The role of families in the educational outcomes of children and young people

Foreword

Families are the foundation of society. They play a decisive role in shaping early childhood environments, and through this they enable young children to develop the cognitive and non-cognitive skills which will influence their lifetime outcomes.1

They are key in establishing aspirations, providing the basis on which children engage with learning at school. And they play a critical role in providing the cultural capital and occupational knowledge that children need as they become adults and
enter the labour market.

Too often, social mobility narratives characterise families in terms of deficits, and efforts are guided by the principle that your background should not determine your outcomes. In practice, however, family plays one of the most important roles in
shaping life trajectories.2 Taken too far, this way of thinking creates a crude picture where ‘non-disadvantaged’ families are seen as all privileged and bestowing unfair advantages on their children, while the children of ‘disadvantaged’ families are seen
as a homogeneous group at risk of poor outcomes, whose families are a hindrance from which they need to escape.

It is true that families are a source of inequality. Their influence is both profound and quantifiable. Differences in early home learning environments have been shown to explain around 10% of the variation in educational outcomes between children
from high- and low-income households throughout their schooling. Parental education, financial resources and household stability all play roles in shaping outcomes for children, and beyond material wealth, a parent’s beliefs, the high expectations and the strength of their social networks sit at the very heart of whether a child can successfully navigate the ladder of opportunity.

However, it is important to recognise that no individual family is average, and the data can hide big differences between families that look similar on paper. Data shows, for example, that children from poorer families are less likely to achieve good GCSE results than their better-off peers: 26% of disadvantaged pupils and 53% of all other pupils achieved a grade 5 and above in both English and maths in the 2023 to 2024 academic year.3  So while 74% of families who were disadvantaged did not achieve those grades, neither did 47% of those who were not disadvantaged. Material circumstances and grades are correlated, not directly
causal.

This is because there are a wide range of factors to consider, above and beyond income. This is supported by the findings of a recent randomised controlled trial, which shows that cash transfers alone do very little to change the outcomes for young children. There is a complex set of circumstances at play, and we need to understand them if we want to help all families provide the best possible start in life for their offspring.

We should also recognise that ‘providing the best start’ or the concept of social mobility might mean very different things to different people. The Commission’s previous research on perceptions and qualitative deep dives of social mobility show that some members of the general public see it as linked to social connectedness and place rather than just occupation or income. While social mobility champions may view families as core to their purpose, families themselves do not necessarily share this perspective. For most, social mobility is not their defining raison d’etre.

Failing to acknowledge this can inadvertently fuel a ‘perfect family’ model against which real-world families are measured and found wanting. This perspective risks framing policy primarily as an intervention to fill a perceived deficit, trying to deliver outcomes for the child that parents are assumed to be unable to provide. This narrow focus treats the family unit as a hurdle to be cleared rather than a foundation to be supported.

There are certainly cases where direct intervention is necessary, but we are not always clear on where that line should be drawn. We must avoid lazy caricatures of families that function well but do not fit the traditional social mobility narrative, such as working-class parents passing on valuable trade skills rather than preparing a child for university. These instances of stable, supportive parenting are fundamentally different from environments where outcomes are consistently poor and support is genuinely absent.

The challenge is to develop an approach to family that recognises and celebrates its fundamental role while also acknowledging differences in values and outlook, and that is also confident about when and why families are not delivering. This kind of balanced approach is currently conspicuous by its absence.

In policy terms, successive governments have struggled with ‘family’. There has been no Cabinet level Secretary focused on families since 2010, and no dedicated policy around actively promoting good family environments. Furthermore, the closest thing we have to a full, rounded assessment of families during the last quarter-century is Frank Field’s Independent Review of Poverty and Life Chances in 2010. No government has had a coherent strategy dedicated to families in and of themselves.

It is worth reflecting on why the role of the family has often been overlooked in broader policy discussions. This hesitation usually stems from the personal nature of the subject; there is a natural caution to avoid being judgmental, leaning into stereotypes or overlooking the challenging circumstances that families face. These are important sensitivities to acknowledge. However, they should not prevent a clear-eyed analysis of the evidence. To truly improve outcomes, we must discuss the complexities of family life with the same balance of empathy and rigour that we apply to any other important social issue.

There is sufficient evidence to start piecing together a proper analysis of the changes in family life in the United Kingdom, and to connect them to the wider social, economic and cultural challenges facing the country. We know, for example, that there are substantial changes in family formation, as well as significant differences based on social background and education. These shifts correlate with regional disparities and are a particular feature of the former industrial regions of the country that have been ‘left behind’ for decades following deindustrialisation. This is consistent with the Commission’s recent findings from the 2025 State of the Nation report, outlining similar areas entrenched or declining into disadvantage across conditions of childhood and labour market opportunities for young people. This evidence identifies a cluster of issues that correlate with one another. What we now need is to build explanations which cut beyond correlations to identify causes.

We clearly need to do more work to understand this and to devise better policy solutions – or target those which we have in a more focused direction. To begin with, however, it is useful to collate what we currently know about families and the role they play in shaping outcomes for children. This is what our report on “The role of Families in the Educational Outcomes of Children and Young People”, sets out to achieve.

The report presents a synthesis of the best available evidence on how families shape children’s educational outcomes from early childhood through adolescence. It starts from the well-established finding that socio-economic gaps in attainment emerge early and widen over time, but it argues that focusing on economic circumstances alone is too narrow. Instead, it examines a wide range of family-related factors and separates what we know causes outcomes from what is only linked to them.

The strongest causal influences identified are parental education (especially mothers’ education) and the quality of the home learning environment, both of which are shown to raise children’s cognitive and socio-emotional outcomes and to contribute directly to reducing educational inequalities. While there are different explanations which outline how this causal link works, we need further work to understand this more precisely.

Beyond these causal mechanisms, the report shows that there are many other aspects of family life that matter but for which causality is harder to establish. Following prior attainment, parental aspirations and expectations are the single strongest factor explaining differences in GCSE results at age 16 between young people from different socio-economic backgrounds and ethnicities. Support from grandparents, access to trusted non-parental adults, diverse social networks and participation in extra-curricular activities are all positively associated with better educational and later-life outcomes, particularly for disadvantaged children. These factors show the subtle realities of family life that one type of data or source can miss.

By contrast, commonly cited factors such as family structure, birth order and family size have indirect rather than direct effects once income, parental education, stress and mental health are taken into account. Apparent disadvantages associated with lone parenthood, larger families or later position in the birth order largely reflect these deeper constraints rather than family form itself. There is a further debate to be had about these aspects. For example, it may be that income parental education, stress and mental health are the four most significant influences on outcomes; however, family structure and size can have a profound impact on these very factors. Indeed, the evidence suggests markedly different patterns of family formation between the average graduate family and the average non-graduate family. Graduates have fewer children, they have them at an older age after first establishing economic security and they are less likely to be non-partnered.

There is clearly a great deal of further work to do, especially if our call for a balanced and rounded strategy for families is to become a realistic policy ambition. As a starting point, however, we should recognise the central message of this report. It argues for moving away from deficit-based views of families and towards an evidence-based, life-course approach. Families are not simply the source of inequality but a key part of the solution: supportive parenting, high aspirations, and rich social and community networks can buffer disadvantage at multiple stages of development.

Effective policy must combine early intervention, support for parenting and home learning, and strong school and community environments. However, it must also address the context of family formation and how this differs for different socio-economic groups in different places.

To move the needle on social mobility, we must pay more attention to the conditions that enable families to function at their best. This includes recognising that raising children takes a wider ‘village’ of support. Beyond the immediate household, wider social networks and the support of the extended family, particularly the role of grandparents, provide the critical framework of stability and resilience that children need to thrive. However, we must be willing to discuss the structure and stability. This is particularly important because the evidence suggests that as family formation patterns have changed over recent decades, understanding these shifts may offer greater insight into the factors driving poor outcomes, child poverty and intergenerational underachievement for the truly disadvantaged.

Alun Francis

Chair, Social Mobility Commission

1 J. Garcia and J. Heckman (2023), Parenting promotes social mobility within and across generations.

2 K. Kiernan and others (2026), The IFS Deaton Review: Families and inequalities.

3 SMC (2025), Attainment at age 16.

4  K. Noble and others (2025), The effect of a monthly unconditional cash transfer on children’s development at four years of age: A randomized controlled trial in the U.S.

5 SMC (2025), Perceptions of social mobility in the UK: Unpacking public attitudes and evolving aspirations; SMC (2025), Deep dives: A local perspective on social mobility. 

Executive summary

Inequalities in educational outcomes between children from different socio-economic backgrounds emerge early in life and persist throughout schooling, with long-term consequences for opportunities in later education, employment and wellbeing. Addressing these inequalities requires an understanding of the many factors shaping children’s development, among which families play a uniquely central role.

Drawing on extensive evidence, this report examines the role of families in shaping the educational outcomes of children and young people across the life course. Specifically, it focuses on non-economic inequalities between families, inequalities within families, positive intervention by parents and caregivers, and the contributions of wider support networks.

The report draws on a range of quantitative evidence, with the strongest able to identify causal relationships between the factors of interest and children’s educational outcomes. Causal evidence provides a high degree of certainty that changing these factors would lead to changes in children’s educational outcomes and potentially help reduce differences across families. Other types of evidence, meanwhile, are typically only able to identify associations between these factors and children’s educational outcomes. This means that we can be less sure that changing them would necessarily change children’s outcomes.

We classify the factors explored in this report into three groups in this summary:

1. those for which there is a clear causal relationship
2. those for which there is good evidence of positive associations, after accounting for other factors that differ between children growing up in different families
3. those for which there is little or no evidence of positive associations after accounting for other factors, or which are unlikely to contribute significantly to differences in outcomes between children growing up in different families

Key findings

Factors that are causally related to children’s outcomes

Parental education: Children whose parents, particularly mothers, have at least GCSE-level (or equivalent) qualifications have significantly higher test scores than children whose parents have no qualifications. This causal relationship is explained by increases in family resources, such as higher levels of maternal employment and home ownership, improved partner matching, and greater household income, as well as healthier behaviours during pregnancy and better access to educational materials at home.

Parent behaviours: The home learning environment is positively associated with children’s educational outcomes throughout primary and secondary school, over and above socio-economic circumstances. Targeted interventions that improve the quality of the home learning environment have been shown to lead to significantly higher maths and literacy skills.

Factors that are positively associated with children’s outcomes

Parent beliefs and aspirations: Parental aspirations and expectations have the largest association with children’s educational outcomes, compared with other forms of parental involvement such as participation in school activities, supporting learning at home and parent–child academic discussions. High parental aspirations are the single strongest factor mediating differences in GCSE results at age 16 between young people from different socio-economic backgrounds and ethnicities.

Extended family: Some evidence shows that support from grandparents, through caregiving, financial support and the intergenerational transmission of socio-economic advantage, is positively associated with children’s educational outcomes. These relationships are often independent of parental factors.

Social networks and communities: Evidence shows that children from low-income families tend to have better adult outcomes when they grow up in communities where friendships are more socially diverse. Friendship with higher-income peers is associated with higher earnings, greater happiness and trust, and lower loneliness in adulthood.

Extra-curricular activities: Participation in extra-curricular or ‘positive’ activities is associated with higher attainment and a reduced likelihood of being NEET (not in education, employment or training), even after controlling for earlier academic performance. Some evidence suggests that these positive effects may arise because extra-curricular opportunities allow children to engage with skilled non-parental adults who can fulfil a mentoring role, which is positively associated with educational, mental health and social outcomes, particularly in adolescence.

Children and young people with limited support networks: Children and young people in contact with social care have persistently lower attainment than their peers. Some evidence suggests that fostering a sense of belonging and connectedness to school is positively associated with better educational outcomes for those experiencing challenging social circumstances.

Factors that are unlikely to be causally related to children’s outcomes or that are unlikely to explain differences in outcomes across families

Family structure: Children born to married couples often appear to have stronger cognitive and socio-emotional outcomes than children born to cohabiting couples or single parents. However, these differences are not driven by family structures. Instead, other factors such as differences in family resources and income, parental education and parental mental health between different family types underpin the differences in children’s educational outcomes by family structure.

Birth order: First-born children and only children appear to have better cognitive scores than later-born children. However, these differences are not observed when family circumstances such as socio-economic background and parental separation are considered. Instead, parental investments, resource availability and structural disadvantage, which differ by birth order, underpin the apparent relationship between birth order and children’s educational outcomes.

Family size: There is limited reliable evidence on the direct relationship between family size and children’s educational outcomes. However, there is some evidence that larger families, particularly, those from lower socio-economic backgrounds report greater financial strain and stress than smaller families or larger families from higher socio-economic backgrounds.

Month of birth: Children born at the end of the academic year (July to August in England) have lower scores on national attainment tests than their peers born at the start of the academic year (September to October). This attainment gap largely reflects developmental maturity at the time of the test. While month of birth is causally related to children’s educational outcomes, it does not play an important role in explaining differences in outcomes between children growing up in different families, because month of birth does not vary systematically by socioeconomic status.

Summary

Overall, this evidence highlights the value of recognising how families and wider social support networks operate within multiple, interacting layers of influence on children and young people’s cognitive and socioemotional development. As highlighted above, only parental education and the quality of the home learning environment were identified as being causally related to children’s educational outcomes. Many other family-related factors, including parental aspirations, relationships with grandparents and trusted non-parental adults, diverse social networks, and supportive school environments, were found to be positively associated with children’s outcomes, after accounting for socio-economic differences across families. However, their relative importance to each other appears to shift as young children progress into adolescence and adulthood. For example, parents and grandparents play a central role in early childhood. Although they remain important in adolescence, peer relationships and wider social networks also become increasingly significant. Nevertheless, the best available evidence emphasises that the family remains central to promoting educational success and social mobility throughout childhood and adolescence.

Introduction

Inequalities in educational outcomes between children from different socio-economic backgrounds emerge early in life. For example, four- and five-year-olds eligible for free school meals (FSM) are, on average, 4.6 months behind their peers in their cognitive and socioemotional development. This has a lasting influence on their later educational outcomes. Research shows that one in four children who were below expected levels at age five did not achieve a pass in English and maths GCSEs at age 16, compared to just one in ten among those who reached a good level of development at the same age.

While there is value in focusing on early intervention to address educational inequalities, it is vital that efforts also take a life-course approach. This is because educational inequalities that appear in early childhood significantly increase as children get older. For example, cross-sectional data shows that the attainment gap between children eligible for FSM and their peers more than doubles to 10.3 months by the end of Key Stage 2 at age 11 and reaches 19.2 months by Key Stage 4 at age 16. Studies suggest that the transition to secondary school is a particularly sensitive period, with high-achieving economically disadvantaged children more likely than their better-off peers to experience a decline in their attainment as well as their attitudes towards school, behaviour and mental health. Of course, not all young people’s learning progresses at the same rate. Indeed, studies have shown that children from ethnic minority backgrounds tend to make faster progress in school than their White British peers, as do those born at the end of the academic year (summer-born children in England). However, on average, the gaps between children from different socio-economic backgrounds widen over time. Children from economically disadvantaged backgrounds are also more likely than their peers to encounter barriers when accessing higher education and employment opportunities, even when they achieve similar levels of attainment.

12 FSM eligibility is a widely used indicator of economic disadvantage.

13 Tuckett and others (2024), Annual report 2024: Disadvantage; ‘socio-emotional development’ refers to children and young people’s ability to form positive relationships with adults and their peers, manage and express their emotions, and regulate their behaviour and attention in socially and culturally appropriate ways; T.G. Halle and K.E. Darling-Churchill (2016), Review of measures of social and emotional development

14 Elliot Major and S. Parsons (2022), The forgotten fifth: Examining the early education trajectories of teenagers who fall below the expected standards in GCSE English language and maths examinations at age 16, CLS Working Papers 2022/6, London: Centre for Longitudinal Studies (CLS), UCL.

15 Carneiro and others (2024), The short- and medium-term impacts of Sure Start on educational outcomes; S. Cattan and others (2021), The health impacts of Sure Start.

16 A. Howard-Jones and others (2012), The timing of educational investment: A neuroscientific perspective; L. Macmillan and L.A. Outhwaite (2025), What are the evidence-based ways to equalise opportunities?

17 Crawford and others (2017), When and why do initially high-achieving poor children fall behind?; C. Farquharson and others (2024), Education inequalities.

18 Tuckett and others (2024), Annual report 2024: Disadvantage.

19 Crawford and others (2014), Progress made by high-attaining children from disadvantaged backgrounds.

20 Jerrim and M.P. Carvajal (2025), What happens to bright 5-year-olds from poor backgrounds? Longitudinal evidence from the Millennium Cohort Study.

21 Farquharson and others (2024), Education inequalities.

22 Crawford and others (2011), Does when you are born matter? The impact of month of birth on children’s cognitive and non-cognitive skills in England.

23 Campbell and others (2022), Matching in the dark? Inequalities in student to degree match; C. Dilnot and others (2025), Inequalities in access to professional occupations; P. Martin (2024), Do participants in widening participation outreach programmes in England progress to selective universities at a higher rate than would otherwise be expected?

The importance of families and wider social networks

Although indicators of economic disadvantage, such as FSM eligibility, are widely used in research, policy, and practice to understand educational inequalities, they may not fully capture the broad range of factors shaping children’s experiences and outcomes. To understand and address educational inequalities across the life course, it is essential to consider wider perspectives that reflect the multiple dimensions of children’s developmental contexts and the central role that families play within them.

Children’s development, including their educational outcomes, is shaped by many layers of influence, including their family, school, peers, community and wider society. These influences interact with each other and can change in relative importance as children get older. For example, the family plays a central role in early childhood. However, although family remains important in adolescence, peer relationships and wider social networks also become increasingly significant.

This report synthesises the best available evidence on different aspects of family influence, including non-economic inequalities between families, inequalities within families, positive intervention by parents and caregivers, and the contributions of wider support networks.

24 Campbell and others (2025), Who has been registered for free school meals and pupil premium in the National Pupil Database?; L.A. Outhwaite (2025), Understanding early inequalities: Multiple dimensions of children’s developmental contexts predict age 3 outcomes.

25 Bronfenbrenner (1977), Toward an experimental ecology of human development.

26 Sameroff (2010), A unified theory of development: A dialectic integration of nature and nurture.

Non-economic inequalities between families

Parental education

Parental education is one of the most consistent predictors of children’s educational outcomes. Large-scale longitudinal data from the British Cohort Study (children born in 1970) and the Millennium Cohort Study (MCS) (children born in 2000 to 2001) shows that parental education is the strongest predictor of children’s cognitive and language outcomes between ages 3 and 14, closely followed by income, occupation and social class. Crucially, although parental education is often interrelated with income, occupation and social class, it continues to be the strongest predictor of children’s outcomes when all these other factors are considered together. These patterns have remained broadly stable across generations in the UK and are consistent with evidence from other countries. For example, studies in the US show that parental education in the first two years of a child’s life is strongly associated with children’s attainment and socioemotional outcomes throughout school and into adulthood.

While studies often differ in how they measure parental education, the relationships with children’s educational outcomes are consistent and well-established. For example, many studies use a linear approach, whereby each additional level of parental education, ranging from GCSEs to postgraduate university qualifications, is incrementally associated with better child outcomes. Other studies adopt a binary approach, often contrasting ‘low or no qualifications’ with ‘higher qualifications’. For example, a 2024 study also used data from the MCS and shows that four- and five-year-olds whose mothers have low educational qualifications, such as GCSEs or below, have significantly lower cognitive outcomes and higher socioemotional difficulties on average than those whose mothers have a university degree. These differences have also been observed in more contemporary data from 2012 and 2022. Using a nationally representative sample, a 2025 study found that by age three, children whose parents have GCSEs or below are, on average, two months behind in their early developmental outcomes compared with those whose parents have A-level qualifications or higher. Later in life, 55% of individuals with at least one parent with a degree graduate from higher education themselves by age 25, compared with 30% among those for whom neither parent has a degree. Importantly, UK-based longitudinal studies have established that the relationship between parents’, particularly mothers’, educational qualifications and children’s attainment is causal. Evidence from these studies shows that having at least GCSE-level (or equivalent) qualifications causes higher test scores among children, compared with the outcomes of children whose parents have no formal qualifications.

Researchers propose that the link between parental education and children’s outcomes operates through multiple pathways. Specifically, higher levels of parental education underpin income and occupation status, as well as influencing a wide range of parenting behaviours and investments which impact on children’s cognitive, socioemotional, and behavioural development. Two key theoretical models help to explain how these processes work in practice. First, the family investment model proposes that higher levels of parental education and income enable families to provide more cognitively enriching resources and stimulating experiences for their children. Meanwhile, the family stress model suggests that lower levels of parental education and income can increase financial strain, leading to greater parental stress and reduced wellbeing, which can undermine parents’ ability to provide sensitive, responsive care and a nurturing home environment.

Among these pathways, the family investment model tends to receive stronger empirical support, with parental education being the driving factor. For example, a study using large-scale longitudinal cohort data covering the UK’s 1972 compulsory schooling reform compared children’s outcomes for mothers who could leave school at age 15 without any qualifications (i.e. mothers born in or before August 1957) with those for children whose mothers were required to stay in education until age 16 and could obtain some formal qualifications (i.e. mothers born in or after September 1957). It found that raising maternal education from no qualifications to a basic level is causally related to significant long-term improvements in children’s cognitive and socioemotional outcomes measured at the ages of four to seven. These improvements are largely driven by increases in family resources, such as higher maternal employment and home ownership, improved partner matching (i.e., assortative mating patterns), and greater household income, alongside healthier behaviours during pregnancy and better access to educational materials at home. Although the reform did not significantly change the amount of time parents spent with their children, the study demonstrates that even modest gains in maternal education can lead to meaningful intergenerational benefits.

27 P.E. Davis-Kean and others (2021), The role of parent educational attainment in parenting and children’s development.

28 E. Thornton and others (2024), Investigating how vocabulary relates to different dimensions of family socio-economic circumstance across developmental and historical time; A. Sullivan and others (2013), Social class and inequalities in early cognitive scores.

29 E. Thornton and others (2024), Investigating how vocabulary relates to different dimensions of family socio-economic circumstance across developmental and historical time; Q. Song and others (2025), Effects of family socioeconomic status on educational outcomes in primary and secondary education: A systematic review of the causal evidence.

30 S.A. Schmitt and others (2023), Early cumulative risk and outcomes in adolescence and adulthood: The role of executive function and behavioral regulation; N.E. Waters and others (2025), Socioeconomic status and academic achievement: Developmental pathways through parenting and children’s executive functions.

31 For example, E. Thornton and others (2024), Investigating how vocabulary relates to different dimensions of family socio-economic circumstance across developmental and historical time; A. Sullivan and others (2013), Social class and inequalities in early cognitive scores.

32 S. Cattan and others (2024), Early childhood inequalities.

33 L.A. Outhwaite (2025), Understanding early inequalities: Multiple dimensions of children’s developmental contexts predict age 3 outcomes.

34  M. Henderson and others (2020), Moving on up: ‘First in family’ university graduates in England.

35 H. Chowdry and others (2009), Drivers and barriers to educational success-evidence from the Longitudinal Study of Young People in England; M. Dickson and others (2016), Early, late or never? When does parental education impact child outcomes?

36 P.E. Davis-Kean and others (2021), The role of parent educational attainment in parenting and children’s development.

37 R. Haveman and B. Wolfe (1994), Succeeding generations: On the effects of investments in children, New York: Russell Sage Foundation.

38 R.D. Conger and G.H. Elder, Jr (1994), Families in troubled times: Adapting to change in rural America, Hawthorne, NY: Aldine de Gruyter.

39 P.E. Davis-Kean and others (2021), The role of parent educational attainment in parenting and children’s development.

40 L. Macmillan and E. Tominey (2023), Parental inputs and socio-economic gaps in early child development.

Family size

There are some differences in family size in the UK based on parental education. Among women born between 1940 and 1969, those with degrees have the smallest completed family sizes (average 1.5 children), compared with women with lower or no qualifications (average three children). These differences appear to be driven, in part, by later childbearing and higher rates of childlessness among degree-educated women. However, more recent data suggests that the overall decline in birth rates in England and Wales, from 1.94 children in 2012 to 1.41 in 2024, has occurred similarly across all socio-economic backgrounds. There is some regional variation in these trends: the West Midlands and London are the only regions where the birth rates have slightly increased (1.56 in 2023 to 1.59 in 2024 and 1.33 in 2023 to 1.35 in 2024, respectively).

Research on family size aims to understand whether having more children affects each child’s educational outcomes. Resource dilution theory suggests that as families grow, parents may need to spread their time, money and resources more thinly across their children, which can limit the opportunities and support available to each child. Several correlational studies appear to support this theory. An increase in the number of siblings is associated with a decrease in early cognitive and language outcomes. for the youngest children within the family, between the ages of two and six. Some evidence also suggests that these effects typically only occur when the older sibling is male. Children with an older sister and first-born children have been shown to have comparable language outcomes. However, estimating the effects of family size is not straightforward because family size often reflects parents’ choices, values and circumstances, which can also influence children’s outcomes. In the UK, policies such as the two-child limit have added another dimension to how family size may affect children’s outcomes. Introduced in April 2017, the two-child limit means that families can claim means-tested benefits only for up to two children, with no additional support for any third or subsequent children born after this date. The policy was removed in April 2026.

Evidence suggests that the two-child limit had very small, non-significant effects on fertility rates among low-income families. This contrasts with earlier studies showing that the expansion of more generous benefits, specifically the tax credit system introduced in 1999, was associated with increased births among partnered women from low-income backgrounds. To explain these asymmetrical differences, Reader and colleagues (2025) highlight qualitative evidence which suggests that many low-income families were unaware of the policy changes and how it may affect them, as well as a disconnect between policy narratives and people’s decisions to have more children.

Evidence of the impact of the policy on children’s outcomes is mixed. Some research shows that the two-child limit policy, alongside the benefit cap, deepened financial hardship for larger families, many of whom are already at a disproportionate risk of experiencing poverty due to greater household needs and lower work intensity. It is argued that this financial strain makes it harder for parents to invest in children’s basic needs, learning and wellbeing. Qualitative studies drawing on the family investment model and the family stress model suggested that by exacerbating economic disadvantage, the two-child limit increased parental stress and reduced the resources and stability children need to thrive, thereby compounding the risks highlighted by resource dilution theory.

In contrast, recent quantitative evidence shows that the policy had no statistically significant impact on the proportion of children achieving a ‘good level of development’ on the Early Years Foundation Stage Profile (EYFS-P), a teacher-based school readiness assessment completed at school when children are four to five years old. However, the absence of an observed effect on this specific educational measure does not imply that the policy had no wider consequences. For example, the impact on children’s health, wellbeing and home learning environments were not evaluated.

41 A. Berrington and others (2015), Educational differences in timing and quantum of childbearing in Britain: A study of cohorts born 1940–1969.

42 ONS (2013), Births in England and Wales: 2012; ONS (2025), Births in England and Wales: 2024 (refreshed populations).

43 J. Ermisch (2024), The recent decline in period fertility in England and Wales: Differences associated with family background and intergenerational educational mobility.

44 ONS (2025), Births in England and Wales: 2024 (refreshed populations).

45 J. Blake (1981), Family size and the quality of children; D.B. Downey (1995), When bigger is not better: Family size, parental resources, and children’s educational performance.

46 L. Gurgand and others (2023), The influence of sibship composition on language development at 2 years of age in the ELFE birth cohort study; H. Peyre and others (2016), Differential effects of factors influencing cognitive development at the age of 5-to-6 years.

47 N. Havron and others (2022), The effect of older sibling, postnatal maternal stress, and household factors on language development in two-to four-year-old children.

48 C.J. Diaz and J.E. Fiel (2021), When size matters: IV estimates of sibship size on educational attainment in the US.

49 M. Reader and others (2025), Does cutting child benefits reduce fertility in larger families? Evidence from the UK’s two-child limit.

50 M. Brewer and others (2012), Does welfare reform affect fertility? Evidence from the UK.

51 M. Reader and others (2025), Does cutting child benefits reduce fertility in larger families? Evidence from the UK’s two-child limit; R. Patrick and K. Andersen (2022), The two-child limit and fertility decision making: When policy narratives and lived experiences collide.

52 M. Reader and others (2025), Does cutting child benefits reduce fertility in larger families? Evidence from the UK’s two-child limit.

53 K. Andersen and others (2025), ‘It’s the kids that suffer’: Exploring how the UK’s benefit cap and two-child limit harm children; S. Motiani and others (2024), Lost opportunities: Parents’ perspectives on how the two-child limit policy is affecting their children’s early learning and development.

Family structure

Family structure has changed significantly over recent decades, with important implications for our understanding of inequalities in children’s educational outcomes. First, in the UK and other high-income countries people are generally becoming parents at increasingly older ages. The average maternal age in England and Wales rose from 26.1 years in 1996 to 28.8 years in 2016. This trend was particularly marked for university-educated women, with the average maternal age increasing from 30.0 to 32.9 in the same period. Studies also show that between 2010 and 2020, women who attained higher education levels than their parents (i.e. those who were upwardly socially mobile in educational terms) were more likely to delay parenthood than were women who were considered downwardly socially mobile. By the end of the decade, the timing of childbearing among upwardly mobile women closely resembled that of women who were already highly educated and whose parents were also more highly educated. Overall, delaying parenthood can provide parents with more time to gain higher qualifications, progress in the labour market and achieve greater economic security, all of which can support better outcomes for their children.

In addition, patterns of partnership have shifted. Modern couples are more likely to cohabit before or instead of marrying, and the proportion of children born to cohabiting parents, relative to those born to parents who are married or in a civil partnership, has increased substantially. Meanwhile, the share of children born to parents who are not living together has remained relatively stable over time. There is some evidence of an educational gradient to these trends. Research shows that marriage remains the predominant context for childbirth among first-time mothers with degrees but cohabitation is the most common arrangement among those without a degree. There is also some geographical variation, with martial births more prevalent in London and the South East of England than in other regions of the UK.

Large-scale longitudinal evidence shows that children born to cohabiting couples have slightly lower cognitive and socioemotional outcomes than those born to married couples, with the largest differences seen in attainment in later secondary school. Children in single-parent households have also been found to have worse developmental outcomes than those in two-parent families, although more recent analyses do not replicate this finding when other indicators of children’s developmental contexts, such as children’s health, the home learning environment and ethnicity, are also considered.

Overall, studies suggest that differences in educational outcomes are not causally driven by family structures but rather reflect underlying differences in the characteristics of individuals in different types of relationships. For example, married couples are more likely to be older and university educated and to own their own home, compared with cohabiting and single parents. These characteristics are shown to be related to children’s educational outcomes. For example, a 2020 study looked at the link between lone parenthood and children’s outcomes at age 11 in three birth cohorts (1958, 1970 and 2000). It found significant differences in outcomes between children with single parents versus those with couple parents, and these differences were remarkably stable over time. Some, though not all, of the gap was explained by differences in parent characteristics at the time of the child’s birth, with the remaining differences entirely mediated by contemporaneous characteristics, particularly family income. In other words, children living in single-parent families have worse outcomes, on average, than their peers living in dual-parent families because lone parenthood is associated with other characteristics that are negatively related to children’s outcomes, such as lower income and poorer mental health, rather than because there was any direct effect of lone parenthood per se on children’s outcomes.

Studies have also examined how family transitions, such as parental separation, re-partnering and blended families, can affect children’s outcomes. For example, recent analyses of the MCS and other international datasets show that children who have experienced family transition have lower cognitive and socioemotional outcomes in primary and secondary school, compared with their peers in more stable family environments. However, family transitions per se are not the key mechanisms underpinning the association with children’s outcomes. Instead, the relationship between family transitions and children’s outcomes is largely driven by the instability hypothesis. This theory posits that difficulties such as reduced family resources, increased stressors, the need for adjustments and parental depression contribute to adverse educational outcomes. A systematic review of 39 studies, mostly in the US, found that a range of protective factors, including strong parent–child relationships, good social support networks, stable routines and financial security, helped to mitigate some of these difficulties.

54 S. Cattan and others (2025), What is the effect of the two-child limit on children’s school readiness?

55 ONS (2018), Mean age of mother at birth of first child, by highest achieved educational qualification, 1996 to 2016, England and Wales.

56 J. Ermisch (2024), The recent decline in period fertility in England and Wales: Differences associated with family background and intergenerational educational mobility.

57 K. Kiernan and others (2024), Families and inequalities; see Section 3.1.

58 A. Pelikh and others (2022), Make up or break up? Partnership transitions among young adults in England and Wales.

59 K. Kiernan and others (2024), Families and inequalities.

60 N. Peri-Rotem and J. Scott (2017), Differences in partnership and marital status at first birth by women’s and their partners’ education: Evidence from Britain 1991–2012.

61 K. Kiernan and others (2024), Families and inequalities.

62 C. Crawford and others (2013), Cohabitation, marriage, relationship stability and child outcomes.

63 S. Cattan and others (2024), Early childhood inequalities; S. Harkness and others (2020), The rise in single-mother families and children’s cognitive development: Evidence from three British birth cohorts.

64 L.A. Outhwaite (2025), Understanding early inequalities: Multiple dimensions of children’s developmental contexts predict age 3 outcomes.

65 C. Crawford and others (2013), Cohabitation, marriage, relationship stability and child outcomes; S. Harkness and others (2020), The rise in single-mother families and children’s cognitive development: Evidence from three British birth cohorts.

66 K. Kiernan and others (2024), Families and inequalities.

67 S. Harkness and others (2020), The rise in single-mother families and children’s cognitive development: Evidence from three British birth cohorts.

68 J.A. Jarvis and others (2023), Family structure and child behavior in the United Kingdom; A. Solaz and others (2024), Does family structure account for child achievement gaps by parental education? Findings for England, France, Germany and the United States.

69 K. Hadfield and others (2018), Do changes to family structure affect child and family outcomes? A systematic review of the instability hypothesis.

Inequalities within families

Birth order

There is mixed evidence on the relationship between birth order and educational outcomes. Large-scale studies in the UK and the US show that relative to their older siblings, later-born children have lower cognitive outcomes from infancy through to adolescence, including when controlling for socio-economic circumstances. In line with resource dilution theory, these differences are associated with fewer family investments per child, such as reduced cognitive stimulation and fewer prenatal health investments, for later-born children.

However, these effects of birth order are not always consistent. A study using large-scale longitudinal data from the US showed that in early childhood, the first-born advantage is observed only in English-speaking families who were not experiencing poverty, poor mental health or single parenthood. Within these families, first-born children scored between 0.07–0.13 standard deviations higher than their later-born siblings in early cognitive, literacy and maths skills. In contrast, later-born children from more disadvantaged backgrounds have early educational outcomes equivalent to or even slightly better than those of their first-born siblings. Similarly, analyses across four British cohorts born between 1946 and 2000 to 2002 found that 10- and 11-year-olds without siblings perform as well as children from two-child families and better than those with two or more siblings. However, these differences substantially reduced when family circumstances, including socio-economic background and parental separation, were considered. Specifically, children from lower socio-economic backgrounds or who had experienced parental separation had lower cognitive outcomes than their peers, regardless of whether they were only children or had siblings. This suggests that family circumstances, rather than birth order or family size, are the primary factors influencing educational success.

Similarly, studies show that birth order and number of siblings are not significantly related to socioemotional outcomes. The quality of the sibling relationship (e.g., infrequent bullying behaviours between siblings) is a more important predictor, and many parents reported similar levels of emotional support for their children, regardless of birth order.

While many of these studies adjust for parent age and family size when estimating birth order effects, recent work by Clark and Cummins (2024) adopted a different approach by estimating the net effect of birth order. The net effect aims to capture the overall influence of birth order, separate from overlapping factors. For example, birth order is highly correlated with parent age, which, when considered together, can cancel each other out. Using this approach with genealogical data from 1680 to 2024 in England, the authors found no consistent birth order effects for the general population. Advantages for first-born children were limited to those born before 1929 to elite families. However, they were focused primarily on outcomes relating to broad social status indicators, such as elite university attendance and elite occupations. They were not able to capture cognitive or socioemotional outcomes typically used to assess children’s educational development. Also, their sample consisted of rare-surname families, which may limit generalisability.

Together, this evidence suggests that while birth order may influence educational outcomes throughout childhood, the direction and strength of these effects depend on family context, socio-economic factors and methodological choices in analysis. Instead, the underlying mechanisms, such as different family circumstances, parental investments, resource availability and structural disadvantage, are more likely to be factors directly influencing children’s educational outcomes.

70 J.Y.K. Lehmann and others (2018), The early origins of birth order differences in children’s outcomes and parental behavior; D. Muslimova and others (2024), Gene–environment complementarity in educational attainment.

71 J. Blake (1981), Family size and the quality of children; D.B. Downey (1995), When bigger is not better: Family size, parental resources, and children’s educational performance.

72 J.Y.K. Lehmann and others (2018), The early origins of birth order differences in children’s outcomes and parental behavior.

73 Luo and others (2022), A closer look at the birth order effect on early cognitive and school readiness development in diverse contexts.

74 A. Goisis and others (2023), Only children and cognitive ability in childhood: A cross-cohort analysis over 50 years in the United Kingdom.

75 J.Y.K. Lehmann and others (2018), The early origins of birth order differences in children’s outcomes and parental behavior; D. Yucel and A.V. Yuan (2015), Do siblings matter? The effect of siblings on socio-emotional development and educational aspirations among early adolescents.

76 D. Yucel and A.V. Yuan (2015), Do siblings matter? The effect of siblings on socio-emotional development and educational aspirations among early adolescents .

77 J.Y.K. Lehmann and others (2018), The early origins of birth order differences in children’s outcomes and parental behavior.

78 G. Clark and N. Cummins (2024), Birth order and social outcomes, England, 1680–2024.

Month of birth

In England, where the academic year runs from September to August, children born in the summer months, particularly in July and August (at the end of the academic year), tend to experience poorer educational outcomes than their autumn-born peers (who are born early in the academic year). For example, studies show that children born in August have lower scores on national attainment tests, other measures of cognitive and socioemotional skills, and parent and teacher perceptions of academic performance, relative to those born in September. Also, teachers are more likely to report that these younger children exhibit language delays, behavioural challenges and lower school readiness at the end of the Reception year. In some cases, these perceptions translate into higher rates of teacher-identified special educational needs (SEN), particularly for children with moderate learning difficulties.

Although evidence suggests that these effects tend to diminish over time, they are not limited to the early years. Large-scale longitudinal data from the UK shows that, while they catch up to some extent with their autumn-born counterparts, summer-born children continue to show lower average attainment in secondary school and report poorer socioemotional outcomes during adolescence compared with their older peers. Notably, these month-of-birth differences were observed to be of very similar magnitude across all socio-economic groups, despite some evidence that parents of August-born children are, on average, more likely to provide a richer home learning environment at age five than parents of September-born children. Given that the home learning environment plays an essential role in children’s early learning and development, this suggests that there may be other mechanisms underpinning the association between month of birth and educational outcomes. These include age of starting school, number of months of schooling before assessment, relative age in the classroom and age at test.

A comprehensive 2014 analysis concluded that age at test is the dominant driver of these effects: younger children perform less well because they are developmentally less mature at the time of assessment than their older peers. The combined contribution of the other three mechanisms was found to be minimal. In addition to developmental factors, adult perceptions may also play a role. For example, while summer-born children are more likely to be identified by teachers as having learning difficulties, these differences disappear when more rigorous diagnostic assessments are applied. Overall, international evidence shows that the effect of birth month per se can be ruled out, because similar disparities are observed between children born at the start and end of the academic year in countries with different academic year dates.

Furthermore, despite the early disadvantages associated with being summer-born, and the significant differences found in GCSE attainment, evidence of long-term impacts into adulthood is limited. A postgraduate study found no significant relationship between term of birth and university attendance overall, although other research has reported that summer-born students are marginally less likely to attend a Russell Group institution. The impact of month of birth on degree choice and outcomes was not evaluated. In the longer term, factors such as family instability and socio-economic background appear to have a far greater influence on educational and life outcomes than month of birth. Moreover, month of birth is unlikely to play a key role in explaining differences in outcomes between children growing up in different economic circumstances, because it does not vary systematically by socio-economic status.

79 Crawford and others (2011), Does when you are born matter? The impact of month of birth on children’s cognitive and non-cognitive skills in England; S.E. Rose and C.M. Barlow (2024), The impact of relative age effects on psychosocial development: A systematic review.

80 C.F. Norbury and others (2016), Younger children experience lower levels of language competence and academic progress in the first year of school: Evidence from a population study.

81 G. Squires and others (2012), The identification of special educational needs and the month of birth: Differential effects of category of need and level of assessment.

82 C. Crawford and others (2011), Does when you are born matter? The impact of month of birth on children’s cognitive and non-cognitive skills in England.

83 P. Patalay (2015), The extent and specificity of relative age effects on mental health and functioning in early adolescence.

84 C. Crawford and others (2011), Does when you are born matter? The impact of month of birth on children’s cognitive and non-cognitive skills in England.

85 L.A. Outhwaite (2025), Understanding early inequalities: Multiple dimensions of children’s developmental contexts predict age 3 outcomes; K. Sylva and others (2013), The effects of early experiences at home and pre-school on gains in English and mathematics in primary school: A multilevel study in England; K. Toth and others (2020), Home learning environment across time: The role of early years HLE and background in predicting HLE at later ages.

86 C.Crawford and others (2014), The drivers of month-of-birth differences in children’s cognitive and non-cognitive skills.

87 G.Squires and others (2012), The identification of special educational needs and the month of birth: Differential effects of category of need and level of assessment.

88 P. Givord (2020), How a student’s month of birth is linked to performance at school: New evidence from Pisa.

89 C. Wright (2015), A multilevel analysis of post-16 and higher education in England, PhD thesis, University of Bristol.

90 C. Crawford and others (2011), Does when you are born matter? The impact of month of birth on children’s cognitive and non-cognitive skills in England.

91 C. Wright (2015), A multilevel analysis of post-16 and higher education in England, PhD thesis, University of Bristol.

Positive intervention by parents

Parent behaviours

It is well established that the quality of the home learning environment and positive parental behaviours have a lasting impact on children’s cognitive and socioemotional outcomes. In the early years, high-quality home learning environments are often characterised by stimulating and responsive social interactions and cognitive stimulation. Many large-scale longitudinal studies, including the MCS and the Effective Pre-school and Primary Education (EPPE), measure the early home learning environment through the extent to which parents engage in activities such as reading to children, painting and drawing, visiting the library, playing with letters or numbers, and engaging in songs, poems or nursery rhymes.

Data from the MCS and the EPPE show that richer home learning environments, particularly in the early years, are strongly associated with improved cognitive and socioemotional outcomes throughout primary education. Other longitudinal studies, in the UK and the US, have also demonstrated that the home learning environment in early childhood is associated with educational outcomes throughout adolescence into adulthood, with cumulative benefits when both home and formal childcare environments are of high quality. Notably, the association between a high-quality home learning environment on children’s educational outcomes was similar in size to that observed for having a university-educated mother. In both cases, the relationship was modest, with correlations of around 0.33.

Some UK-based evidence indicates that university-educated mothers living in higher-income households and with good mental health typically spend more time reading to and engaging in early learning activities with their children than mothers without these characteristics when the child is aged three years old. Similarly, differences in the home learning environment early in a child’s life, captured at ages three, five and seven, have been shown to explain around 10% of the difference in educational outcomes at ages 7, 11 and 16 between children from high- and low-income households, highlighting the potential long-term benefits of these early investments in children’s learning environments.

However, other studies find only a modest correlation between parents’ qualifications and home learning environment quality in the early years, and some report limited differences in parental support for maths and writing activities across socio-economic groups at ages 7 to 11. This suggests that while the home learning environment is an important mechanism in explaining the causal relationship between parents’ education and children’s outcomes, parents with lower formal qualifications can and do provide highly enriching home learning environments that benefit their children. This highlights the importance of supporting all families to engage in activities that promote children’s learning and development from a young age.

Encouragingly, there is also evidence that the home learning environment can be positively shaped by policy and practice interventions. For example, early evidence on the impact of access to Sure Start centres highlighted improvements in parenting behaviours and the home learning environment, which more recent evidence on the longer-term benefits of access to Sure Start centres hypothesised to be one of the key pathways through which the programme has improved subsequent educational attainment. This chimes with research on highly intensive, highly targeted programmes in the US, such as the Perry Preschool and Abecedarian projects; these have also been shown to improve parenting behaviours and the home learning environment, and wider evidence suggests that this might have been a key component of their success in generating long-lasting, even intergenerational, effects. More broadly, meta-analyses of randomised controlled trials have demonstrated that targeted interventions to improve the quality of the home learning environment lead to significantly higher maths and literacy outcomes at ages three to nine.

The characteristics of a high-quality home learning environment and the form of parental involvement tend to evolve as children progress through their education. For older children and young people, important aspects include parents’ participation in school activities, helping with homework, regular parent–child discussions about school, and the availability of material resources such as access to a computer and the internet, a quiet study space and private tuition. Evidence from Next Steps (formerly known as the Longitudinal Study of Young People in England) shows that among these factors, material resources are most strongly associated with higher GCSE outcomes. Specifically, young people with access to both a computer and the internet at home score, on average, 5.5 percentiles higher than those without. Only parental aspirations were found to have a stronger association with attainment (see next section). These inequalities were particularly pertinent during the COVID-19 pandemic, when young people from disadvantaged families reported studying for fewer days per week and hours per day, received less support at home and through private tuition, and felt more held back by school closures as a result.

Other studies using the Next Steps dataset also show that parental involvement behaviours are not static. A 2010 study found that around half of parents of Year 9 pupils (age 13–14 years) reported fluctuations in their involvement in their child’s school life over a two-year period. These changes were strongly linked to factors such as the parents’ relationships with the school, the child’s attitude to learning, attendance at parents’ evenings, and wider family circumstances, including ethnicity, parent qualifications and changes in employment.

Positive parental involvement often continues into university, particularly through financial support. Data from the Student Income and Expenditure Survey shows that students from higher-income families typically receive more help from their families than their lower-income peers. While means-tested financial support for disadvantaged young people can significantly benefit their educational outcomes, evidence shows that low-income university students are disproportionately affected by the current declining real value of student loans and rising accommodation and maintenance costs.

However, while parental behaviours and resources matter in secondary school and higher education, prior attainment at age 11 has been found to explain nearly four times as much of the variation in educational outcomes at age 16. Similarly, studies show that inequalities in access to higher education largely reflect existing attainment gaps across socio-economic status, gender and ethnicity. Overall, this emphasises the importance of positive family behaviours and investments starting from an early age.

92 L. Boonk and others (2018), A review of the relationship between parental involvement indicators and academic achievement; Daucort and others (2021), The home math environment and math achievement: A meta-analysis; OECD (2020), Early learning and child well-being: A study of five-year olds in England, Estonia, and the United States.

93 K. Sylva and others (2013), The effects of early experiences at home and pre-school on gains in English and mathematics in primary school: A multilevel study in England; A. Sullivan and others (2013), Social class and inequalities in early cognitive scores; K. Toth and others (2020), Home learning environment across time: The role of early years HLE and background in predicting HLE at later ages.

94 E. Del Bono and others (2016), Early maternal time investment and early child outcomes; K. Sylva and others (2013), The effects of early experiences at home and pre-school on gains in English and mathematics in primary school: A multilevel study in England; A. Sullivan and others (2013), Social class and inequalities in early cognitive scores.

95 H. Beshir and others (forthcoming), Explaining human capital inequalities: The role of parents, schools and pupils, CLS Working Paper, London: CLS, UCL; A. A. Whitaker and others (2023), Predicting adolescent and young adult outcomes from emotional support and cognitive stimulation offered by preschool-age home and early care and education settings.

96 R.J. Duncan and others (2019), Additive and synergistic relations of early mother–child and caregiver–child interactions for predicting later achievement; K. Sylva and others (2013), The effects of early experiences at home and pre-school on gains in English and mathematics in primary school: A multilevel study in England.

97 K. Sylva and others (2013), The effects of early experiences at home and pre-school on gains in English and mathematics in primary school: A multilevel study in England.

98 S. Cattan and others (2024), Early childhood inequalities.

99 H. Beshir and others (forthcoming), Explaining human capital inequalities: The role of parents, schools and pupils, CLS Working Paper, London: CLS, UCL.

100 K. Sylva and others (2013), The effects of early experiences at home and pre-school on gains in English and mathematics in primary school: A multilevel study in England.

101 SMC (2024), Childhood origins of social mobility: Revised and updated.

102 L. Macmillan and E. Tominey (2023), Parental inputs and socio-economic gaps in early child development.

103 N. Archer and C. Oppenheim (2021), The role of early childhood education and care in shaping life chances The changing face of early childhood in the UK

104 DfE (2018), Improving the home learning environment: A behaviour change approach.

105 E. Melhuish and others (2008), Effects of fully-established Sure Start Local Programmes on 3-year-old children and their families living in England: A quasi-experimental observational study.

106 P. Carneiro and others (2025), The short- and medium-term effects of Sure Start on children’s outcomes.

107 Baulos and others (2024), Perry Preschool at 50: What lessons should be drawn and which criticisms ignored?

108 J. Garcia and J. Heckman (2023), Parenting promotes social mobility within and across generations.

109 A. Cahoon and others (2024), Meta-analyses and narrative review of home-based interventions to improve literacy and mathematics outcomes for children between the ages of 3 and 5 years old, Review of Educational Research, 94(6), 803–842; G. Nelson and others (2024), A meta-analysis and quality review of mathematics interventions conducted in informal learning environments with caregivers and children.

 110 H. Chowdry and others (2011), The role of attitudes and behaviours in explaining socio-economic differences in attainment at age 16; C. Farquharson and others (2024), Education inequalities.

111 H. Chowdry and others (2011), The role of attitudes and behaviours in explaining socio-economic differences in attainment at age 16.

112 J. Anders and others (2023), Inequalities in late adolescents’ educational experiences and wellbeing during the Covid-19 pandemic; A. Andrew and others (2020), Family time use and home learning during the COVID-19 lockdown.

113 E. Skaliotis (2010), Changes in parental involvement in secondary education: An exploration study using the longitudinal study of young people in England.

114 DfE (2023), Student income and expenditure survey 2021 to 2022.

115 R. Murphy and G. Wyness (2023), Testing means-tested aid; G. Wyness and R. Murphy (2024), Higher education funding: What’s the problem and what are the potential solutions?

116 H. Chowdry and others (2011), The role of attitudes and behaviours in explaining socio-economic differences in attainment at age 16.

117 C. Crawford and E. Greaves (2015), Socio-economic, ethnic and gender differences in HE participation.

Parent beliefs and aspirations

Positive parent behaviours are underpinned by parents’ beliefs about their perceived role in their child’s education and their confidence in their ability to help. The Revised Parental Involvement Process (R-PIP) model highlights three core psychological factors shaping parents’ involvement: their role construction (i.e. beliefs about whether they should be involved in their child’s learning), their self-efficacy (i.e. confidence in their ability to make a difference), and the degree to which they receive invitations and opportunities to participate from schools and children themselves.

A large body of international evidence shows that when parents had a strong sense of role and efficacy, they were more likely to engage in their child’s learning and development. However, parents’ anxieties about their abilities can also undermine their confidence to support children’s learning. For example, around 20% of UK adults report feeling anxious about mathematics, which is distinct from general anxiety. Evidence from the US shows that when maths-anxious parents provide frequent, well-meaning homework help, their children often develop higher maths anxiety themselves and make less progress in the subject. Invitations and opportunities from school are particularly influential for parents who may otherwise feel unsure about how and why to engage with their child’s education.

In addition to beliefs about themselves, parents’ aspirations and expectations, such as whether they believe their child will stay in education or progress to university, have consistently been shown to be among the strongest predictors of children’s academic success. A meta-analysis of 98 studies published between 2000 and 2017 found that parental academic expectations have the largest effect on children’s educational outcomes compared with other forms of parental involvement, such as participation in school activities, supporting learning at home and parent–child academic discussions. Parental aspirations and expectations also play a key role in explaining differences in educational attainment between groups. Evidence from Next Steps shows that after prior attainment, parental aspirations and expectations are the single strongest factor mediating differences in GCSE results at age 16 between young people from different socio-economic backgrounds, compared with positive family–child interactions and material resources including private tuition and computer and internet access. Analysis of the same data also found that these measures are among the key reasons that young people from ethnic minority backgrounds have higher educational attainment than young people of White British ethnicity, particularly among those from low socio-economic backgrounds.

Further studies in the US show that these beliefs can be shaped by the wider family network. Research using data from the Youth Development Study found that grandparents’ educational expectations for their children during adolescence predicted those children’s parental expectations for own their children’s educational outcomes and university attendance decades later. However, studies also suggest that a mismatch between parents’ and children’s aspirations can hinder young people’s educational progress. For example, analyses of the Next Steps study and the British Birth Cohort Study show that incongruent educational aspirations are associated with a lower likelihood of participating in and completing post-16 and higher education, compared to families in which both parents and children hold high aspirations, even after controlling for prior attainment and school motivation.

Among the incongruent families, young people with high aspirations but low-aspiration parents are less likely to remain in education after age 16 than their peers with low aspirations but high-aspiration parents. However, by age 26 this pattern is reversed: young people with high aspirations but low-aspiration parents are more likely to achieve a university degree than their low-aspiration peers with high-aspiration parents. Such mismatches are more common in families with lower parental education and among young people with lower school motivation. 

Ethnic minority students are more likely than their White British peers to share high aspirations with their parents. These high aspirations among ethnic minority families may, in part, explain why more ethnic minority students progress to university, regardless of their prior attainment. However, despite these high educational aspirations, other studies show that significant inequalities based on ethnicity remain for education and employment outcomes. For example, graduates from Asian, Black and Mixed ethnic minority backgrounds are less likely than White British graduates to be awarded a first or upper second-class degree. Ethnic minority students are also less likely to be offered a graduate role in professional occupations, despite being overrepresented in the applicant pool. These inequalities persist even after controlling for prior educational attainment. This evidence suggests that while family aspirations play an important role in supporting young people, there are also issues within education and employment systems and wider society that must be addressed to ensure equality of opportunity.

118 K.V. Hoover-Dempsey and H.M. Sandler (1995), Parental involvement in children’s education: Why does it make a difference?; K.V. Hoover-Dempsey and H.M. Sandler (1997), Why do parents become involved in their children’s education?

119 A.M. Albanese and others (2019), The role of parental self-efficacy in parent and child well-being: A systematic review of associated outcomes; K.V. Hoover-Dempsey and others (2005), Why do parents become involved? Research findings and implications; J. Goodall (2013), Parental engagement to support children’s learning: A six point model.

120 Maths Anxiety Trust (2018), Maths Anxiety Summit 2018.

121 M. Malanchini and others (2017), The genetic and environmental aetiology of spatial, mathematics and general anxiety.

122 E.A. Maloney and others (2015), Intergenerational effects of parents’ math anxiety on children’s math achievement and anxiety; C. Carkoglu and others (2023), Building the parent and child math anxiety network model from empirical evidence.

123 K.V. Hoover-Dempsey and H.M. Sandler (1997), Why do parents become involved in their children’s education?; E. Skaliotis (2010), Changes in parental involvement in secondary education: An exploration study using the longitudinal study of young people in England.

124 L. Boonk and others (2018), A review of the relationship between parental involvement indicators and academic achievement; A. Fiskerstrand (2022), Literature review: Parent involvement and mathematic outcome.

125 C.Y. Tan and others (2020), Academic benefits from parental involvement are stratified by parental socioeconomic status: A meta-analysis.

126 H. Chowdry and others (2011), The role of attitudes and behaviours in explaining socio-economic differences in attainment at age 16.

127 S. Strand (2014), Ethnicity, gender, social class and achievement gaps at age 16: Intersectionality and ‘getting it’ for the white working class.

128 J.T. Mortimer and M. Lee (2021), How do grandparents’ and parents’ educational attainments influence parents’ educational expectations for children?

129 I. Schoon and K. Burger (2021), Incongruence between parental and adolescent educational aspirations hinders academic attainment.

130 I. Schoon and K. Burger (2021), Incongruence between parental and adolescent educational aspirations hinders academic attainment.

131 P. Martin (2025), Examining the relationship between ethnicity, school attainment and higher education participation in England.

132 J.T. Richardson and others (2020), The role of gender, social class and ethnicity in participation and academic attainment in UK higher education: An update.

133  C. Dilnot and others (2025), Inequalities in access to professional occupations.

Intervention from wider support networks

Extended family

A growing body of evidence suggests that grandparents may play a significant, albeit complex, role in shaping their grandchildren’s educational outcomes from early childhood through to adolescence. This influence can occur through direct caregiving, financial contributions and emotional support, as well as through the transmission of socio-economic advantage. A systematic review of 206 studies showed wide variation in how grandparent involvement is conceptualised, ranging from co-residence and custodial care to informal caregiving roles and financial support, with mixed effects on children’s outcomes.

For example, with co-residence, evidence from the MCS shows that compared with children in two-parent households, those living in three-generational households have poorer cognitive and behavioural outcomes at ages three to five. This negative effect was attributed to a combination of factors, including less stimulating grandparenting activities, role ambiguity, and increased household conflict and stress, which outweigh the potential benefits. However, by middle childhood and early adolescence, the influence of three-generational living arrangements, relative to two-parent households, on educational outcomes had diminished.

In contrast, other studies examining informal grandparental care have reported more positive outcomes in early childhood. A 2018 study drawing on the MCS found that children receiving informal care from their grandparents have better language outcomes at 18 months than children who attended formal childcare settings. This is likely related to more stable relationships between children and grandparents and more opportunities for one-to-one interactions. However, these results were stronger for children from higher-income households. Meanwhile, children attending formal childcare settings have better school readiness, non-verbal reasoning and maths outcomes, and these results are stronger for children from lower-income backgrounds.

Longitudinal evidence from the US further highlights how grandparents’ socio-economic status can shape early academic skills. A 2011 study found that children with university-educated grandparents begin school with stronger literacy and numeracy skills than those with less-educated grandparents. However, these grandchildren are also more likely to live in two-parent, higher-income households and to have older mothers. This suggests that the observed benefits may reflect broader contextual advantages rather than a distinct ‘grandparent effect’.

The ‘grandparent effect’ describes the direct influence of grandparents’ socio-economic status on grandchildren’s educational outcomes, independent of parental factors, although there is debate about how large this effect may be. A systematic review of 69 analyses showed that 58% of studies reported significant ‘grandparent effects’, with grandparents’ influence averaging about one-quarter that of parents. For example, a 2013 study examined relative social mobility patterns across three generations in the UK between 1946 and 1970 and found that, controlling for parents’ social class, grandchildren whose grandparents were in the professional-managerial class are at least 2.5 times more likely to enter that class than those whose grandparents were unskilled manual workers. Notably, this effect persisted after adjusting for parents’ education, income and wealth. Overall, the ‘grandparent effect’ is stronger for maternal grandmothers, particularly when parents’ financial resources or education levels are low.

However, studies in other countries have not always replicated this ‘grandparent effect’. For example, large-scale evidence from Sweden and the Netherlands shows that when detailed parental income and cognitive characteristics are accounted for, grandparental influence on grandchildren’s education and earnings largely disappears. Similarly, differences have been found in the magnitude of the ‘grandparent effect’ across European countries. While there were methodological differences across the studies (e.g. in how parent characteristics were measured), these contrasting findings may, in part, reflect important contextual differences across countries. Evidence shows that the relationship between parental education and children’s outcomes in the UK and USA is more than double that of countries such as Sweden and the Netherlands. Extended family resources, such as those provided by grandparents, may play a greater role in contexts where there are higher educational inequalities.

134 A.F. Sadruddin and others (2019), How do grandparents influence child health and development? A systematic review.

135 J. He and J. Wang (2021), When does it matter? The effect of three-generational household arrangement on children’s well-being across developmental stages.

136 D. Del Boca and others (2018), The role of grandparenting in early childcare and child outcomes.

137 J.L. Ferguson and D.D. Ready (2011), Expanding notions of social reproduction: Grandparents’ educational attainment and grandchildren’s cognitive skills.

138 M. Colagrossi and others (2020), Like (grand) parent, like child? Multigenerational mobility across the EU.

139 L.R. Anderson and others (2018), Grandparent effects on educational outcomes: A systematic review.

140 T.W. Chan and V. Boliver (2013), The grandparents effect in social mobility: Evidence from British birth cohort studies.

141 L.R. Anderson and others (2018), Grandparent effects on educational outcomes: A systematic review; H. Lehti and others (2019), Tying the extended family knot: Grandparents’ influence on educational achievement.

142 T. Bol and M. Kalmijn (2016), Grandparents’ resources and grandchildren’s schooling: Does grandparental involvement moderate the grandparent effect?; P. Engzell and others (2020), It’s all about the parents: Inequality transmission across three generations in Sweden.

143 M. Colagrossi and others (2020), Like (grand) parent, like child? Multigenerational mobility across the EU.

144 M. Corak (2013), Income inequality, equality of opportunity, and intergenerational mobility; J. Jerrim and L. Macmillan (2015), Income inequality, intergenerational mobility, and the Great Gatsby Curve: Is education the key?

Social networks and communities

‘Social capital’ refers to the strength of an individual’s social network and community. Research using large-scale social network data in the US has shown that children from low socio-economic families earn, on average, 20% more in adulthood if they grew up in communities with social networks similar in socio-economic status to those of their more affluent peers. Further research has found similar patterns in the UK, for example showing, using aggregated data on over 20 million Facebook users, that although people’s friendship networks are less divided by class than in the US, social capital remains unevenly distributed. Specifically, individuals from better-off backgrounds are more likely to be friends with others like themselves. Meanwhile, those from less well-off families tended to have fewer friendships with individuals from higher-income groups, meaning that they have access to fewer social connections that can support opportunity and mobility. These social connections are associated with later outcomes. Young people from low-income backgrounds who grew up in areas where friendships crossed economic lines go on to earn 38% more as adults than similar young people in areas where social networks were more economically segregated. Having more connections to higher-income groups is also linked to wider benefits, including higher reported levels of happiness and trust and lower feelings of loneliness and social isolation.

145 B. Sacerdote (2011) Peer Effects in Education: How Might They Work, How Big Are They and How Much Do We Know Thus Far? In ‘Handbook of the Economics of Education’

Extra-curricular activities

Participation in extra-curricular or ‘positive’ activities, including sport, cultural engagement, music, political participation, volunteering and youth clubs, is consistently associated with better education outcomes. Research shows that engagement in such activities at age 14 is linked to higher attainment at Key Stages 3 and 4 (ages 14 and 16), and a reduced likelihood of being NEET at age 17, even after controlling for earlier academic performance. Other studies also find longer-term benefits. Secondary school pupils who attend extra-curricular activities show greater odds of progressing to higher education and employment in early adulthood than those who do not attend.

Although the exact mechanisms driving these effects remain unclear, some commentators suggest that these extra-curricular opportunities allow children to engage with skilled non-parental adults who can fulfil a mentoring role. A meta-analysis of 70 studies shows that mentoring, particularly in adolescence, has a positive relationship with a range of educational, mental health and social outcomes. There is also some evidence that young people from lower socio-economic backgrounds benefit more from mentoring programmes than their more affluent peers. Although participation rates in positive activities are generally reported to be high, children with more highly educated mothers and those whose parents are more involved in school are more likely to engage in such activities. Meanwhile, children with SEN, those from lower-income households and those in care are less likely to participate. Qualitative evidence shows that many schools attempt to close the gap in access to extra-curricular activities through pupil premium funding. However, these approaches have not been formally evaluated.

148 H. Chowdry and others (2009), Drivers and barriers to educational success-evidence from the Longitudinal Study of Young People in England; A. Vignoles and E. Meschi (2010), The determinants of non-cognitive and cognitive schooling outcomes. Report to the Department of Children, Schools and Families.

149 D. Robinson (2024), Access to extra-curricular provision and the association with outcomes.

150 R. Gilligan (2007), Spare time activities for young people in care: What can they contribute to educational progress?

151 E.B. Raposa and others (2019), The effects of youth mentoring programs: A meta-analysis of outcome studies.

152 R.B. Thompson and others (2013), A longitudinal study of family socioeconomic status (SES) variables as predictors of socio-emotional resilience among mentored youth.

153 H. Chowdry and others (2009), Drivers and barriers to educational success-evidence from the Longitudinal Study of Young People in England; E. Holt-White and C. Cullinane (2023), Social mobility: The next generation; D. Robinson (2024), Access to extra-curricular provision and the association with outcomes; B. Taylor (2022), Pupil premium for looked after children: Its allocation, use and impact on educational outcomes for children aged 5–12.

154 S. Read and others (2020), Effective use of Pupil Premium Plus to improve educational outcomes for looked after children; B. Taylor (2022), Pupil premium for looked after children: Its allocation, use and impact on educational outcomes for children aged 5–12.

Children and young people with limited family support networks

For children and young people with limited family support networks, the wider community plays a highly significant role in supporting educational outcomes. For example, in England nearly half a million children are in contact with social care, including over 7,000 unaccompanied asylum-seeking children. Research shows that many of these children have persistently lower attainment than other children not in care.

A systematic review of seven studies in England shows that children in care have higher absence and exclusion rates, more frequent school moves, and reduced access to high-quality schooling, all of which are negatively associated with educational attainment. Similarly, recent evidence from London suggests that children in care and refugee and asylum-seeking children have a weaker sense of belonging in school relative to their peers. Fostering a sense of belonging and connectedness to school is important, as other studies have shown that there is a small but positive relationship between sense of belonging and better cognitive and socioemotional outcomes, particularly for children who may have been experiencing challenging social circumstances.

Several studies show that supportive school environments and relationships with trusted adults can mitigate some of the negative educational outcomes experienced by children in care and those affected by family difficulties. More broadly, a systematic review of 22 studies, primarily in the US, shows that interventions targeting students’ strengths and encouraging positive interactions among students and between school staff and students have an overall positive effect on young people’s sense of belonging in school, including for those from disadvantaged backgrounds.

155 DfE (2024), Children in need; DfE (2024), Children looked after in England including adoptions.

156 Berridge and others (2020), Children in need and children in care: Educational attainment and progress; A. O’Higgins (2019), Analysis of care and education pathways of refugee and asylum-seeking children in care in England: Implications for social work.

157 M.A. Jay and L. McGrath-Lone (2019), Educational outcomes of children in contact with social care in England: A systematic review.

158 Brown and others (2024), Inclusion, belonging, and safety in London schools.

159 H. Korpershoek and others (2020), The relationships between school belonging and students’ motivational, social-emotional, behavioural, and academic outcomes in secondary education: A meta-analytic review; I.D. Rose and others (2024), The relationship of school connectedness to adolescents’ engagement in co-occurring health risks: A meta-analytic review.

160 D. Berridge and others (2020), Children in need and children in care: Educational attainment and progress; L. Larsen and others (2025), Children’s school reluctance and satisfaction: The combined role of teacher support and parent–child relation.

161 K.-A. Allen and others (2022), Impact of school-based interventions for building school belonging in adolescence: A systematic review.

Conclusion

This report highlights the central role of families and the wider social networks surrounding them in shaping children’s educational outcomes across the life course. In particular, parental education, particularly mothers, is causally related to children’s educational outcomes: children whose parents have at least GCSE-level (or equivalent) qualifications have higher attainment than children whose parents do not have any qualifications. Targeted interventions that improve the quality of the home learning environment have also been shown to lead to significantly higher maths and literacy skills.

Parental aspirations and expectations are also strongly associated with children’s long-term academic success and socioemotional development, over and above family socio-economic circumstances. Beyond the immediate family, broader relationships, including those with grandparents, trusted non-parental adults, diverse social networks and supportive school environments, also provide valuable support, guidance and opportunities that are associated with better educational outcomes.

While family structure, size and birth order may have some relationship with educational outcomes, evidence suggests that their apparent effects are often driven by other underlying mechanisms, such as financial strain, time and resource constraints, and parental stress and mental health difficulties. The role of children’s month of birth in explaining differences in outcomes across families is limited, as month of birth does not vary systematically by socio-economic circumstances. Overall, this evidence highlights the value of recognising how families and wider social support networks operate within multiple, interacting layers of influence on children and young people’s educational outcomes. Although the relative importance of family-related factors may shift as young children progress into adolescence and adulthood, they remain a key source of support for children’s long-term success and social mobility.