As a window into inequalities in early learning and development, this chapter examines the associations between children’s outcomes in IELS 2025 and their gender and socio-economic, immigration and language backgrounds. It also explores the alignment between gaps in assessed outcomes and parents’ and teachers’ perspectives on children’s skills.
Building Strong Foundations for Life
Results from the 2025 Early Learning and Child Well-being Study
Abstract
At age 5, girls outperform boys across the three dimensions of early learning and development measured in IELS 2025. The differences are more pronounced in social and emotional development and less salient in foundational learning. On average across jurisdictions, girls score between 20 and 31 points higher than boys in the domains of emotion identification, emotional attribution, pro-social behaviour and non-disruptive behaviour. Emergent numeracy is the domain where girls’ and boys’ outcomes are most similar. In executive function, differences exist in inhibition and mental flexibility. Gender gaps in favour of girls tend to be larger among children with the weakest performance in the IELS assessments.
IELS 2025 finds pervasive socio-economic gaps in learning and development outcomes by age 5. These gaps are most substantial in foundational learning: socio-economically advantaged children score on average between 60 and 70 points higher than disadvantaged peers in emergent literacy and emergent numeracy. Socio-economic gaps are smaller in executive function and social and emotional development but remain consistent across all domains.
Developmentally resilient children achieve strong early learning and development outcomes despite growing up under unfavourable socio-economic circumstances. On average across jurisdictions, about 15% of socio-economically disadvantaged five-year-olds perform strongly in emergent literacy and emergent numeracy. Across executive function and social and emotional development domains this percentage ranges from 15% to 22%.
Across jurisdictions and assessment domains, gaps in early learning and development between children with and without an immigration background are small compared to other aspects of equity. Differences in outcomes by immigration background are larger and more often statistically significant in foundational learning, but rare in the executive function and social and emotional development. Accounting for socio-economic status and home language reduces both the magnitude and significance of these gaps.
On average, gaps between children whose primary home language is the same or different differs from the language of assessment are also relatively small and not visible in many jurisdictions after children’s socio-economic status is taken into account. They appear mainly in the foundational learning domains of emergent literacy and emergent numeracy, in the executive function domain of mental flexibility, and in the social and emotional development domain of emotion identification.
The perceptions of parents and early childhood education and care (ECEC) or school teachers about five-year-olds’ cognitive and motor skills, socio-emotional skills and global skills generally reflect the same socio-demographic gaps that emerge in children’s assessments. Both parents and teachers perceive girls as having stronger skills than boys, and socio-economically advantaged children as having stronger skills than disadvantaged peers. However, ECEC/school teachers’ perceptions of the skills of children with and without an immigration background sometimes go in the opposite direction to gaps in assessments.
The jurisdictions participating in IELS 2025 vary markedly in their economic development, demographic profiles and social protection systems, resulting in differences in the levels of resources and opportunities that public policies can provide to support young children and families (see Annex A). However, most of the variation in five-year-olds’ early learning and development outcomes lies within rather than between jurisdictions (see Chapter 3). This highlights the importance of children’s individual and family characteristics in shaping the distribution of outcomes across child populations.
This chapter describes how variation in children’s early learning and development differs according to gender, socio-economic status, and immigration and language backgrounds. The chapter considers these factors as central components of children’s social identities (i.e. who children are), while subsequent chapters examine the influences of home environments (i.e. what parents do) (Chapter 5) and early education and care services (i.e. what early education systems do) (Chapter 6).
In this chapter, socio-demographic gaps in early learning and development outcomes are examined through an equity lens. To a significant extent, these gaps are socially conditioned rather than naturally occurring (as, for instance, age-related gaps) and can therefore represent important policy targets. By adopting an equity lens, IELS views early skill development and the distribution of early learning and development outcomes as shaped by social contexts and influenced by “circumstances” that children themselves cannot control, for example, their gender, socio-economic status, race or ethnicity, immigrant background, family structure or place of residence. From this perspective, more equitable early years systems are those that succeed in reducing the influence and minimising any adverse impact of these circumstances on children’s early learning and development.
Gender is an important dimension for understanding children’s early learning and development. Nonetheless, the relationship between a child’s gender and their early skills is neither univocal nor biologically determined. The size and direction of gender differences in skills can change as children grow older, since the relationship is moderated by other factors, chiefly the social environments in which children live (Etchell et al., 2018[1]) (Eliot et al., 2021[2]).
Research suggests that gender-related differences exist in the pace of skill development in early childhood with female children having, on average, stronger early language and communication abilities, as well as an advantage in social and emotional skills, compared to male children, whereas the direction of gaps in motor skills development often vary between fine (in favour of girls) and gross (in favour of boys) (Eriksson et al., 2011[3]) (Rinaldi et al., 2021[4]) (Bando et al., 2024[5]). Yet, while maturational differences between male and female children can drive gaps in their early skill development, the magnitude of these gaps changes across social and cultural contexts, which indicates that other factors are also at play.
Gender-differentiated parenting is another potential influence on gender gaps in early learning and development. The time and activities that mothers and fathers engage in with their children can vary depending on the gender of the child; at the same time, children’s gender-differentiated behaviours can induce parents to treat boys and girls differently, with either of these channels resulting in varying types of interactions, including differential vocalisation, displays of affect, play or compliance (Raley and Bianchi, 2006[6]) (Morawska, 2020[7]). For instance, the fact that girls and boys reach pretend‑play milestones at different times and exhibit differential development in fine and gross motor skills is likely to reflect socio-cultural expectations and practices about how each gender is encouraged to play (Ertem et al., 2018[8]) (Dinkel and Snyder, 2020[9]).
In IELS 2025, the even gender distribution of all jurisdictions’ samples allows robust comparisons regarding the skills development of boys and girls by age 5 (Table B.4.1).
Associations between gender and children’s early learning and development outcomes
Girls outperform boys across the three dimensions of early learning and development measured in IELS 2025, with differences being most pronounced in social and emotional development and least visible in foundational learning (Figure 4.1). On average across jurisdictions, girls’ scores are between 20 and 31 points higher than boys’ scores in emotion identification, emotional attribution, pro-social behaviour and non-disruptive behaviour, and 14 points higher in trust. Compared across jurisdictions, gaps in this dimension tend to be largest among children in England (United Kingdom) and smallest among children in Baku and Sumgait (Azerbaijan). In other jurisdictions, their size varies inconsistently across domains.
Both the prevalence and magnitude of gender gaps are smaller in executive function. Across the three domains in this dimension, the average gap in favour of girls is about 10 points across jurisdictions; however, in both inhibition and working memory, the differences between girls’ and boys’ scores are not statistically significant in half or more of the jurisdictions. Gender gaps in executive function domains are largest among children in Korea and the United Arab Emirates.
In foundational learning domains, statistically significant gaps in emergent literacy are only visible among children in the Netherlands, Ceará, Pará and São Paulo (Brazil) and the United Arab Emirates, with a size of around 10 points, and in England (United Kingdom), where the gap reaches 18 points. Emergent numeracy is the domain in which five-year-old girls and boys have the most similar scores, with significant differences observed only among children in the United Arab Emirates (in favour of girls) and in Korea (in favour of boys).
It is also important to examine gender gaps at the extremes of the performance distribution because the average spread of scores is greater among boys than girls in most of the early learning and development domains assessed in IELS. This can be done by comparing the average scores in the lowest- and highest-performing 10% of girls and boys within each jurisdiction. This comparison shows that gender gaps, always in favour of girls, are generally larger among children with the weakest performance in the IELS assessments. This pattern holds among children in many jurisdictions in the social and emotional development domains of emotion identification and emotional attribution; in the two foundational learning domains of emergent literacy and numeracy; and in the executive function domains of working memory and mental flexibility, and more commonly in the jurisdictions where the average gender gaps are larger. By contrast, across all three dimensions, differences between the highest-performing girls and boys tend to be smaller and less often statistically significant than gaps for those with an average performance (Tables B.4.2, B.4.3 and B.4.4).
Score-point differences between girls and boys in mean scores of foundational learning, executive function, and social and emotional development domains
Note: *Estimates for Brazil correspond to the average across the three participating states. Gaps calculated as girls’ mean scores minus boys’ mean scores. Solid triangles indicate statistically significant differences. For more information, the IELS 2025 Technical Report (OECD, 2026[10]). Jurisdictions are ranked in ascending order of the size of the gap for each domain.
Source: OECD, IELS 2025 Database, Tables B.4.2, B.4.3 and B.4.4.
Parent and ECEC teacher perspectives on the skills of five-year-old girls and boys
Differences in the perceptions of parents and teachers about the skills of male and female 5-year-olds align with the gender gaps observed in their assessment scores (Table 4.1). Across all the jurisdictions in IELS 2025, significant gender differences emerge in parent and ECEC teacher ratings of children’s cognitive and motor skills as well as social and emotional skills: in general, both parents and ECEC teachers perceive girls as having stronger skills than boys. However, differences between boys and girls in the scale of global capabilities (from teachers only) are smaller and not statistically significant in Baku and Sumgait (Azerbaijan), England (United Kingdom), Korea, Malta and the Netherlands. Overall, results suggest a high degree of consistency between the gender gaps observed in children’s assessments and parents’ and teachers’ reports about children’s skills.
Tests of statistically significant differences in indices of parent and ECEC staff perspectives between girls and boys
Note: OECD members are listed in black font. Partner countries/economies are listed in blue font. Sub-national jurisdictions are listed in italics. *Estimates for Brazil correspond to the average across the three participating states. **IELS Adjudicated Region. Blue (grey) shading indicates that girls are perceived as having stronger (weaker) skills than boys. White shading indicates a lack of statistically significant differences. Results based on parent and ECEC staff contextual indices. For more information, see the IELS 2025 Technical Report (OECD, 2026[10]). Consistency ratings based on the number of indices where differences show the same sign as differences in assessments. Jurisdictions are ranked in alphabetical order.
Source: OECD, IELS 2025 Database, Table B.4.5.
Socio-economic status is a concept widely used in social and educational research to capture an individual’s access to family resources (i.e. economic, social and cultural forms of capital) and the relative social and economic position of an individual or household (Cowan et al., 2012[11]) (Avvisati, 2020[12]). Families’ socio-economic status strongly influences children’s health, cognitive and socio-emotional development from before birth into adulthood, through a range of mechanisms that include differences in time and material investments, parenting practices, and stress exposure (Bradley and Corwyn, 2002[13]) (Duncan, Magnuson and Votruba‐Drzal, 2015[14]).
In IELS, a child’s socio-economic background is measured by an index combining information about their parents’ highest levels of education and occupational status and about the household’s income, all collected through the Parent questionnaire. This report relies on the IELS index of socio-economic status derived from Item Response Theory (IRT) (see the Reader’s guide and the IELS 2025 Technical Report (OECD, 2026[10])). The index enables comparisons between children with different socio-economic profiles. In this report, five-year-olds are considered socio-economically advantaged if they are among the 25% of children with the highest values on the index in their jurisdiction, and socio-economically disadvantaged if they are among the 25% with the lowest values.
While the index is centred within each jurisdiction to a mean of zero, its standard deviation can vary and serve as a proxy of the variation in children’s socio-economic conditions within each of the IELS 2025 national samples. According to this measure, the extent of variation in socio-economic status is largest among children in England (United Kingdom), the Flemish Community (Belgium) and Malta, and lowest among children in Korea (Table B.4.1).
Associations between socio-economic background and children’s early learning and development outcomes
In IELS 2025, children’s socio-economic background accounts for a small but non-negligible share of the variation in their early learning and development outcomes. This system-level socio-economic gradient is more pronounced in the foundational learning dimension, accounting for 7% of the variation in emergent literacy and 9% of the variation in emergent numeracy scores, on average across jurisdictions. This association is strongest in England (United Kingdom), the Flemish Community (Belgium) and the Netherlands and lowest in Baku and Sumgait (Azerbaijan) and the United Arab Emirates. In the executive function and social and emotional development dimensions, however, children’s socio-economic background accounts for a more modest share of the variation in outcomes: below 5% in any of the corresponding domains (Tables B.4.6, B.4.7 and B.4.8).
Another relevant equity indicator is the size of the gaps in outcomes between socio-economically advantaged and disadvantaged children within each jurisdiction. Across dimensions of early learning and development, these gaps consistently favour five-year-olds from families with a high socio-economic status (Figure 4.2). IELS 2025 provides unequivocal evidence that differences in children’s learning and development related to socio-economic inequalities are already visible and sizeable by age 5.
Socio-economic gaps are most substantial in foundational learning domains. On average across jurisdictions, socio-economically advantaged children score 61 and 67 points above disadvantaged peers in the emergent literacy and emergent numeracy domains, respectively. These differences are smaller among children in Ceará, Pará and São Paulo (Brazil), and Korea, and larger among children in England (United Kingdom) and the Flemish Community (Belgium). The case of Korea shows that a strong average performance in this dimension is compatible with comparatively small socio-economic gaps.
In the social and emotional development dimension, socio-economic gaps are also smaller than in foundational learning, although relatively large in emotional identification (43 points) compared to the trust, and non-disruptive behaviour domains (in both cases, below 20 score points), on average across jurisdictions. The differences in social and emotional development domains associated with families’ socio-economic position are also generally smaller among children in Baku and Sumgait (Azerbaijan), Ceará, Pará and São Paulo (Brazil) and Korea, and larger among children in the Flemish Community (Belgium).
Overall, the magnitude of socio-economic gaps in early learning and development across jurisdictions mirrors the degree of variation in children’s socio-economic circumstances, as captured by the standard deviation of the IELS index of socio-economic status (Table B.4.1).
Score-point differences in foundational learning, executive function, and social and emotional development domains between socio-economically advantaged and disadvantaged children
Note: *Estimates for Brazil correspond to the average across the three participating states. Gaps calculated as mean scores of socio-economically advantaged children minus mean scores of socio-economically disadvantaged children. Advantaged and disadvantaged children are those in the top and bottom quarters, respectively, of the distribution of the IELS index of socio-economic status (SES) within their jurisdiction. Solid triangles indicate statistically significant differences. For more information, see the IELS 2025 Technical Report (OECD, 2026[10]). Jurisdictions are ranked in ascending order of the size of the gap for each domain.
Source: OECD, IELS 2025 Database, Tables B.4.6, B.4.7 and B.4.8.
Socio-economic gaps tend to be smaller in executive function domains. On average across jurisdictions, differences in scores between socio-economically advantaged and disadvantaged children amount to 46 points in working memory, 37 points in mental flexibility, and 14 points in inhibition, the domain in which these gaps are smallest across the 10 outcomes assessed by IELS.
Parent and ECEC teacher perspectives about the skills of five-year-olds from different socio-economic backgrounds
The perceptions held by parents and ECEC teachers about children’s skills are systematically associated with their socio-economic backgrounds. Across all the jurisdictions in IELS 2025, socio-economically advantaged children are perceived to have stronger skills than disadvantaged peers as measured by significant differences in the cognitive and motor skills, social and emotional skills and global capabilities indices based on adults’ reports (Table 4.2). While parental perceptions can be heavily conditioned by the socio-economic position they share with their own children, this result holds also for ECEC teachers, who generally encounter children from more diverse socio-economic backgrounds as part of their work in ECEC settings. Overall, the results also suggest a strong consistency between children’s assessments and parents’ and teachers’ perspectives regarding differences in the skills of five-year-olds from different socio-economic backgrounds.
Tests of statistically significant differences in indices of parent and ECEC staff perspectives between advantaged and disadvantaged children
Note: OECD members are listed in black font. Partner countries/economies are listed in blue font. Sub-national jurisdictions are listed in italics. *Estimates for Brazil correspond to the average across the three participating states. **IELS Adjudicated Region. Blue (grey) shading indicates that socio-economically advantaged children are perceived as having stronger (weaker) skills than socio-economically disadvantaged children. White shading indicates a lack of statistically significant differences. Results based on parent and ECEC staff contextual indices. For more information, see the IELS 2025 Technical Report (OECD, 2026[10]). Consistency ratings based on the number of indices where differences show the same sign as differences in assessments. Jurisdictions are ranked in alphabetical order.
Source: OECD, IELS 2025 Database, Table B.4.9.
A focus on developmentally resilient five-year-olds
In alignment with other OECD studies, IELS defines developmentally resilient children as those from low socio-economic backgrounds (i.e. in the bottom quarter of the distribution of the index of socio-economic status within their jurisdiction) who nevertheless achieve strong early learning and development outcomes (i.e. whose scores are in the top quarter of the distribution of a domain within their jurisdiction). Given the strong association between socio-economic status and early learning and development, these children are considered developmentally resilient because they achieve better outcomes in IELS than would be predicted by the relatively unfavourable socio-economic conditions in which they grow up. Within jurisdictions, the percentage of resilient children is another indicator of the extent to which early learning and development are associated with socio-economic status: higher levels of resilience among disadvantaged children can be interpreted as a sign of a weaker relationship between children’s circumstances and outcomes. The indicator is also related to the degree of variation in socio-economic conditions within jurisdictions, as captured by the IELS index of socio-economic status.
Percentage of socio-economically disadvantaged children in the top quarter of the performance distribution in foundational learning, executive function and social and emotional development domains in their jurisdiction
Note: *Estimates for Brazil correspond to the average across the three participating states. Developmentally resilient children are those in the bottom quarter of the distribution of the IELS index of socio-economic status (SES) within their jurisdiction, and whose scores are in the top quarter of a domain of assessment within their jurisdiction. For more information, see the IELS 2025 Technical Report (OECD, 2026[10]). Jurisdictions are ranked in ascending order of the percentage of developmentally resilient children in each domain.
Source: OECD, IELS 2025 Database, Tables B.4.6, B.4.7 and B.4.8.
Figure 4.3 shows the percentage of developmentally resilient disadvantaged 5-year-olds in each of the 10 domains of assessment in IELS 2025. In the foundational learning dimension, and on average across jurisdictions, around 15% of socio-economically disadvantaged children perform strongly in emergent literacy and emergent numeracy, with variation of close to five percentage points between the jurisdictions with the highest and lowest shares. In both the executive function and social and emotional dimensions, the percentage of resilient disadvantaged children ranges between 15% to 22% across domains. Reflecting the generally smaller social-economic gaps in scores observed in these jurisdictions, developmental resilience tends to be higher among disadvantaged 5-year-olds in Korea, Malta, Baku and Sumgait (Azerbaijan) and Ceará, Pará and São Paulo (Brazil).
Rising international migration is reshaping the makeup of populations worldwide. In 2020, the global number of international migrants reached 281 million, including 36 million children (UNICEF, 2025[15]). By 2022, 145 million people living in OECD countries were born abroad, an increase of 25% over a decade, representing more than 10% of the population in two‑thirds of these countries (OECD, 2023[16]). Meanwhile, between 2010 and 2024, the global number of children displaced due to conflict and violence nearly tripled, rising from approximately 17.0 million to 48.8 million (UNICEF, 2025[17]) .
These trends are also visible in the composition of child populations participating in ECEC. Data from TALIS Starting Strong 2024 show that children with an immigration background represent more than 10% of the children attending over one quarter of pre-primary settings in 11 out of 15 participating entities. In the Flemish Community (Belgium), immigrant children accounted for over 50% of the children in 1 out of 6 settings (OECD, 2025[18]).
Young children with an immigration background often face a range of structural, linguistic, and socio‑cultural disadvantages that can shape their early educational experiences and long-term developmental outcomes. Immigrant children are less likely to access high‑quality early childhood programs due to barriers such as limited programme availability, insufficient bilingual or culturally responsive services, and a lack of institutional outreach tailored to their families, limiting children’s access to early learning and development opportunities (Karoly and González, 2011[19]). Research shows that immigration-related gaps in later educational outcomes are often rooted in skills gaps in the early years (Washbrook et al., 2012[20]) (Becker and Klein, 2021[21]).
However, within most countries, immigrants tend to be a highly diverse, rather than uniform, population group. Children with an immigration background vary with regard to their countries of origin, cultural and linguistic traditions, socio‑economic circumstances, and the amount of time they have spent in the host country. Intersecting social identities means these children can face a double disadvantage if they are more likely than peers without an immigration background to also experience socio‑economic hardship, compounding the effects of cultural and social barriers (Kruzik et al., 2023[22]). In some cases, though, children with an immigration background come from relatively affluent and educated families. That is why accounting for socio-economic status is critical in examining the relationship between an immigration background and children’s outcomes.
Increasing multilingualism in child populations is one of the major consequences of international migration flows. In some countries, this overlaps with the presence of multiple official languages rooted in longstanding historical and cultural traditions. A limited familiarity with the language used in early education settings can follow for both children with an immigrant background – especially those having arrived recently – and children from local cultural communities where other languages are predominantly spoken.
Given the centrality of language for early learning and development, multilingualism in the early years represents an important consideration for equity. Opportunities for early language development can differ between monolingual and multilingual children, depending partly on how much they are exposed to the majority language in their home environments (Lauro, Core and Hoff, 2020[23]) (Persici et al., 2022[24]). In a shift from deficit- to strength-based approaches to diversity, research increasingly highlights how multilingual children can flexibly draw on their full linguistic repertoire to respond to social and cognitive demands (Bonacina-Pugh, da Costa Cabral and Huang, 2021[25]).
Overall, minimising any potentially adverse impact of children’s immigration or diverse language backgrounds on their early learning and development outcomes is central to promoting equity in the early years and a way of enhancing social cohesion in host communities.
Children with an immigration background and with a different home language
In IELS 2025, children with an immigration background are defined as children having two parents born outside the jurisdiction in which the child participated in IELS (or one parent, in the case where information about only one parent was provided). Among children with an immigration background, a further distinction can be made between first- and second-generation migration, depending on whether only the parents or both the parents and the child were born outside the IELS jurisdiction. Further, children are considered to have a different home language if they primarily speak at home a language other than the language of assessment in IELS. In each participating jurisdiction, the language of assessment was the language of the ECEC centre/school that the child attends1.
Figure 4.4 shows substantial variation across jurisdictions in the shares of children with an immigration background and children with another home language in IELS 2025. At one extreme, almost 60% of the participating children in the United Arab Emirates come from families with international origins, with more than half being first-generation. Other jurisdictions with notably diverse child populations are England (United Kingdom) and the Flemish Community (Belgium), where about one in four and one in five children, respectively, have an immigration background. In both jurisdictions, the majority of these children are second-generation migrants. The most homogeneous child populations in IELS 2025 are those of Korea, Ceará, Pará and São Paulo (Brazil) and Baku and Sumgait (Azerbaijan), where less than 3% of participating five-year-olds have an immigration background.
Similarly, the prevalence of multilingualism among children in IELS 2025 varies widely across jurisdictions. It is highest in Malta and the United Arab Emirates, where 57% and 31% of children, respectively, primarily speak a language other than the language in which they were assessed in IELS at home. By contrast, in Korea and Ceará, Pará and São Paulo (Brazil) virtually all children speak the same language in their homes and their ECEC centres (Figure 4.4). On average across jurisdictions, more than one in four children have two parents whose first language differs from the language of assessment in IELS 2025. In each jurisdiction, this percentage is higher than that of children speaking primarily another language at home, a result that may be explained by the presence of siblings or the adoption, by these children, of the language used in their ECEC centres/schools (Table B.4.1).
Associations between immigration background and children’s early learning and development outcomes
Across jurisdictions, gaps between children with and without an immigration background in the early learning and development outcomes included in IELS 2025 are small compared to other equity gaps, particularly after taking into account children’s socio-economic status and home language. Differences in outcomes by immigration background are larger and more often statistically significant in foundational learning domains, but rare in the executive function and social and emotional development dimensions (Figure 4.5).
In foundational learning, on average across jurisdictions, the mean scores in emergent literacy and emergent numeracy of children with an immigration background are 37 points and 15 points lower, respectively, than those of children without such a background. Once children’s socio-economic and linguistic backgrounds are controlled for, the average adjusted mean difference decreases to 24 points in emergent literacy and disappears in emergent numeracy. These gaps are largest among children in the Flemish Community (Belgium), the Netherlands and England (United Kingdom). In opposition to the general pattern, gaps in emergent numeracy are in favour of children with international origins in Malta and the United Arab Emirates.
In executive function domains, only raw comparisons show differences between five-years-olds with and without an immigration background; these gaps disappear in virtually all cases after other child characteristics are taken into account. In social and emotional development domains, gaps persist in emotion identification and pro-social behaviour but only in a small number of jurisdictions.
Percentage of children with a first- and second-generation immigration background, and percentage of children who primarily speak at home a language other than the language of assessment
Note: *Estimates for Brazil correspond to the average across the three participating states. First-generation immigration background refers to children born outside their IELS jurisdiction; second-generation immigration background refers to children born in their IELS jurisdiction whose parents were born abroad. Home language refers to the language primarily spoken at home by the child. The percentage of children speaking primarily a language other than the language of assessment appears above jurisdictions’ names. For more information, see the IELS 2025 Technical Report (OECD, 2026[10]). Jurisdictions are ranked in ascending order of the percentage of first-generation immigrant children.
Source: OECD, IELS 2025 Database, Table B.4.1.
Score-point differences in foundational learning, executive function, and social and emotional development domains between children with and without an immigration background, before and after accounting for socio-economic status and home language
Note: Ceará, Pará and São Paulo (Brazil) excluded due to the small share of children with an immigration background. Gaps calculated as mean scores of children with an immigration background minus mean scores of children without such background. An immigration background refers to having two parents born outside the jurisdiction in which the child participated in IELS. Children’s socio-economic background as measured by the IELS index of socio-economic status (SES). Home language refers to the language primarily spoken at home by the child. Solid triangles indicate statistically significant differences. For more information, see the IELS 2025 Technical Report (OECD, 2026[10]). Jurisdictions are ranked in ascending order of the size of the gap for each domain.
Source: OECD, IELS 2025 Database, Tables B.4.10, B.4.11 and B.4.12.
Parent and ECEC teacher perspectives about the skills of five-year-olds with and without an immigration background
Contrary to results for gender and socio-economic status, the perceptions of adults concerning the relative level of skills of children with and without an immigration background show are not always aligned with the gaps in IELS 2025 assessments (Table 4.3). When parental views on their children’s skills are compared, non-immigrant children tend to be perceived as having stronger skills than children with an immigration background in most of the jurisdictions with sizeable shares of immigrant children, while the opposite holds in the United Arab Emirates. Parental perceptions are therefore broadly consistent with actual gaps in assessment. However, when this comparison is made based on the views of ECEC teachers, children with an immigration background are generally perceived to have stronger skills than non-migrant children in several jurisdictions, including England (United Kingdom), the Flemish Community (Belgium) and the Netherlands. This is a noteworthy result given that, in these jurisdictions, children with an immigration background tend to have lower scores in several early learning and development domains. This result does not necessarily mean that teachers do not accurately assess the skills of five-year-olds in their ECEC centres/schools. Instead, teachers, who see the learning and developmental progressions of both groups of children, may contextualise their outcomes in relation to children’s immigration status in different ways than parents. For both parents and ECEC staff, differences in perceptions remain largely similar after accounting for children’s socio-economic status and home language.
Tests of statistically significant differences in indices of parent and ECEC staff perspectives between children with and without an immigration background, after accounting for socio-economic status and home language
Note: OECD members are listed in black font. Partner countries/economies are listed in blue font. Sub-national jurisdictions are listed in italics. Ceará, Pará and São Paulo (Brazil) excluded due to small sample sizes. **IELS Adjudicated Region. Blue (grey) shading indicates that children without an immigration background are perceived as having stronger (weaker) skills than children with an immigration background. White shading indicates a lack of statistically significant differences. Results based on parent and ECEC staff contextual indices. For more information, see the IELS 2025 Technical Report (OECD, 2026[10]). Consistency ratings based on the number of indices where differences show the same sign as differences in assessments. Jurisdictions are ranked in alphabetical order.
Source: OECD, IELS 2025 Database, Table B.4.13.
Score-point differences in foundational learning, executive function, and social and emotional development domains between children who primarily speak the language of assessment at home and children who primarily speak another language, before and after accounting for socio-economic status
Note: Ceará, Pará and São Paulo (Brazil) and Korea excluded from this analysis due to small sample sizes. Children’s socio-economic background as measured by the IELS index of socio-economic status (SES). Home language refers to the language primarily spoken at home by the child. Differences calculated as mean scores of children speaking primarily another language at home minus mean scores of children speaking primarily the language of assessment at home. Solid triangles indicate statistically significant differences. For more information, see the IELS 2025 Technical Report (OECD, 2026[10]). Jurisdictions are ranked in ascending order of the size of the gap for each domain.
Source: OECD, IELS 2025 Database, Tables B.4.14, B.4.15 and B.4.16.
Associations between home language and children’s early learning and development outcomes
On average across jurisdictions, the mean scores of children whose primary language at home is different from the language of assessment in IELS 2025 only differ from those of children speaking the same language in the foundational learning domains of emergent literacy and emergent numeracy, and in the social and emotional development domain of emotion identification; in these cases, the score differences range between 13 and 30 points, after accounting for children’s socio-economic status (Figure 4.6). However, few statistically significant differences are observed between the two groups across the remaining domains of assessment. Gaps in favour of children with a different home language are observed only among children in Baku and Sumgait (Azerbaijan) and Malta in some domains, for instance mental flexibility. Unsurprisingly, gaps by home language tend to be largest in emergent literacy, with a 48 point or larger advantage in favour of children speaking the same language at home and in their ECEC centre among children in England (United Kingdom), the Flemish Community (Belgium), and the Netherlands.
References
[12] Avvisati, F. (2020), “The measure of socio-economic status in PISA: a review and some suggested improvements”, Large-scale Assessments in Education, Vol. 8/1, https://doi.org/10.1186/s40536-020-00086-x.
[5] Bando, R. et al. (2024), “Gender Differences in Early Child Development: Evidence from Large-Scale Studies of Very Young Children in Nine Countries”, Journal of Economics, Race, and Policy, Vol. 7/2, pp. 82-92, https://doi.org/10.1007/s41996-023-00131-1.
[21] Becker, B. and O. Klein (2021), “The primary effect of ethnic origin – rooted in early childhood? An analysis of the educational disadvantages of Turkish-origin children during the transition to secondary education in Germany”, Research in Social Stratification and Mobility, Vol. 75, p. 100639, https://doi.org/10.1016/j.rssm.2021.100639.
[25] Bonacina-Pugh, F., I. da Costa Cabral and J. Huang (2021), “Translanguaging in education”, Language Teaching, Vol. 54/4, pp. 439-471, https://doi.org/10.1017/s0261444821000173.
[13] Bradley, R. and R. Corwyn (2002), “Socioeconomic Status and Child Development”, Annual Review of Psychology, Vol. 53/1, pp. 371-399, https://doi.org/10.1146/annurev.psych.53.100901.135233.
[11] Cowan, C. et al. (2012), Improving the Measurement of Socioeconomic Status for the National Assessment of Educational Progress, National Center for Education Statistics, https://nces.ed.gov/nationsreportcard/pdf/researchcenter/socioeconomic_factors.pdf (accessed on 15 January 2026).
[9] Dinkel, D. and K. Snyder (2020), “Exploring gender differences in infant motor development related to parent’s promotion of play”, Infant Behavior and Development, Vol. 59, p. 101440, https://doi.org/10.1016/j.infbeh.2020.101440.
[14] Duncan, G., K. Magnuson and E. Votruba‐Drzal (2015), Children and Socioeconomic Status, Wiley, https://doi.org/10.1002/9781118963418.childpsy414.
[2] Eliot, L. et al. (2021), “Dump the “dimorphism”: Comprehensive synthesis of human brain studies reveals few male-female differences beyond size”, Neuroscience & Biobehavioral Reviews, Vol. 125, pp. 667-697, https://doi.org/10.1016/j.neubiorev.2021.02.026.
[3] Eriksson, M. et al. (2011), “Differences between girls and boys in emerging language skills: Evidence from 10 language communities”, British Journal of Developmental Psychology, Vol. 30/2, pp. 326-343, https://doi.org/10.1111/j.2044-835x.2011.02042.x.
[8] Ertem, I. et al. (2018), “Similarities and differences in child development from birth to age 3 years by sex and across four countries: a cross-sectional, observational study”, The Lancet Global Health, Vol. 6/3, pp. e279-e291, https://doi.org/10.1016/s2214-109x(18)30003-2.
[1] Etchell, A. et al. (2018), “A systematic literature review of sex differences in childhood language and brain development”, Neuropsychologia, Vol. 114, pp. 19-31, https://doi.org/10.1016/j.neuropsychologia.2018.04.011.
[19] Karoly, L. and G. González (2011), “Early Care and Education for Children in Immigrant Families”, The Future of Children, Vol. 21/1, pp. 71-101, https://futureofchildren.princeton.edu/sites/g/files/toruqf2411/files/media/immigrant_children_21_01_fulljournal.pdf (accessed on 12 January 2026).
[22] Kruzik, C. et al. (2023), “The Early Emergence of SES Achievement Gaps: Disparities Across Race, Ethnicity, and Immigrant Status”, Race and Social Problems, Vol. 16/1, pp. 116-132, https://doi.org/10.1007/s12552-023-09402-w.
[23] Lauro, J., C. Core and E. Hoff (2020), “Explaining Individual Differences in Trajectories of Simultaneous Bilingual Development: Contributions of Child and Environmental Factors”, Child Development, Vol. 91/6, pp. 2063-2082, https://doi.org/10.1111/cdev.13409.
[7] Morawska, A. (2020), “The Effects of Gendered Parenting on Child Development Outcomes: A Systematic Review”, Clinical Child and Family Psychology Review, Vol. 23/4, pp. 553-576, https://doi.org/10.1007/s10567-020-00321-5.
[10] OECD (2026), IELS 2025 Technical Report, http://oecd.org/en/about/projects/international-early-learning-and-child-well-being-study.
[18] OECD (2025), Results from TALIS Starting Strong 2024: Strengthening Early Childhood Education and Care, TALIS, OECD Publishing, Paris, https://doi.org/10.1787/20af08c0-en.
[16] OECD (2023), International Migration Outlook 2023, OECD Publishing, Paris, https://doi.org/10.1787/b0f40584-en.
[24] Persici, V. et al. (2022), “Vocabulary and reading speed in the majority language are affected by maternal language proficiency and language exposure at home: a study of language minority bilingual children in Italy”, International Journal of Bilingual Education and Bilingualism, Vol. 25/10, pp. 3729-3744, https://doi.org/10.1080/13670050.2022.2076552.
[6] Raley, S. and S. Bianchi (2006), “Sons, Daughters, and Family Processes: Does Gender of Children Matter?”, Annual Review of Sociology, Vol. 32/1, pp. 401-421, https://doi.org/10.1146/annurev.soc.32.061604.123106.
[4] Rinaldi, P. et al. (2021), “Gender differences in early stages of language development. Some evidence and possible explanations”, Journal of Neuroscience Research, Vol. 101/5, pp. 643-653, https://doi.org/10.1002/jnr.24914.
[17] UNICEF (2025), Child displacement, https://data.unicef.org/topic/child-migration-and-displacement/migration/ (accessed on 3 February 2026).
[15] UNICEF (2025), International Migrant Stock 2020, https://data.unicef.org/topic/child-migration-and-displacement/migration/ (accessed on 3 February 2026).
[20] Washbrook, E. et al. (2012), “The Development of Young Children of Immigrants in Australia, Canada, the United Kingdom, and the United States”, Child Development, Vol. 83/5, pp. 1591-1607, https://doi.org/10.1111/j.1467-8624.2012.01796.x.
← 1. The languages of assessment in each jurisdiction were: Azerbaijani and Russian, depending on the ECEC centre/school, in Baku and Sumgait (Azerbaijan); Portuguese in Ceará, Pará and São Paulo (Brazil); English in England (United Kingdom); Dutch in the Netherlands and the Flemish Community (Belgium); Arabic and English, depending on the ECEC centre/school, in the United Arab Emirates and Abu Dhabi (as an IELS adjudicated region); Maltese and English, depending on the ECEC centre/school, in Malta; Mandaring in Hangzhou (China), and Korean in Korea.