What does the latest data tell us about school exclusions?

The data is published by the Scottish Government on a biennial basis, and is freely available online. The most recent figures are for the 2022/23 school year, the 2020/21 data is of limited use because of the obvious impact of Covid-era school closures on exclusion rates. This means that in most cases, exclusions appear to have increased according to the most recent publications, but a more accurate picture can be seen by comparing information from 2022/23 to the data from 2018/19, the last undisrupted year.

Finally, it is worth noting that although exclusion data is considered reliable, concerns have been raised about the use of informal exclusions in Scottish schools, especially in relation to pupils with additional support needs or those looked after by the care system. These would not be reflected in the official data.

The national picture

Across Scotland, exclusions from school have fallen dramatically over the past fifteen years – but prior to that they had been steadily increasing.

Between 2002/03 and 2006/07 the exclusion rate per 1000 pupils rose from 49 to 64, but from then on it declined year on year, usually by significant margins.

As a result, the exclusion rate for the country as a whole now stands at 17 per 1000 pupils, which represents a drop of nearly 75%.

Exclusion rates by local authority

The first thing to notice about the local authority data on exclusions is that there are enormous variations across the country, but it’s worth adding a small caveat here: there is often a great deal of variation within councils as well as between them.

When looking at the 2022/23 data, we can see that while East Renfrewshire had 3 instances of exclusion per 1000 pupils, the figure for Aberdeenshire was ten times that at 31 per 1000 pupils.

The local authority data over time, however, shows that many councils have seen significant reductions in exclusions rates over the past fifteen years.

In 2008, Dundee City Council were reporting 112 instances of exclusion per 1000 pupils. Glasgow (97), North Ayrshire (91), and Inverclyde (86) councils weren’t too far behind. At this stage the national figure for exclusions was 58 per 1000 pupils.

Since then the picture has changed dramatically.

Having had one of the highest rates of exclusions in 2007/08 (77 instances per 1000 pupils), the data for Clackmannanshire shows that no pupils were excluded in the 2022/23 school year. Argyll and Bute has gone from 61 exclusions per 1000 pupils in 2007/08 to just 10 per 1000 pupils now. Large reductions were also recorded in Dundee, Glasgow, North Ayrshire, Inverclyde, West Dunbartonshire and West Lothian. Only Orkney reported 2022/23 exclusion levels higher than those from 2007/08.

One of the most interesting changes in council exclusion data is that the amount of variation between councils has significantly reduced, reflecting policy shifts driven by the increasing national focus both on exclusion itself and on the educational experiences of groups like young people from deprived area.

Read more:

Reducing school exclusions needs a system-wide approach

Detail of exclusion

The government’s data allows us to analyse some of the specifics behind the instances of exclusion across the 2022/23 academic year.

We know, therefore, that the overwhelming majority of school exclusions last for just one or two days – in fact, this accounts for 70% of all exclusions across the country.

We can also see how exclusion rates are affected by the structure of school terms thanks to a breakdown of exclusions per week across the academic year. Mapped out, this shows what many would expect: exclusions rise after holidays but begin to fall before them, suggesting that, in at least some cases, schools may hold off processing an exclusion if a natural break in the year is imminent.

Another important aspect of exclusion is the reason behind it, and the official data also gives us an insight into this area. Whether looking at actions against fellow pupils or actions against teachers, an assault with no weapon is by far the most common reason for pupils to be excluded from Scottish schools, followed in both cases by the threat of violence with no weapon.

Less-common reasons for exclusion include theft and the sending of malicious communications, and a very small (but nonetheless serious) number of exclusions are for threats of sexual violence.

Breakdown by characteristic

Another way in which we can try to contextualise exclusions data is by analysing it in relation to information about characteristics such as additional support needs, deprivation, ethnicity and care experience. In each case, trawling through the data highlights some clear concerns.

ASN

Government data includes a wide range of additional support needs including dyslexia, autism spectrum disorders, hearing impairments, being a young carer, English not being a first language, bereavement, being a ‘more able pupil’ and quite a bit more. One reason for a pupil to have additional support needs is that they are ‘at risk of exclusion’.

By definition, pupils in this category have additional needs that must be met in order for them to access education and have the same opportunities as other pupils. Yet pupils with a registered additional support need are much more likely to be excluded from school than those who do not.

For pupils with no ASN, the latest data shows an exclusion rate of 7 per 1000 pupils, but for those with some form of ASN the rate is 35 per 1000 pupils.

The data can also be broken down even further to differentiation between those who spend all, none, or some of their time in mainstream classes. The latter group comes out worst by a significant margin – there are 92 exclusions per 1000 pupils when it comes to those who spend some, but not all, of their time in a mainstream environment, compared with just 34 per 1000 pupils spending all of their time there.

The particular problem with ASN gaps is that they are a direct indictment of education policy, because it’s simple really: if large numbers of pupils with additional support needs are being excluded from school, then large numbers of pupils are clearly not having their needs met.

Read more:

What are additional support needs and who has them?

Deprivation

Another key factor when analysing exclusions is the relative deprivation of pupils. Official publications provide two main sets of data in this area.

The most recent figures show that those from the most deprived 20% of Scotland are three-and-a-half times more likely to be excluded from than those from the most affluent fifth of the country.

It’s a stark statistic based on the breakdowns from the Scottish Index of Multiple Deprivation (SIMD). This system breaks the country up into nearly 7000 small areas which are then categorised based on factors such as income levels, educational provision and crime. SIMD is extremely useful but, in this context, suffers from a key limitation: it looks at geography, not individual human beings.

However, free school meals data is based on individual circumstances, and it paints a strikingly similar picture that makes the pattern undeniable. Pupils who did not receive free school meals were excluded at a rate of 18 per 1000; those who did experienced an exclusion rate of 55 per 1000.

The data is absolutely clear: poorer pupils are far more likely to be excluded than richer ones.

You might be wondering why any of this is news, because of course kids living in poverty are more likely to be excluded, right?

Well, ok – but hang on a minute. If a pupil’s socio-economic background – for which they are not, and cannot be, responsible – is such a powerful factor in exclusions that we don’t even really think about it, then perhaps we need to ask ourselves the extent to which excluding deprived pupils means punishing children for things that aren’t their fault?

Ethnicity

Exclusion data based on pupil ethnicity is also published by the Scottish government, although it attracts less attention for one main reason: the overwhelming majority of pupils in Scotland are identified as being from a single ethnic group. Of the 705,528 pupils identified in the 2022/23 census, more than 70% were ‘White – Scottish’. The next two largest groups were ‘White – British’ and ‘White – Other’, and combined all three account for nearly 85% of young people in Scottish schools.

The fact that ethnicity is either ‘Not known’ or ‘Not disclosed’ for a significant minority of young people also causes problems, while the relatively low variations between groups means that this data is often of little interest.

But it does offer one very striking insight into exclusions across Scottish schools, and that is the impact on the Gypsy/Traveller community, whose children are far more likely to be excluded than any other group.

'Looked after' children

Analysing exclusions of looked after young people turned out to be more complicated than was the case for other categories. The data for this group is gathered and provided in a different way, and as a result the most recent figures are not for 2022/23 but rather for 2020/21 – the year in which school closures had an obvious effect on exclusion rates.

Given the pattern across other areas, where a dip in 2020/21 was followed by an increase two years later, it is possible that exclusion figures for looked after children will show an increase when they become available later this year; however, the overall trend may still be one of consistent decline.

In 2009/10, the exclusion rate amongst all pupils was 45 per 1000, but for looked after pupils it was 397 per 1000. That is not a typo.

Read more:

Agenda: We must end exclusions for care experienced children

Since then, the figure has markedly declined and, as of 2018/19, stood at 152 per 1000 pupils. This is, clearly, an extremely significant reduction, but it is still nearly seven times the exclusion rate recorded for pupils who are not looked after.

In Scotland, the recent independent care review stated that the ‘formal and informal exclusion of care experienced children from school must end’, but an investigation by Who Cares? Scotland found that most councils have not set a target date to achieve this. Careful analysis of the available data also reveals that some, but not all, care experienced young people receive specific additional support on that basis.

Finally, we should note that current data does not actually relate to ‘care experienced’ young people – instead, it focuses on those who are classed as ‘looked after’. The gap between the definitions for these terms has significant implications going forward.