Reducing Inequalities by Investing in Early Childhood Education and Care

Data informing the Starting Strong VIII report and its supplementary outputs (country notes) were derived from various sources:

Country notes were produced for the five countries that engaged in the policy review in greater depth: Australia, Bulgaria, Ireland, Japan and Korea. These country notes follow a standardised format and address a common set of issues but vary in focus, as they explore questions deemed of particular relevance to these countries. The notes were prepared by the OECD Secretariat and reviewed by the countries. The preparation of the notes followed the same methodological procedures implemented for the main report.

Figures 1.5, 3.5, 3.6, 5.1, 5.2, 5.3 and 6.1, and Table 1.2 present results from tests for statistically significant differences between estimates to understand whether observed differences in sampled data are likely to represent actual differences within the population. Consequently, for each sample estimate, there is an associated degree of uncertainty expressed in a standard error.

In this report, differences among sample estimates are labelled as statistically significant when a difference would be observed less than 5% of the time if there were no difference in corresponding population values (statistical significance at the 95% level). In other words, the risk of reporting a difference as significant when such difference, in fact, does not exist, is contained at 5%.

Reported standard errors and differences between estimates were calculated in line with the methodology of each of the source databases; and considering independence between groups or lack thereof.

Figures 1.5, 5.6 and 9.3, and Table 1.1 present an association between two variables. For each pair of variables, the association is calculated as a linear regression, and coefficients are estimated using Ordinary Least Squares. When controlling for another variable, this variable is added as an independent variable in the regression, and its results are not presented.

In Figure 1.5 and Table 1.1, the estimates shown correspond to the linear regression coefficients. In Figures 5.6 and 9.3, the fitted line represents the regression coefficient graphically, while the value shown corresponds to the r-squared value associated with the linear regression. Since the regression is bivariate, the r-squared value corresponds to the squared value of the correlation between the two variables.

Figure 7.3 relies on data from TALIS Starting Strong 2018, a large-scale international survey that focuses on the ECEC workforce. Questionnaires were administered to staff and leaders to collect data on their characteristics, practices at work and views on the ECEC sector, with an emphasis on aspects that promote conditions for children’s learning, development and well‑being.

Nine countries participated in TALIS Starting Strong 2018: Chile, Denmark, Germany, Iceland, Israel, Japan, Korea, Norway and Türkiye. All of these countries collected data from staff and leaders in pre-primary education (ISCED Level 02) settings. In addition, four of the nine countries (Denmark, Germany, Israel and Norway) collected data from staff and leaders in settings serving children under age 3. For each level of ECEC in which these countries participated, the study aimed to survey a representative sample of ECEC staff and centre leaders.

Data in Figure 7.3 refer to data on diversity of children within ECEC centres. Presented data show the percentage of centres whose leaders reported 10% or more of the children in the setting each having one of the following characteristics: children from socio-economically disadvantaged homes, children with special education needs, children with a different first language, and children who are refugees. Several dimensions of diversity can accumulate within a given ECEC centre. Data for ECEC settings for children under age 3 are limited to centre-based settings to ensure comparability with ISCED 02 (home-based settings are excluded). Denmark did not meet the technical standards on response rates; its results are therefore not shown in Figure 7.3.

For more information, see the TALIS Starting Strong 2018 Technical Report (OECD, 2019[1]).

Figure 3.2 relies on data from IELS, an international survey that assessed the skills of children at age 5 attending early childhood education centres or schools in Estonia, the United Kingdom (England) and the United States in 2018. The study aimed to identify key factors that drive or hinder the development of early learning.

IELS data in Figure 3.2 are disaggregated by various measures of children’s social and economic backgrounds. Information on parental/guardian’s education comes from the parent questionnaire, with levels of parental education classified following ISCED. The measure of socio-economic status (SES index score) was derived nationally, based on three indices: i) highest parental occupational status of parents; ii) highest educational level of parents (in years of education according to ISCED); and iii) household income. The number of books in the home refers to the number of children’s books that parents reported as present in the home environment. Disadvantaged refers to families in the bottom quartile of the national distribution of the socio-economic status (SES) index, whereas advantaged refers to families in the top SES quartile nationally.

For more information, see the IELS Technical Report (OECD, 2021[2]) and the Improving Early Equity report (OECD, 2022[3]).

Figures 1.4, 1.5 and 3.5, and Tables 1.1 and 1.2 rely on data from the PISA 2015 and 2022 assessments. PISA is a triennial test and survey of 15-year-old students that assesses the extent to which they have acquired key knowledge and skills in mathematics, reading and science that are essential for full participation in social and economic life. In addition, PISA uses student questionnaires to collect information from students on various aspects of their home, family and school background, and school questionnaires to collect information from schools about various aspects of organisation and educational provision in schools. PISA 2015 was conducted in the current 38 OECD member countries (members have increased since 2015: Lithuania joined in 2018, Colombia in 2020, and Costa Rica in 2021) and 34 partner countries and economies, whereas PISA 2022 was conducted in 37 OECD member countries (all but Luxembourg), and 44 non-OECD member countries and economies.

Results in these Figures and Tables refer to students’ socio-economic background, measured through the PISA index of economic, social and cultural status (ESCS). This index is based on three variables related to family background: i) parents’ highest level of education (PAREDINT); ii) parents’ highest occupational status (HISEI); and iii) home possessions, including books in the home (HOMEPOS). Data for these Figures and Tables refer to students as “advantaged” or “disadvantaged”. A socio-economically disadvantaged (or advantaged) student is a student in the bottom (or top) quarter of the ESCS index in their own country. The socio-economic gap for a variable refers to the difference in value for that variable between advantaged and disadvantaged students. To make PAREDINT scores for PISA 2015 comparable to PAREDINT scores for PISA 2022, new PAREDINT scores were created for each student who participated in previous cycles using the coding scheme used in PISA 2022. These new PAREDINT scores were used in the computation of trend ESCS scores. Estimates obtained with this methodology may deviate slightly from estimates in international PISA reports published before 2022 or in national reports.

Results in Figures 1.4 and 1.5, and in Table 1.1 refer to students’ participation in ECEC. Students were asked about their participation in ECEC and the amount of time they participated. In Figure 1.4 and Table 1.1, the socio-economic gap is calculated for 2015 and 2022, and countries are classified as “increased gap” where the difference grew more than 3 percentage points between 2015 and 2022, “no change” where the difference was between -3 and 3 percentage points, and “narrowed gap” where the difference changed by a number less than -3 percentage points (equivalent to narrowing by more than 3 percentage points). Data from PISA and the European Union Statistics on Income and Living Conditions (EU-SILC) in Figure 1.4 and Table 1.1 are not directly comparable, as questions on participation in ECEC in the two surveys are different and target different respondents (students versus parents), and the surveys follow different methodologies.

For more information, see the PISA 2015 Technical Report (OECD, 2017[4]) and the PISA 2022 Technical Report (OECD, 2024[5]).

Figures 1.1, 4.4, 5.6 and 9.7 rely on data from the OECD Family Database. The database contains cross-national indicators on family outcomes and policies across OECD countries, partners, and EU member states. It includes indicators on the structure of families, families’ labour market position, public policies for families and child outcomes.

Data in Figure 1.1 show poverty rates, which are defined as in the OECD Income Distribution Database: the percentage of the national population living under the poverty threshold, excluding lump-sum payments. The poverty threshold is set at 50% of the median disposable income in each country (relative threshold). Child poverty is defined as the percentage of families with 0-17-year-olds with an equivalised household disposable income below the poverty threshold.

Figure 5.6 refers to data from the OECD Family Database, Indicator PF3.2, on ECEC enrolment for 0-2-year-olds. Data for 0-2-year-olds generally include children enrolled in early childhood education services (ISCED 2011 Level 0) and other registered ECEC services (outside the scope of ISCED 0, because they are not in adherence with all ISCED-2011 criteria), except for

• Denmark, Finland and Spain (enrolment in ECEC only for ISCED 0 services).

• Belgium, Czechia, France, Hungary, Greece, Ireland, Italy, Latvia, Luxembourg, the Netherlands, Poland, the Slovak Republic, Switzerland, the United Kingdom, Bulgaria, Croatia and Romania (enrolment in ECEC regardless of whether services are ISCED 0 or recognised).

• The United States (enrolment in ECEC regardless of whether the services are paid, registered or ISCED 0).

Data for Bulgaria, Croatia, Czechia, Denmark, Greece, Iceland, the Netherlands, Poland, the Slovak Republic, Slovenia, Spain and Sweden refer to 2017; data for Argentina, Australia, Austria, Brazil, Chile, Colombia, Estonia, France, Germany, Hungary, Italy, Korea, Lithuania, Mexico, Norway, Romania, Türkiye and the United Kingdom refer to 2018; data for Japan refer to 2019; data for New Zealand and Portugal refer to 2020; and data for Latvia refer to 2021.

Figures 4.4 and 9.7 refer to data on social expenditure per child. The data do not always fully capture non-central government spending. Cash benefits are adjusted for direct tax, but in-kind benefits and spending on education are not. Health-related spending by age is omitted since it lacks comparability across countries. Data for family benefits include cash benefits (family allowances, maternity and parental leave, and other cash benefits), and benefits in kind (ECEC, home help/accommodation, and other benefits in kind).

See the OECD Family Database Indicator PF3.2 (OECD, 2024[6]) and Social Expenditure Database Manual (OECD, 2019[7]) for more information.

Figure 5.5 relies on data from the OECD Net Childcare Costs Indicator and Tax-Benefit model. Data reflect the gross childcare fees and net costs of full-time care in a typical childcare centre for a two-earner two-child family, where both parents are aged 40, in full-time employment and the children are aged 2 and 3. The components of the cost are shown separately, even when they are deducted in practice, and they are all considered in the calculation of the net childcare costs. The data are based on the following definitions and assumptions:

• Gross earnings for the two earners in the family are set equal to 100% of average earnings for the first earner, and 67% of average earnings for the second earner. Both parents are assumed to be working full-time. Average earnings/the average wage (AW) refers to the gross wage earnings paid to average workers, before deductions of any kind (e.g. withholding tax, income tax, private or social security contributions and union dues).

• Families are assumed to use full-time centre-based care.

• Gross childcare costs are the fees charged to parents after any public subsidies received by the provider but before any fee reductions or discounts provided to users based on their characteristics.

• Childcare benefits are childcare allowances or fee rebates that are explicitly designed to reduce the financial costs of childcare.

• Impacts in taxes include tax concessions conditional on childcare use or childcare expenses, as well as other changes in taxes resulting from childcare use.

• Impacts in other benefits show the changes in all other benefits (except childcare benefits) resulting from childcare use, notably the loss of homecare allowances, which usually require recipients to not use formal childcare services, thus increasing the net childcare costs.

• Where benefit entitlements change over time, calculations refer to the second month of benefit receipt. If housing benefits are included in the calculations, these are calculated assuming a household renting in the private market paying rent equal to 20% of the average wage. Rent levels are the same for all family types.

• Net childcare costs are calculated as the difference in “family net income” between a family that uses centre-based childcare services and an otherwise identical family that does not. Family net income is the sum of gross family earnings plus cash benefits, minus income taxes and social contributions paid by workers. The methodology takes into account gross childcare fees, childcare-specific supports designed to reduce the costs faced by parents, and the interaction between childcare-specific policies and any other tax and benefit policies. Results are then presented in percentage of average earnings.

Where benefit rules are not determined on a national level but vary by region or municipality, results refer to a “typical” case. Concerned countries have data based on a region or municipality instead of the whole country: Australia (New South Wales), Austria (Vienna), Belgium (French Speaking Community), Croatia (Zagreb), Czech Republic (Prague), Estonia (Tallinn), Germany (Berlin), Greece (Athens), Hungary (Budapest), Iceland (Reykjavik), Italy (Rome), Latvia (Riga), Lithuania (Vilnius), Norway (Oslo), Poland (Warsaw), Slovak Republic (Bratislava), Spain (Madrid), Sweden (Stockholm), Switzerland (Zurich), United Kingdom (England), United States (Michigan).

For more information, see OECD Net Childcare Cost Indicator and the OECD TaxBEN: Tax and Benefit simulation model: Methodology, user guide and policy applications (OECD, 2024[8]), and the OECD calculator of taxes and benefits.

Figures 1.3, 1.4, 4.2, 6.2, 9.1, 9.2, 9.3, 9.4, 9.5 and 9.6, and Table 1.1 rely on data from the OECD EAG. EAG is a source of data on the state of education for OECD member, partner and accession and countries. It is produced annually since 1997, and covers indicators on the output of educational institutions, the impact of learning, access, participation and progression in education, investment in education, and teachers, the learning environment, and the organisation of schools. Data from EAG used in Starting Strong VIII were taken from the 2012, 2014, 2017, 2018, 2022, 2023 and 2024 editions of EAG.

Figure 1.3 refers to the enrolment rates in education by age for 0-5-year-olds in 2022. This indicator is taken from the OECD EAG 2024 report, table B1.1, for all countries except Bulgaria. Data for Bulgaria are taken from the EAG 2024 database, updated on 25 November 2024, due to an update of population data.

Figure 1.4 and Table 1.1 show enrolment levels in both ECEC (ISCED 0) and primary education (ISCED 1) at age 4 by country from EAG 2018 Table B2.1b, the latest data available for this indicator. However, the data source, indicator or year of reference differ for some countries:

• For Brazil, Denmark, New Zealand and the United Kingdom, data for 2005 comes from EAG 2017, Table C2.1.

• For Australia, Greece, Korea and Sweden, data from 2005 comes from EAG 2014, Table C2.1.

• For Argentina and Ireland, the year of reference differs from 2005: 2010; and comes from EAG 2012, Table C2.1.

• For Bulgaria and Croatia, the year of reference 2015 differs: 2013; the age group of 4 differs: 3 to 5; and the education level differs from ISCED 0 and ISCED 1: only ISCED 0. The data source is EAG 2022, Table B2.1.

In Figures 9.1 and 9.2, related to trends on expenditure, Australia was omitted due to a number of changes in data sources and methodology in 2019 that caused a significant break in the series.

Figures 9.2 and 9.3 refer to data on private expenditure on education. Private expenditure can be categorised according to sources of education funds: expenditure by households and expenditure by other private entities. Expenditure by households includes transfers to households and students used for tuition fee payments to educational institutions, payments for ancillary services provided by educational institutions, and costs borne by private households for the purchase of educational goods and services outside of educational institutions. It excludes the living expenses of students. Expenditure by other private entities consists of direct payments to educational institutions and subsidies to students or households. Data are shown after public transfers, which includes household subsidies and subsidies to other private entities.

Figures 6.2 and 9.3 show data referring to private institutions, that comprise government-dependent and independent institutions. A government-dependent private institution is one that receives 50% or more of its core funding from government agencies or whose teaching personnel are paid by a government agency. An independent private institution is one that receives less than 50% of its core funding from government agencies and whose teaching personnel are not paid by a government agency.

Figures 9.4 and 9.5 show data on public expenditure on education. Public expenditure is defined as spending by public authorities at all levels, excluding expenditure not directly related to education, unless the activities/services are provided as ancillary services by educational institutions. It includes expenditure on education by other ministries or equivalent institutions, as well as subsidies provided to households and other financial entities which can be attributable to educational institutions or not. It can come from central (national) government, regional governments or local governments. Inter-governmental transfers of funds are transfers of funds specifically designated for education from one level of government to another. They are defined as net transfers from a higher level to a lower level of government.

Figure 9.6 shows data on salaries of pre-primary teachers relative to earnings of tertiary-educated workers. Data refer to the ratio of salary, using annual average salaries (including bonuses and allowances) of full-time teachers in public institutions relative to the earnings of workers with similar educational attainment (weighted average) and to the earnings of full-time, full-year workers with tertiary education, for pre-primary education. Where the year of reference for the earnings of tertiary-educated workers and the salaries of teachers differ, the earnings of tertiary-educated workers have been adjusted to the reference year used for salaries of teachers using deflators for private final consumption expenditure. Annual salaries are provided in national currencies and converted into USD using purchasing power parity for private consumption.

For more information, see Education at a Glance 2012 (OECD, 2012[9]), Education at a Glance 2014 (OECD, 2014[10]), Education at a Glance 2017 (OECD, 2017[11]), Education at a Glance 2018 (OECD, 2018[12]), Education at a Glance 2022 (OECD, 2022[13]), and Education at a Glance 2024 (OECD, 2024[14]); as well as the Sources, Methodologies and Technical Notes (OECD, 2024[15]), and the Handbook for Internationally Comparative Education Statistics (OECD, 2018[16]) for definitions.

Figures 3.3 and 3.4 rely on data from the OECD Income Distribution Database (IDD). The IDD contains data on levels and trends in income inequality and poverty. It is updated on a bi- or tri-annual basis. The latest available version, used for Starting Strong VIII, was the July 2024 update.

These Figures are based on equivalised household disposable income, measured as the income after taxes and transfers, adjusted for household size. Data in Figure 3.3 refer to income inequality, which is measured through the Gini coefficient on household disposable income. The Gini coefficient expresses the difference between the cumulative share of households and the cumulative share of disposable income. The coefficient varies between 0 (where all the population has the same income) and 1 (where all income goes to one individual). Data in Figure 3.4 show poverty rates, which are defined as the percentage of the national population living under the poverty threshold, excluding lump-sum payments. The poverty threshold is set at 50% of the median disposable income in each country (relative threshold).

The data shown in each Figure correspond to the following years:

• Data shown for 2022 refer to 2022 for all countries except: Costa Rica and United States (2023), Japan, Switzerland (2021), Australia and Germany (2020); Denmark, (2019); Iceland (2017). 2022 data for the Netherlands and 2023 data for the United States are provisional.

  • Survey estimates for 2020 are subject to additional uncertainty and are to be treated with extra caution, as in most countries the survey fieldwork was affected by the Coronavirus (COVID-19) pandemic.

• Data shown for 2019 refer to 2019 for all countries except Australia, Japan and Mexico (2018); Chile (2015 and 2017); Iceland (2016).

• Data shown for 2007 refer to 2007 for all countries except Chile (2009); Australia, France, Germany, Israel, Mexico, Norway, Sweden and the United States (2008); Brazil and Japan (2006).

• Additionally, for Romania, the value of goods produced for own consumption was excluded from the income definition due to methodological issues.

  For more information, see the OECD Framework for Statistics on the Distribution of Household Income, Consumption and Wealth (OECD, 2013[17]), and the OECD Income Distribution Database (OECD, 2024[18]).

The EASIE database is the data collection activity of the European Agency for Special Needs and Inclusive Education, an independent organisation that facilitates collaboration between education ministries of its 31 member countries and jurisdictions across Europe. The Agency provides an international report with indicators on access and placement in inclusive education for levels from pre-primary to upper secondary, based on the International Standard Classification of Education (ISCED). Data from EASIE in Starting Strong VIII is taken from the 2020/2021 cross-country report.

Figure 7.2 is based on data from EASIE on children with an official decision of special education needs (SEN) and their enrolment in mainstream ECEC. The enrolment rate (in percentage) of children with an official decision of SEN in inclusive education is calculated as the number of children with an official decision of SEN educated with their peers in mainstream groups for 80% or more of the time, divided by the overall number of children with an official decision of SEN at the pre-primary level. The identification rate (in percentage) of children with an official decision of SEN is calculated as the overall number of children with an official decision of SEN, divided by the number of children enrolled in any form of recognised education at the pre-primary level (ISCED 02).

For more information, see the European Agency Statistics on Inclusive Education: 2021/2022 School Year Dataset Cross-Country Report (European Agency for Special Needs and Inclusive Education (EASIE), 2024[19]).

Figures 1.3, 5.1, 5.2, 5.3, 5.7 and 6.1, and Table 1.1 rely on data from Eurostat’s EU-SILC, an instrument that collects comparable cross-sectional and longitudinal microdata on income distribution, poverty and social exclusion, as well as policies on poverty and living conditions. The data used in Starting Strong VIII referring to 2023 correspond to EU-SILC 2023 for all countries, except for Germany (2022) and Switzerland (2021); data referring to 2010 correspond to EU-SILC 2010 for all countries.

These Figures and Tables show data on inequalities in ECEC participation by income tertile. Income based on EU-SILC data is measured as equivalised disposable household income (variable: EQ_INC): the disposable (post tax and transfer) income of the household divided by the number of household members in equivalised adults (using the OECD equivalence scale). Disadvantaged children refer to children who are in a household in the lowest tertile of income; advantaged children are children who are in a household in the highest tertile of income. The socio-economic gap is the difference in value for advantaged and disadvantaged children.

Figure 1.4 and Table 1.1 present data from EU-SILC on gaps in participation in ECEC based on EU-SILC household data on children’s current participation in ECEC (variable: RL010). Participation is reported as a weighted average for 3-5-year-olds. The socio-economic gap is classified as “narrowed” if its reduction over time is of at least 3 percentage points; “increased” when the gap widens by at least 3 percentage points; and “no change” when it is between -3 and 3 percentage points. When EU-SILC data are not available, they are replaced by data from PISA. Data from EU-SILC and PISA in Figure 1.4 and Table 1.1 are not directly comparable, as questions on participation in ECEC in the two surveys are different and target different respondents (students versus parents), and the surveys follow different methodologies.

Figures 5.1, 5.2, 5.3, 5.7 and 6.1 rely on data on participation in one or several of the different types of ECEC and informal care. The types are:

• Regulated centre-based and home-based ECEC (variables: RL010, RL020, RL040): refers to children using regulated centre-based services (e.g. nurseries or day care centres and preschools, both public and private), organised family day care, and care services provided by (paid) qualified childminders organised and controlled by a structure, regardless of whether the service is registered or ISCED-recognised.

• Unregulated childminder care (variable: RL050): refers to children using care services provided by childminders who are not organised and controlled by a structure (e.g. babysitters, au pairs).

• Informal care (variable: RL060): refers to children benefitting from unpaid care provided by grandparents, household members other than parents, other relatives, friends or neighbours.

• After-school care (variable: RL030): refers to children benefitting from care in centre-based services outside preschool hours – only the hours of care before and after preschool are reported (cultural and sport activities outside preschool hours, such as a club, music lessons, etc. are not included as far as they are not used as a childcare service but rather for the child’s leisure).

For more information, see the variable descriptions in the scientific use files of the EU-SILC 2024 data release (Eurostat, 2024[20]). The responsibility for all conclusions drawn from the data lies entirely with the authors.

Figures 3.6 and 7.1, and Tables 1.1 and 1.2 rely on data from TIMSS 2011 and 2019. TIMSS is an international assessment of student achievement in mathematics and science at the fourth and eighth grade levels, conducted every four years since 1995. In TIMSS 2011 at the fourth-grade level, 52 countries and 7 benchmarking entities (jurisdictions) participated in the assessment. In TIMSS 2019, 58 countries and 6 benchmarking entities (jurisdictions) participated in the fourth-grade assessment. Where TIMSS was not available, data were taken from PIRLS 2011 and 2021. PIRLS is an international survey on students’ reading achievement in fourth grade, conducted every five years since 2001. PIRLS 2011 had 48 participating countries and 9 benchmarking entities (jurisdictions), while PIRLS 2021 had 57 participating countries and 8 benchmarking entities (jurisdictions). For both TIMSS and PIRLS, data used in Starting Strong VIII refer to fourth-grade students (approximately 10-year-old students, depending on countries’ education institutions). The only country referring to data from both TIMSS (2011) and PIRLS (2021) is Canada for participation in ECEC.

Both surveys comprise five context questionnaires, which collect information about the students’ lives and the home and school context in which students learn. All Figures based on TIMSS and PIRLS in Starting Strong VIII use data from the context questionnaires (for socio-economic background, ECEC attendance, multilingualism, and early literacy and numeracy activities), and Table 1.2 uses data on assessment scores as well. Some countries do not have available data for one or more of these context questionnaires, even when they participated in the assessments, and are therefore not shown in the Figures. Standard errors on all surveys were calculated according to the methodologies described in the user guides and used 150 replicate weights for their calculation to make estimates comparable across trends.

Data in Table 1.1 and in Figure 7.1 refer to participation in ECEC for more than two years. ECEC attendance is derived from the early learning survey questionnaire, answered by parents. Data from TIMSS/PIRLS, EU-SILC and PISA in Table 1.1 are not directly comparable, as questions on participation in ECEC in the three surveys are different and target different respondents (students versus parents), and the surveys follow different methodologies.

Results in Figure 3.6 and Table 1.2 refer to students’ socio-economic background, measured with the Home Resources for Learning scale (variable: ASBGHRL). The scale is built from the responses to five items: i) number of books at home; ii) number of home study supports (taken from student responses in the student context questionnaire); iii) number of children’s books at home; iv) highest level of education of either parent; v) highest level of occupation of either parent (taken from parent responses in the early learning survey questionnaire). A socio-economically disadvantaged (or advantaged) student is a student in the bottom (or top) quarter of the ESCS index in his or her own country. The socio-economic gap for a variable refers to the difference in value for that variable between advantaged and disadvantaged students.

Data in Table 1.1 refer to the association between attendance of ECEC for more than two years and mathematics scores. The association is classified as having increased when the increase over time was of at least 5 score points, as having decreased when the decrease over time was of at least 5 score points, and as not having changed when the change over time was between +5 and -5 score points.

Results in Figure 3.6 refer to high frequency of early literacy and numeracy activities. This frequency is measured using the early literacy and numeracy activities scale (variable: ASDHLNT). This scale is built from parent responses in the home questionnaire to 18 items on the frequency of activities they conducted with their children before primary school at home. Students were classified as “often” on this scale if their parents report conducting 9 of the 18 activities often and the other 9 “sometimes”, on average.

Data in Figure 7.1 refer to multilingual children. This variable is built from parent responses in the early learning survey. Parents were asked which languages their children spoke before beginning primary school (variable: ASBH03, items 1-6), and the variables were recoded to produce the number of languages children spoke before beginning primary school. Multilingual children are those whose parents indicated they spoke any two or more languages before beginning primary school. In most countries, “non-native speakers” of the TIMSS mathematics and science test language were excluded from the test. Non-native speakers are students who are unable to read or write in the language of the test and would be unable to overcome the language barrier and, typically, are students who have received instruction in the test language for less than one year.

For more information, see TIMSS 2011 User Guide for the International Database (Foy, Arora and Stanco, 2013[21]), PIRLS 2011 User Guide for the International Database (Foy and Drucker, 2013[22]), TIMSS 2019 User Guide for the International Database (Fishbein, Foy and Yin, 2021[23]) and PIRLS 2021 User Guide for the International Database (Fishbein, Yin and Foy, 2024[24]); as well as Methods and procedures in PIRLS and TIMSS 2011 (Martin and Mullis, 2012[25]), Methods and procedures: TIMSS 2019 Technical Report (Martin, von Davier and Mullis, 2020[26]) and Methods and procedures: PIRLS 2021 Technical Report (von Davier et al., 2023[27]).

Data in Figure 5.6 on social norms on working mothers comes from the World Values Survey (WVS) Wave 7: 2017-2022. The WVS (www.worldvaluessurvey.org) is a global research initiative studying shifts in human beliefs and values, and their influence on social and political dynamics. It has been operating in cycles since 1981, gathering nationally representative and comparable data in more than 120 countries.

For Figure 5.6, only data on the item “when a mother works for pay, the children suffer”, available for OECD member and accession countries, was used. The percentage shown corresponds to the percentage of people who reported they agree or strongly agree with this belief by country. The statement was presented in the questionnaire as stated above.

Data for Argentina, Bulgaria, Croatia, Czechia, Denmark, Greece, Iceland, the Netherlands, Poland, the Slovak Republic, Slovenia, Spain, and Sweden refer to 2017; for Australia, Austria, Brazil, Chile, Colombia, Estonia, France, Germany, Hungary, Italy, Korea, Lithuania, Mexico, Norway, Romania, Türkiye and the United Kingdom to 2018; for Japan to 2019; for New Zealand and Portugal to 2020; and for Latvia to 2021.

For more information, see the World Values Survey Wave 7 (2017 – 2022), WVS Database (Haerpfer, 2022[28]).

Assistants (or ECEC assistants): Refers to ECEC staff whose role is to provide support to the teachers or lead staff member with a group of children. Assistants usually have lower qualification requirements than teachers, ranging from no formal requirements to, for instance, vocational education and training. This role does not exist in every country.

Centre leader (or ECEC centre leader): Refers to the person in an ECEC centre with the most responsibility for administrative, managerial and/or pedagogical leadership. They may also be called the Head or Principal of the ECEC centre. Centre leaders may be responsible for the monitoring of children; the supervision of other staff; contact with parents and guardians; and/or the planning, preparation and carrying out of the pedagogical work in the centre. Leaders may also spend part of their time working with children.

Child-centred (beliefs, attitudes and practices): Refers to staff approaches and views which assume that learning is an active and co-operative process where children develop their own solutions to given problems.

Children’s development and learning: Refers to children’s academic and socio-emotional development, including children’s cognitive and non-cognitive development, which helps in the acquisition of skills, abilities, competencies, values and attitudes necessary for children to know themselves, build and maintain relationships with others, engage with life’s joys and complexities, and meet challenges in everyday life. Sometimes referred to as outcomes.

Curriculum/curriculum framework: Curriculum frameworks are overarching documents setting out the principles, standards, guidelines and approaches that could be used by ECEC staff to foster children’s development, learning and well-being. Curriculum frameworks may be broad, aiming to achieve several goals, embracing varied pedagogical approaches, covering several age groups or addressing only a particular age group. The implementation of curriculum frameworks is tightly linked with pedagogy, which can denote the theoretical foundation of a curricular approach.

ECEC: Refers to early childhood education and care. It includes all arrangements providing care and education for children under compulsory school age, regardless of setting, funding, opening hours or programme content (see also ECEC setting).

ECEC provider: Refers to the organisation that provides early childhood education and care services as its main objective. This can be a public institution as well as a private company, or a non-profit organisation.

ECEC quality: A multidimensional concept covering structural characteristics and process quality. Conceptualisations cover global aspects (such as warm climate), and domain-specific stimulation in learning areas such as literacy, emerging mathematics and science (see definitions for structural quality and process quality).

ECEC setting: Refers to the place where early childhood education and care (ISCED Level 0) is delivered. Most settings typically fall into one of the following categories:

• Home-based ECEC: Home-based settings refer to early childhood education and care that is provided in a home setting rather than a centre. These settings may or may not have an educational function and be part of the regular ECEC system. The minimum requirements defined for home-based settings vary widely across countries. Registered home-based setting providers are generally accredited to take care of children in their own homes.

• Regular centre-based ECEC: More formalised ECEC centres typically belong to one of these three sub-categories:

  • Age-integrated centre-based ECEC for children from birth or 1-year-old, up to the beginning of primary school: Can be called kindergarten, preschool, or pre-primary, and offers a holistic pedagogical provision of education and care (often full-day). To an increasing degree, these settings are linked to the educational system.

  • Centre-based ECEC for children aged 0-2: Often called “crèches”, these settings may have an educational function, but are typically attached to the social or welfare sector and are associated with an emphasis on care.

  • Centre-based ECEC for children aged 3+: Often called kindergarten or preschool, these settings tend to be more formalised and linked to the education system. Many of them are part-time and provided in schools, but they can also be provided in designated ECEC centres.

ISCED: The International Standard Classification of Education (ISCED) is the reference classification for organising education programmes and related qualifications by education levels and fields. The classification was revised in 2011 and is referred to as ISCED 2011 (see OECD/European Union/UNESCO-UIS, 2015, http://dx.doi.org/10.1787/9789264228368-en).

• ISCED 0 (or early childhood education and care): Refers to early childhood programmes that have an intentional education component and aim to develop cognitive, physical and socio-emotional skills necessary for participation in school and society. Programmes at this level target children below the age of entry into ISCED levels and are often differentiated by age.

• ISCED 01 – Early childhood educational development: Provides educational content designed for younger children (in the age range of 0 to 2 years). The learning environment is visually stimulating and language-rich and fosters self-expression with an emphasis on language acquisition and the use of language for meaningful communication. There are opportunities for active play so that children can exercise their co-ordination and motor skills under supervision and in interaction with staff.

• ISCED 02 – Pre-primary education: Designed for children from age 3 to the start of primary education. Through interaction with peers and educators, children improve their use of language and their social skills, start to develop logical and reasoning skills, and talk through their thought processes. They are also introduced to alphabetical and mathematical concepts, understanding and use of language, and are encouraged to explore their surrounding world and environment. Supervised gross motor activities (i.e. physical exercise through games and other activities) and play-based activities can be used as learning opportunities to promote social interactions with peers and to develop skills, autonomy and school readiness.

• ISCED 1 (or primary education): Designed to provide a sound basic education in reading, writing and mathematics and a basic understanding of some other subjects. Primary education usually begins between the ages of 5 and 7 and has a typical duration of six years.

Teachers (or ECEC teachers): Refers to individuals with the most responsibility for a group of children at the class- or playroom-level. They may also be called core practitioners, pedagogues, educators, pedagogical staff, preschool, pre-primary, kindergarten or early childhood teachers. In small settings, teachers may also be head of the setting while still working with children.

Private setting: Refers to a setting administered/owned directly or indirectly by a non-governmental organisation or private person/organisation (church, trade union, business or other concern). Private settings may be publicly subsidised or not. Private non-publicly-subsidised settings receive no funding from the public authorities and are independent in their finances and governance. Private publicly-subsidised settings operate completely privately but receive some or all their funding from public authorities – if more than 50% of their core funding comes from government agencies, they can be considered government-dependent private ECEC settings.

Process quality: Refers to the nature of the daily classroom and centre experiences of children in ECEC and concerns the more proximal processes of children’s experiences in their programme. Process quality includes all the proximal processes of children’s everyday experience – in addition to the interactions between children and ECEC staff, process quality concerns the interactions among children and the interactions of children with parents, the community and space and materials. While written curricula are considered a structural aspect, the actual activities provided in the ECEC centre are an aspect of process quality. The implementation of written curriculum is a central factor in the configuration of the child’s daily experience at the ECEC centre. Interactions between adults (staff-to-staff, parents and community) are also relevant factors influencing ECEC process quality.

Public settings: Refers to an ECEC centre managed by a public education authority, government agency, or municipality.

Staff (or ECEC staff): Refers to individuals whose professional activity involves the care and transmission of knowledge, attitudes and skills to children enrolled in an ECEC setting. This definition does not depend on the qualification held by the ECEC staff or on the delivery mechanism. ECEC staff may include teachers, educators, assistants or staff working with individual children, among other categories (see the definitions for teacher and assistant).

Staff-child ratio: Refers to the number of children per full-time member of staff. This can be a maximum (regulated) number, which indicates the maximum number of children that one full-time member of staff is allowed to be responsible for; or it can be an average: the average number of children a full-time staff member can be responsible for. Ratios can be either for main staff only (such as teachers or caregivers), commonly reported as teacher-child or teacher-student ratios, but can also include auxiliary staff, such as assistants.

Structural quality in ECEC: Refers to the distal factors that are typically regulated, such as children-to-staff ratio, group size and staff training/education, and create the framework for the experiences of children in ECEC. These characteristics are not only part of the ECEC location in which children participate, but also part of the environment that surrounds the ECEC setting, e.g. the community. Structural factors are an important precursor to the overall domain of process quality and to its subdomains. Additionally, structural features generally have indirect effects on children’s development, learning and well-being (through its influence on process quality). Structural quality is partly determined by legislation, policy and funding and is a major factor in the macroeconomic costs of ECEC. See also the definition for process quality.

[19] Dral, P., A. Lenart and A. Lecheval (eds.) (2024), European Agency Statistics on Inclusive Education: 2020/21 School year dataset cross-country report., Odense, Denmark, https://www.european-agency.org/resources/publications/EASIE-2020-2021-cross-country-report (accessed on 15 November 2024).

[20] Eurostat (2024), EU Statistics on Income and Living Conditions microdata, 2024 release, data, https://doi.org/10.2907/EUSILC2004-2023.

[23] Fishbein, B., P. Foy and L. Yin (2021), TIMSS 2019 User Guide for the International Database (2nd ed.), Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Boston College, https://timssandpirls.bc.edu/timss2019/international-database/ (accessed on 13 June 2024).

[24] Fishbein, B., L. Yin and P. Foy (2024), PIRLS 2021 User guide for the international database (2nd ed.), Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Boston College, https://pirls2021.org/data (accessed on 13 June 2024).

[21] Foy, P., A. Arora and G. Stanco (eds.) (2013), TIMSS 2011 User Guide for the International Database, Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Boston College, https://timss.bc.edu/timss2011/downloads/T11_UserGuide.pdf (accessed on 13 June 2024).

[22] Foy, P. and K. Drucker (eds.) (2013), PIRLS 2011 User Guide, Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Boston College, https://timssandpirls.bc.edu/pirls2011/downloads/P11_UserGuide.pdf (accessed on 13 June 2024).

[28] Haerpfer, C. (ed.) (2022), World Values Survey: Round Seven – Country-Pooled Datafile Version 6.0, Madrid, Spain & Vienna, Austria: JD Systems Institute & WVSA Secretariat, https://doi.org/10.14281/18241.24.

[25] Martin, M. and I. Mullis (eds.) (2012), Methods and procedures in TIMSS and PIRLS 2011, Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Boston College, https://timss.bc.edu/methods/index.html (accessed on 13 June 2024).

[26] Martin, M., M. von Davier and I. Mullis (eds.) (2020), Methods and procedures: TIMSS 2019 Technical Report, Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Boston College, https://timssandpirls.bc.edu/timss2019/methods (accessed on 13 June 2024).

[15] OECD (2024), Education at a Glance 2024 Sources, Methodologies and Technical Notes, OECD Publishing, Paris, https://doi.org/10.1787/e7d20315-en.

[14] OECD (2024), Education at a Glance 2024: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/c00cad36-en.

[6] OECD (2024), Family Database, PF3.2: Enrolment in childcare and pre-school: Definitions and methodology, http://www.oecd.org/els/family/database.htm (accessed on 12 September 2024).

[18] OECD (2024), Income and wealth distribution databases, https://www.oecd.org/en/data/datasets/income-and-wealth-distribution-database.html (accessed on 16 December 2024).

[5] OECD (2024), PISA 2022 Technical Report, PISA, OECD Publishing, Paris, https://doi.org/10.1787/01820d6d-en.

[8] OECD (2024), TaxBEN: The OECD tax-benefit simulation model: Methodology, user guide and policy applications, https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/income-support-redistribution-and-work-incentives/OECD-TaxBEN-methodology-and-manual.pdf (accessed on 16 January 2025).

[13] OECD (2022), Education at a Glance 2022: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/3197152b-en.

[29] OECD (2022), Family Database, PF3.4: Childcare support: Definitions and methodology, http://www.oecd.org/els/family/database.htm (accessed on 12 September 2024).

[3] OECD (2022), Improving Early Equity: From Evidence to Action, OECD Publishing, Paris, https://doi.org/10.1787/6eff314c-en.

[2] OECD (2021), International Early Learning and Child Well-being Study: Technical Report, OECD, Paris, https://www.oecd.org/education/school/early-learning-and-child-well-being-study/Technical%20Report_with%20covers_v1.pdf (accessed on 16 December 2024).

[1] OECD (2019), TALIS Starting Strong 2018 Technical Report, OECD, Paris, https://www.oecd.org/education/talis/TALIS-Starting-Strong-2018-Technical-Report.pdf (accessed on 16 December 2024).

[7] OECD (2019), The OECD Social Expenditure Database (SOCX) Manual, https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/social-spending/socx_manuel_2019.pdf (accessed on 16 December 2024).

[12] OECD (2018), Education at a Glance 2018: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2018-en.

[16] OECD (2018), OECD Handbook for Internationally Comparative Education Statistics 2018: Concepts, Standards, Definitions and Classifications, OECD Publishing, Paris, https://doi.org/10.1787/9789264304444-en.

[11] OECD (2017), Education at a Glance 2017: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2017-en.

[4] OECD (2017), PISA 2015 Technical Report, OECD, Paris, https://www.oecd.org/content/dam/oecd/en/about/programmes/edu/pisa/publications/technical-report/PISA2015_TechRep_Final.pdf (accessed on 6 November 2024).

[10] OECD (2014), Education at a Glance 2014: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2014-en.

[17] OECD (2013), OECD Framework for Statistics on the Distribution of Household Income, Consumption and Wealth, OECD Publishing, Paris, https://doi.org/10.1787/9789264194830-en.

[9] OECD (2012), Education at a Glance 2012: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2012-en.

[27] von Davier, M. et al. (eds.) (2023), Methods and procedures: PIRLS 2021 technical report, Chestnut Hill, MA: TIMSS & PIRLS International Study Center, Boston College, https://pirls2021.org/methods (accessed on 13 June 2024).