Disadvantage

Socio-economic disadvantage continues to shape children’s educational outcomes from the earliest years through to 16-19 education and beyond. EPI’s research has consistently shown that children from disadvantaged families face cumulative barriers to learning that can leave them increasingly behind their peers as they progress through education. The government’s schools white paper recognises the scale of this challenge and has committed to halving the disadvantage gap.

Tracking progress against that ambition requires a clear and consistent measure of the gap. In this section we report the disadvantage gap at each key stage of education from reception to 16-19 education, and consider how this has changed over time. We define the gap as the difference in attainment between disadvantaged learners and their peers, using eligibility for free school meals (FSM) as a proxy for disadvantage. This year we have developed a new ‘EverFSM’ measure to have a more consistent measure of the gap over time. Our previous Ever6 measure was increasingly distorted due to the roll-out of Universal Credit. See methodology page for details.

Early years foundation stage

The attainment gap between disadvantaged pupils and their peers is already significant by age five. In 2025, 17.4 per cent of reception-aged pupils were eligible for free school meals, down from 17.7 per cent in 2024, but still a higher proportion than before the pandemic. Despite this reduction in eligibility, the attainment gap in 2025 was the widest recorded for over a decade, returning to the level last seen in 2014. Only the first year in our series, 2013, recorded a higher rate.

Figure D.1 shows the early years disadvantage gap grew from 4.7 months in 2024 to 4.9 months in 2025. This is an acceleration of the upwards trend since 2023 (when the gap was 4.6 months), leaving disadvantaged children significantly further behind than they were in the years before the pandemic.

Figure D1: The disadvantage gap for pupils in reception year has grown further in 2025 and is now the widest since 2014

Primary school 

By the end of primary school, 31.4 per cent of pupils were recorded as disadvantaged in 2025, based on having ever been eligible for free school meals during primary school. This was higher than in 2024 (30.4 per cent) and above its post-pandemic low (of 29.7 per cent) in 2022.  

On this measure, the disadvantage gap in year 6 remained largely unchanged from 2024 at 10.0 months. This was below its post-pandemic peak of 10.2 months in 2022 and 2023 but still substantially above pre-pandemic levels, when the gap reached a low of 9.2 months in 2018.  

Notably, the gap had already begun to widen before the pandemic, following a marked long-term narrowing between 2011 and 2018. 

Figure D2: The disadvantage gap at key stage 2 was unchanged in 2025 but still substantially wider than before the pandemic 

 

Secondary school 

By the end of secondary school, one-third (33.4 per cent) of pupils were recorded as disadvantaged in 2025, based on having ever been eligible for free school meals during primary or secondary school. This is higher than the proportion at the end of primary school (31.4 per cent), as pupils have had a longer time to be recorded as eligible by year 11. This proportion has fallen each year since its peak (of 34.2 per cent) in 2022.  

On this measure, the disadvantage gap widened to 19.0 months in 2025, up from 18.8 months in 2024 and returning to its 2023 level. This is its joint-highest level since the start of the series in 2013, and a full month wider than before the pandemic. EPI’s recent research on the drivers of the gap shows that the post-pandemic increase in the gap over the period 2019 to 2023 can be entirely explained by the higher levels of absence among disadvantaged pupils. Before the pandemic, the gap had been narrowing between 2013 and 2017, after which progress stalled.  

Figure D3: In 2025, the disadvantage gap in GCSE English and maths increased to 19.0 months – its joint-highest level since the start of the series in 2013 

16-19 education 

In the 16-19 phase, 31 per cent of students were defined as disadvantaged in 2025, based on their free school meal eligibility between reception and year 11. This is slightly lower than in earlier phases, partly because not all young people continue into 16-19 education after secondary school. 

The measured 16-19 disadvantage gap has been broadly stable at around 3.3 to 3.4 grades across students’ best three qualifications, except in 2022 when grading arrangements had not fully returned to pre-pandemic norms. However, this measure only includes students who enter 16-19 education, so it is likely to understate the overall scale of educational inequality at this stage. 

Figure D4: The 16-19 disadvantage gap across students’ best three qualifications remained at pre-pandemic levels in 2025 

 

Due to a change in definitions by the Department for Education, the cohorts of students finishing 16-19 study in 2020 and 2021 were not directly comparable, so we have not shown the disadvantage gap for these years on our chart.

What is happening to the post-16 participation gap? 

Participation here means studying towards a substantial qualification or apprenticeship at the beginning of year 12. Disadvantaged young people (based on free school meal eligibility in the previous 6 years) are less likely than their non-disadvantaged peers to be participating in this way. 

This participation gap between disadvantaged and non-disadvantaged young people was stable at around 10 percentage points between 2015 and 2020, but has since widened. 

Since 2020, non-participation has increased overall, but the increase has been much sharper among disadvantaged students. Among young people who finished their GCSEs in 2023, 23.0 per cent of disadvantaged students were not participating in post-16 education or training at the beginning of year 12, compared with 9.7 per cent of non-disadvantaged students. This represents a gap of 13.3 percentage points, compared with a gap of 10.5 percentage points in 2020. 

This affects interpretation of the attainment gap. As disadvantaged students are less likely to participate, they are also less likely to be included in attainment data. These trends sit within a wider policy focus on post-16 participation. In its 2025 skills white paper, the government confirmed plans to introduce automatic enrolment into college for young people who have not secured a place elsewhere. The interim report from the Milburn review also focuses on young people’s transitions between education, employment and training, including the role of cumulative disadvantage and weak transitions between services. The final report is expected in late 2026. 

Figure D5: The year 12 participation gap between disadvantaged and non-disadvantaged students has widened in recent years

 

Disadvantage gaps have narrowed slightly across all level 3 qualification types since 2022. However, gaps in A levels and applied general qualifications remain wider than before the pandemic. Since 2023, the largest narrowing has been in other level 3 qualifications, although these trends are likely to be affected by changes in which qualifications are included, the gradual defunding of some qualifications, and falling participation among disadvantaged 16-19 year olds. 

Despite this narrowing within qualification types, the overall 16-19 disadvantage gap has remained broadly unchanged. This reflects changes in who enters different qualifications, and who participates in post-16 education or training, rather than clear overall improvements in outcomes. Disadvantaged students continue to be under-represented in A levels and over-represented in level 2 or lower study post-16. As a result, gaps within qualification types show differences between students taking the same qualifications, but do not capture the full picture of disadvantage across all young people. 

Figure D6: Disadvantage gaps by 16-19 qualification have narrowed slightly in recent years, but for A levels and applied general remain wider than they were prior to the pandemic 

 

Disadvantage gaps by qualification type are based on average points per qualification rather than total points over students’ best three qualifications. 

Persistent disadvantage 

In our previous reports we have shown not only that disadvantaged pupils tend to do worse than their peers but that those who are disadvantaged for the longest duration do worst of all.  

Persistently disadvantaged pupils consistently demonstrate some of the lowest educational outcomes in the system, falling, on average, nearly two years behind their peers by the time they sit their GCSEs. 

Our research has also shown that pupils have been increasingly experiencing long-term (or persistent) disadvantage and this trend emerged even prior to the pandemic. This is concerning and consistent with wider evidence of rising child poverty in recent years.  

And while there was progress in gap-narrowing for the headline gap leading up to the pandemic, there was no long-term progress in narrowing the persistent gap, which has remained stubbornly high over successive years. 

For these reasons, children and young people experiencing persistent disadvantage remain a key group of concern. However, since 2018, there has been an increasing number of pupils who are eligible for FSM due to Universal Credit (UC) transitional protections, rather than their financial circumstances. This means that our method for identifying persistent disadvantage is capturing a larger group of pupils than those in the deepest poverty. Just as FSM itself does not fully reflect those in income poverty, our persistent disadvantage measure no longer reflects those in long-term poverty. For example, a pupil who is eligible for free school meals in reception would remain FSM-eligible throughout their time in primary school – and hence be labelled as in persistently disadvantaged – even if their financial circumstances had improved significantly.  

By this year’s data, 2025, the scale of the UC distortions mean that we cannot meaningfully report on the persistent disadvantage gap, though we will return to this in future reports following the end of UC protections later this year.  

 

 

Explore the report