scale Assessments in Education, Vol. 11/1, https://doi.org/10.1186/s40536-023-00171-x.
[45] Fleming, B. and J. Harford (2021), “The DEIS programme as a policy aimed at combating educational disadvantage: fit for purpose?”, Irish Educational Studies, Vol. 42/3, pp. 381-399, https://doi.org/10.1080/03323315.2021.1964568.
[40] Gilleece, L. and G. McHugh (2022), “Validating school-based measures of educational disadvantage in Ireland”, Education Policy Analysis Archives, Vol. 30, https://doi.org/10.14507/epaa.30.7245.
[35] Gilleece, L. and S. Nelis (2023), Ireland’s 2021 National Assessments of Mathematics and English Reading: Exploring the home backgrounds, classrooms and schools of pupils in Urban DEIS schools, Educational Research Centre.
[47] Government of Ireland (2018), Education (Admission to Schools) Act 2018, https://www.irishstatutebook.ie/eli/2018/act/14/enacted/en/html (accessed on 3 June 2024).
[16] Government of Ireland (1998), Education Act, 1998, https://www.irishstatutebook.ie/eli/1998/act/51/enacted/en/html (accessed on 29 November 2023).
[36] GrantThornton and SVP (2023), The role of voluntary contributions in post-primary schools in Ireland, https://www.svp.ie/wp-content/uploads/2023/05/Grant-Thornton-The-role-of-voluntary-contributions-in-post-primary-schools-in-Ireland-St.-Vincent-de-Paul-2023.pdf (accessed on 29 February 2024).
[26] Gustafsson, J., T. Nilsen and K. Hansen (2018), “School characteristics moderating the relation between student socio-economic status and mathematics achievement in grade 8. Evidence from 50 countries in TIMSS 2011”, Studies in Educational Evaluation, Vol. 57, pp. 16-30, https://doi.org/10.1016/j.stueduc.2016.09.004.
[18] Hanushek, E. (1981), “Throwing Money at Schools”, Journal of Policy Analysis and Management, Vol. 1/1, p. 19, https://doi.org/10.2307/3324107.
[55] Higher Education Authority (2019), A Spatial & Socio-Economic Profile of Higher Education Institutions in Ireland, https://hea.ie/assets/uploads/2019/10/Higher-Education-Spatial-Socio-Economic-Profile-Oct-2019.pdf (accessed on 6 May 2024).
[49] Houses of the Oireachtas (2022), Education (Voluntary Contributions) Bill 2021, https://www.oireachtas.ie/en/bills/bill/2021/113/ (accessed on 4 March 2024).
[48] Houses of the Oireachtas (2019), Education (Student and Parent Charter) Bill 2019, https://www.oireachtas.ie/en/bills/bill/2019/67/ (accessed on 3 June 2024).
[44] Karakolidis, A. et al. (2021), “Examining educational inequalities: insights in the context of improved mathematics performance on national and international assessments at primary level in Ireland”, Large-scale Assessments in Education, Vol. 9/1, https://doi.org/10.1186/s40536-021-00098-1.
[30] Mathis, W. (2016), The Effectiveness of Class Size Reduction, https://nepc.colorado.edu/sites/default/files/publications/Mathis%20RBOPM-9%20Class%20Size.pdf (accessed on 29 February 2024).
[56] National Association of School Psychologists (2020), The Relationship Between Mental Health and Academic Achievement, NASP Research Summaries, https://www.nasponline.org/ (accessed on 7 May 2024).
[20] Nelis, S. et al. (2021), Beyond achievement: home, school and wellbeing findings from PISA 2018 for students in DEIS and non-DEIS schools, Educational Research Centre.
[6] OECD (2023), Education at a Glance 2023: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/e13bef63-en.
[29] OECD (2023), Equity and Inclusion in Education: Finding Strength through Diversity, OECD Publishing, Paris, https://doi.org/10.1787/e9072e21-en.
[38] OECD (2023), PISA 2022 Results (Volume I): The State of Learning and Equity in Education, PISA, OECD Publishing, Paris, https://doi.org/10.1787/53f23881-en.
[7] OECD (2022), OECD Economic Surveys: Ireland 2022, OECD Publishing, Paris, https://doi.org/10.1787/46a6ea85-en.
[51] OECD (2022), Quality and Equity of Schooling in the German-speaking Community of Belgium, Reviews of National Policies for Education, OECD Publishing, Paris, https://doi.org/10.1787/9a6b6f3a-en.
[23] OECD (2021), Education at a Glance 2021: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/b35a14e5-en.
[27] OECD (2018), Equity in Education: Breaking Down Barriers to Social Mobility, PISA, OECD Publishing, Paris, https://doi.org/10.1787/9789264073234-en.
[5] OECD (2017), Education at a Glance 2017: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2017-en.
[32] OECD (2017), The Funding of School Education: Connecting Resources and Learning, OECD Reviews of School Resources, OECD Publishing, Paris, https://doi.org/10.1787/9789264276147-en.
[4] OECD (2016), Education at a Glance 2016: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2016-en.
[3] OECD (2012), Education at a Glance 2012: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2012-en.
[58] OECD (2010), Closing the Gap for Immigrant Students: Policies, Practice and Performance, OECD Reviews of Migrant Education, OECD Publishing, Paris, https://doi.org/10.1787/9789264075788-en.
[2] OECD (2008), Education at a Glance 2008: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2008-en.
[1] OECD (2005), Education at a Glance 2005: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2005-en.
[34] Poesen-Vandeputte, M. and I. Nicaise (2014), “Rich schools, poor schools. Hidden resource inequalities between primary schools”, Educational Research, Vol. 57/1, pp. 91-109, https://doi.org/10.1080/00131881.2014.983722.
[46] Trinity College Dublin (2018), Growing Up in Ireland: Key findings on the lives of 9-year-olds launched, https://www.tcd.ie/news_events/articles/growing-up-in-ireland-key-findings-on-the-lives-of-9-year-olds-launched/ (accessed on 29 February 2024).
Annex 3.A. Alternative estimation of the proportion of disadvantaged students in DEIS and non-DEIS schools
Both Delaney et al. (2023[19]) (based on PIRLS 2021) and Nelis et al. (2021[20]) (based on PISA 2018) provide information on the proportion of disadvantaged students within DEIS and non-DEIS schools, as well as the proportion of all students by DEIS status of their school in the respective samples. The combination of these figures allows us to estimate the proportion of disadvantaged students attending DEIS schools at the time of measurement.
Let DD represent the proportion of disadvantaged students in DEIS schools and DN the proportion of disadvantaged students in non-DEIS schools; and let S represent the share of the overall student population attending DEIS schools; then the share of all disadvantaged students attending schools DEIS (SD) equals DD*S / [DD*S + DN*(1-S)].
The relevant parameters can be found in Annex Table 3.A1 for primary education and Annex Table 3.A2 for post-primary education.
The estimation for primary education is based on Delaney et al. (2023[19]). At this level, the formula must be adjusted to take into account the four categories of DEIS schools.
Annex Table 3.A1. Parameters used for estimation of DEIS coverage at primary level
Source: Delaney et al. (2023[19]), PIRLS 2021: Reading results for Ireland, Figure 3.4 and Figure 3.7.
The share of disadvantaged students in primary DEIS schools in 2021 is thus estimated at 42%.
Analogously, at the post-primary level, Nelis et al. (2021[20]) report disadvantaged proportions as shown in Annex Table 3.A2.
Annex Table 3.A2. Parameters used for estimation of DEIS coverage at post-primary level
Source: Nelis et al. (2021[20]), Beyond achievement: home, school and wellbeing findings from PISA 2018 for students in DEIS and non-DEIS schools, Figure 4.4.
The “adjusted” share of disadvantaged students in post-primary DEIS schools in 2018 then equals 39%.
Two assumptions are necessary to arrive at these estimations:
Representativeness of the PIRLS/PISA samples in terms of composition of the DEIS schools; and
Adequacy of the criteria used to identify disadvantaged students (i.e. the bottom quartile of socio‑economic status).
Notes
← 1. The term “baseline funding” is used as the generic set of rules for the funding of education (or schools) irrespective of the additional funding prevailing under DEIS.
← 2. GNI* is defined as GNI less factor income of redomiciled companies, less depreciation on research and development service imports and trade in intellectual property, and less depreciation on aircraft leasing.
← 3. The capitation grant is first based on the number of students in the previous school year and corrected for the revised number of students in April of the current school year. In addition to the capitation grant, there are a number of smaller grants based on a variety of other criteria, which makes the school funding less transparent and equitable than could be expected (GrantThornton and SVP, 2023[36]).
← 4. Whereas most people would agree that inequalities based on social background need to be eliminated, there are differences in the belief how far an individual’s potential can be achieved. From a meritocratic perspective, this potential is constrained by (mainly innate) ability. However, ability is hard to measure accurately and observed ability appears to be correlated with social background, which suggests that the “meritocratic dream” can never be achieved. By contrast, egalitarianists tend to attribute the correlation between ability and social background to influences from the home environment on children’s cognitive development and claim that the social gradient of ability itself can/should be eliminated too (Elford, 2015[53]).
← 5. Ireland is subdivided in 18 919 small areas, each covering just under 100 households.
← 6. Within any DEIS category, the level of some types of support is (partly or fully) proportional to the number of qualifying students, not the severity of needs, while other types of support are independent of the number of qualifying students.
← 7. Note that in the Flemish Community of Belgium, students are given different weights depending on a combination of a set of social characteristics to reflect different degrees of disadvantage at individual level. The additional resources are then allocated to schools on the basis of the overall weight of their student population, rather than the number of disadvantaged students as such. This also means that, among the 49% eligible students, many have a relatively low weight.
← 8. Note that the definitions of “disadvantage” used by Delaney et al. (2023[19]) and Nelis et al. (2021[20]) – who use PIRLS and PISA data, respectively – differ from the DEIS definition. However, we have no reason to believe that the former definitions are less valid than the DEIS definition.
← 9. In the Irish case, bundles of geographical areas are actually “weighted” via the HP Index values of students.
← 10. Until the turn of the millennium, resourcing schemes were mainly based on the geographical location of schools within disadvantaged districts or regions (e.g. Title I in the United States Educational Priority Areas in the United Kingdom, Zones d’Education Prioritaire (Priority Education Zones) in France). In some cases (e.g. in the United States) the geographical demarcation was meant to compensate for inequalities generated by the decentralised funding of schools. In other cases (e.g. Territórios Educativos de Intervenção Prioritária (Priority Intervention Educational Territories) in Portugal) coordinated cross-sectoral policies were conducted to address the cumulation of disadvantage in urban areas of localised disadvantage or remote rural areas. However, the targeting of territories in education generally lacks precision in coverage, leaving large numbers of disadvantaged students residing or going to schools in other areas without additional support (Bernardo and Nicaise, 2000[22]; Demeuse et al., 2008[14]).
← 11. The School Meals Programme is in principle organised by local authorities, and, therefore, probably also partly financed by them. Statistics about the financial contribution of local authorities are not available.
← 12. In 2016 the Department of Education and Skills was responsible for the whole education and training system (including tertiary education and vocational training). The budget breakdown for capital investments by level of education was not published in the budget expenditure report.
← 13. Consumer prices rose 17% between December 2016 and December 2023 in Ireland.
← 14. The North East Inner City Multi-Disciplinary Team model involving school leadership, teachers, NEPS, the Inspectorate and other support services working in eight primary schools can be seen as an example of good practice.
← 15. There are actually two components in the School Meals Programme: a statutory scheme for urban primary schools, organised by local authorities, and a non-statutory scheme operated (or outsourced) by the schools themselves. Both schemes are co-financed by the DSP.
← 16. At the primary level, the grants for school books are universal and therefore not linked to DEIS.
← 17. The label “Matthew effect” is used to denote the regressive distribution effects of public expenditure. The expression was launched by the sociologists Robert Merton and Harriett Zuckerman.
← 18. In PISA, a socio-economically disadvantaged school is a school in the bottom quarter of the index of economic, social and cultural status in the relevant country.
← 19. Data on the percentage of students entitled to an examination fee waiver are available up to 2020 only as examination fees have not been levied since the onset of the COVID-19 pandemic.
← 20. It should be noted that the HP Index element did not incorporate the additional elements included in the refined DEIS identification model (i.e. students residing in International Protection Accommodation Services centres, Emergency Orientation and Reception Centres, those experiencing homelessness, and Traveller and Roma children) who would all have met the threshold for an examination fee waiver. However, these minorities are probably small relative to other disadvantaged groups.
← 21. A horizontal equity issue arises when two or more schools with identical needs are funded unequally. A vertical equity issue refers to a situation where a school with greater needs is funded less than schools with more limited needs.
← 22. Donohue et al. (2023[52]) examined trends in mathematics performance in PISA between the waves of 2012 and 2022. They found a decrease in the average performance gap between the top and bottom socio‑economic quartiles between 2012 and 2018, largely offset by an increase between 2018 and 2022. However, they did not want to examine trends by DEIS status because of changes in the DEIS criteria and selective non-response issues within DEIS schools.
← 23. The DEIS grant is proportional to the number of disadvantaged students in the DEIS school.
← 24. A precise estimate of the extent of misclassification remains hazardous because the HP Index does not cover all Traveller and Roma students, homeless and refugee children, for whom corrections are made in the DEIS definition of disadvantage. However, the resulting error is not expected to be very large. Note also that the PISA sample size is limited and covers only 15-year-olds whereas the DEIS selection covers all students within the selected schools.
← 25. See Chapter 5 for a more extensive discussion of this report.
← 26. This recommendation preceded the recent measures. It was not possible to verify to what extent the most recent measures closed the gap estimated in the report.
← 27. The “baseline” capitation grant is a lump sum cash transfer to schools per student aimed to cover all running expenses of schools. In the Irish case, the mainstream capitation grant is not equal for all students, as enhanced capitation grants apply for students with special educational needs, and Traveller and Roma students. The DEIS grant is a lump sum supplement for students tagged as disadvantaged in DEIS schools.
← 28. In regard to the relationship between mental health and educational achievement, see e.g. “Research summaries: The relationship between mental health and academic achievement” (National Association of School Psychologists, 2020[56]). As for the relationship between migration background and educational achievement, see e.g. “Closing the Gap for Immigrant Students” (OECD, 2010[58]) or “The Impact of National and School Contextual Factors on the Academic Performance of Immigrant Students” (Finch, Hernández Finch and Avery, 2021[57]).
← 29. The HP Index has already been used in studies of progression into higher education (Higher Education Authority, 2019[55]) and further education and training (Connor and Guerin, 2019[54]).
← 30. In 2017, the threshold for students to qualify as disadvantaged was set at -10 for the HP Index. In 2022, the algorithm was extended to students with HP Index scores between -7.5 and -10, who are assigned a weight of 0.5.