Why do some children get excluded from school?

Last Updated on October 23, 2026 by Karl Thompson

In this post I review a range of evidence on school exclusions – from the official statistics to research on ‘off-rolling’ and case studies of excluded pupils – and consider what each type of evidence tells us about why some pupils are excluded, and its limitations.

Exclusions aren’t named on the A-level sociology specification, but the topic is relevant to the sociology of education (in-school processes, labelling, and the effects of marketisation), to crime and deviance, and to research methods, as a good example of the strengths and limitations of official statistics.

Key definitions

• Permanent exclusion – a pupil is removed from the school and will not come back to it.
• Suspension – a pupil is excluded from school for a set period of time, then returns. This used to be called a ‘fixed-period exclusion’. A pupil can be suspended more than once, so the suspension rate counts suspensions, not the number of different pupils suspended.
• Off-rolling – removing a pupil from the school roll without a formal exclusion, for example by persuading parents to withdraw their child or move them elsewhere, when the move is mainly in the interests of the school rather than the pupil. Because no exclusion is recorded, off-rolling doesn’t show up in the exclusion statistics.

A: The official statistics on exclusions

The Department for Education publishes annual statistics on suspensions and permanent exclusions in England. The latest edition, published in July 2026, covers the 2024/25 school year.

	2024/25
Permanent exclusions	9,900 (rate 0.12%, or 12 per 10,000 pupils), down 9% on the year before
Suspensions	913,000 (rate 10.88%, or 1,088 per 10,000 pupils), down 4%
Most common reason for permanent exclusion	Persistent disruptive behaviour (40%), then physical assault against a pupil (16%) and against an adult (12%)
Secondary schools’ share	84% of permanent exclusions; 85% of suspensions

Certain groups of pupils are far more likely to be permanently excluded:

• Pupils eligible for free school meals had a permanent exclusion rate of 0.32, against 0.05 for other pupils – more than six times higher.
• Pupils receiving SEN support had a permanent exclusion rate of 0.37, against 0.06 for pupils with no identified special educational needs.
• Boys’ permanent exclusion rate (0.16) was more than double girls’ (0.07).
• Gypsy/Roma and Traveller of Irish Heritage pupils have had the highest exclusion rates over time, and Chinese and Indian pupils the lowest.

What do the official statistics suggest about why pupils are excluded?

Two things stand out. First, pupils usually have to do something serious to be permanently excluded – persistent disruption or assault. Second, exclusion is strongly patterned: poorer pupils, pupils with special educational needs and boys are far more likely to be excluded.

Taken at face value, this suggests we should look at what it is about these groups of pupils that leads to the behaviour that gets them excluded – material deprivation, unmet special educational needs, or laddish subcultures, for example. HOWEVER, the statistics only tell us what schools recorded. Labelling theorists would ask whether some groups are more likely to be seen as disruptive, and so more likely to be excluded for the same behaviour.

B: Off-rolling – the exclusions that don’t appear in the statistics

The official figures don’t include pupils who leave a school without being formally excluded. Two pieces of research suggest this hidden group is large.

Ofsted’s analysis (2018)

Ofsted looked at pupils who were in Year 10 in a state-funded secondary school in 2016 but weren’t in the same school in Year 11 in 2017. It found that:

• over 19,000 pupils – about 4% of the year group – did not progress from Year 10 to Year 11 in the same school
• 2,900 schools lost some pupils between Years 10 and 11; Ofsted’s model flagged 300 schools with particularly high levels of leavers in both of the last two years
• pupils with special educational needs, pupils eligible for free school meals, looked-after children and some minority ethnic groups were more likely to leave – about 30% of leavers had special educational needs, against 13% of all pupils
• about half of the leavers did not reappear in another state-funded school; more than a quarter went to alternative provision or pupil referral units
• academies, particularly some in multi-academy trusts, appeared to lose proportionately more pupils, and a higher proportion of London schools saw pupil movement.

FFT Education Datalab: Who’s Left? (2018)

Researchers at FFT Education Datalab used the National Pupil Database to track cohorts of roughly 550,000–580,000 pupils through secondary school. For the most recent cohort they studied, they found that about 8,700 pupils had moved into alternative provision and around 22,300 had left state school rolls altogether. After allowing for pupils who moved to independent schools or emigrated, they estimated that between 6,200 and 7,700 pupils were still in England but either didn’t take exams at 16 or weren’t counted in school league tables.

What does this add to our understanding of exclusions?

This research suggests the official exclusion statistics are the tip of the iceberg. Some schools appear to remove pupils without recording an exclusion – and the pupils most likely to leave are the same groups most likely to be formally excluded. This puts more of the focus on schools rather than pupils: in a system where schools are judged on their results, removing pupils who are likely to do badly before they sit their GCSEs improves the school’s figures. That makes off-rolling a clear example of an unintended consequence of marketisation and league tables, and closely related to Gillborn and Youdell’s idea of educational triage.

C: The limitations of the official statistics on exclusions

The exclusion statistics are a good example of why official statistics need to be treated with caution:

• Validity: they don’t capture informal exclusions or off-rolling, so they understate how many pupils are pushed out of mainstream schools.
• Little detail on reasons: only the main reason recorded by the head teacher is given. There is no information on the specific incident, the pupil’s history or the wider context. Labelling theorists would argue that the ‘reason’ is itself a judgement, not a neutral fact.
• Social construction: whether behaviour leads to exclusion depends on the school’s policies and on how staff interpret it, so differences between groups may partly reflect how they are treated rather than how they behave.
• Time lag: the figures are published after the end of the school year they cover, and many exclusions follow a long history of earlier incidents, so the data describe events that may be well in the past by the time you read them.

Independent research like Ofsted’s and Datalab’s, which tracks the same pupils over several years, can give a more valid picture than the annual exclusion figures – a good example of the value of longitudinal data.

D: Qualitative evidence – the experiences of excluded pupils

Case studies of pupil referral units, such as the BBC’s documentary series filmed over a school year in an alternative provision academy in West London (programme website), give a very different, more interpretivist picture. Interviews with excluded pupils often reveal complicated back stories – family breakdown, trauma and chaotic lives – which the statistics can’t show.

Sources like these tend to make us more sympathetic to excluded pupils than the official statistics do, and they show why a mixed-methods approach gives a fuller understanding. HOWEVER, a documentary is a secondary source made for television, so the pupils shown may not be typical, and the material is selected and edited for an audience.

Why are some pupils excluded? Bringing the evidence together

Each type of evidence points to a different explanation:

• The official statistics point to the characteristics of pupils – poverty, special educational needs and gender.
• The off-rolling research points to schools, and to the pressures of league tables and marketisation.
• The qualitative case studies point to pupils’ lives outside school.

The best answer combines them: pupils facing difficulties outside school are more likely to behave in ways schools find hard to manage, and schools under pressure to improve their results have reasons to remove them – formally or informally. This is a good example of in-school and outside-school factors interacting, which is exactly the argument the strongest exam answers make.

Revision bundle

If you like this sort of thing, then you might like my sociology of education revision bundle, written by Karl Thompson, who taught A-level sociology for eighteen years and spent twelve as an AQA examiner, marking all three papers on the current specification.

Signposting

Related posts:

• Official statistics in sociology – strengths and limitations as a research method.
• Gillborn and Youdell: Rationing Education – educational triage and league tables.
• Labelling theory and the self-fulfilling prophecy

Please click here to return to the homepage – ReviseSociology.com

Sources

• Department for Education (2026) Suspensions and permanent exclusions in England, 2024/25 (9 July 2026).
• Ofsted (2018) Off-rolling: using data to see a fuller picture, Ofsted blog, 26 June 2018.
• FFT Education Datalab (2018) Who’s Left? 2018, using the National Pupil Database.