The latest exclusion figures

We finally have the latest exclusion figures. These cover the academic year 2021/22. This was the first school year without a lockdown period since 2018/19. The figures that have been released include suspension figures, but I am more interested in the debate around permanent exclusions. Also, I will be looking at the claims that dominate the debate, i.e. whether exclusions are rising or discriminatory.

Every year it is claimed that exclusions are rising, regardless of whether they are or not. This year they actually are, although that is in the context of exclusions having reached an all-time low in 2020/21.

As we can see, exclusions are still far short of pre-pandemic levels. Further context is available; we can go further back if we look at the absolute numbers, rather than the percentage rate for exclusions.

Permanent exclusions have risen from the all-time low point in the previous figures, but by historical standards, they are in no way high. There were 6495 permanent exclusions from state schools in 2021/22. For a pupil population of well over 8 million, this is a rate of just 0.08%.

As for unfairness, those looking to claim that permanent exclusions disproportionately affect ethnic minority pupils are further away than ever from having good evidence for this. Despite the overall rate of 0.08%; the exclusion rate for ethnic minority pupils was just 0.06%. White British pupils had an exclusion rate of 0.09%, which means they are 47.5% more likely to be excluded than ethnic minority pupils. The last time ethnic minority pupils were more likely to be excluded than white British pupils was in 2010/11. A lot of coverage of this issue has been particularly concerned with whether there were discrepancies affecting black pupils. The latest figures are the first to show clearly that black pupils are less likely to be excluded than white British pupils (the previous 2 years showed black pupils to be only marginally less likely to be excluded). The last time black pupils were more likely to be excluded than white British pupils was 2018/19. The racial disparities that were blatant and alarming in the 00s have now disappeared. Much of the debate in the media has not caught up, although the Guardian, often the worst newspaper for repeating claims that black pupils are more likely to be excluded, reported accurately on these rates.

Regional differences in exclusions remain dramatic, with exclusions far lower in London and the South East. This probably reflects differences in socio-economic status between regions, although it should be noted that London politicians have been among the most vocal opponents of exclusion.

A curious point about this is that opponents of permanent exclusions have repeatedly used alleged disparities between white British and black pupils, or white British and ethnic minority pupils, as a reason to reduce exclusions. However, when looking at the regions of England, these racial disparities are found only where the exclusion rates are lowest. It is only in Inner London, where the exclusion rates are the lowest in England, that ethnic minority pupils are more likely to be excluded than white British pupils.

It is only in Inner London, Outer London and the South East (the three regions with the lowest rate of exclusions) where black pupils are more likely to be excluded than white British pupils.

While I won’t claim to know the factors affecting these differences, it seems apparent that low exclusion rates have not prevented the racial disparities that the advocates of lowering exclusions claim to be concerned about. We can probably conclude that preventing, or reducing exclusions, is not a policy that can be reasonably supported on grounds of reducing these disparities.

Beyond this, most patterns have remained as expected. There are some smaller ethnic groups with higher rates of exclusion than white British pupils. Some of these are due to what appears to be random variation. Some will be due to the way some of the most disadvantaged groups have consistently higher exclusion rates. SEND pupils remain more likely to be excluded, but this is mainly related to SEMH. FSM pupils, boys and older pupils are all much more likely to be excluded.

The reasons for exclusions have changed a little. This is the second year of figures that have allowed schools to give more than one reason for exclusion. I have calculated each reason as a percentage of total exclusions. This differs from the percentages published on the DfE website which are based on the total number of reasons given for exclusions.

The most noticeable changes are an increase in “Persistent Disruptive Behaviour” as a reason for exclusion and a decrease in “Use or threat of use of an offensive weapon or prohibited item”. The former might be a result of not having lockdown as a break in the persistence of the behaviour. I can’t think of an explanation for the latter. Debate rages over what “Persistent Disruptive Behaviour” means (or at least we continue to hear from non-teachers convinced it must mean something trivial). Still, it is noticeable that all the other major categories seem obviously very serious, with the two types of assault very well represented in the data.

Obviously, in this post, I’ve looked only at the issues that get the most media coverage. There may be exciting details in the figures that I have not yet spotted;  I’ll try to keep you informed about this. I will also be putting in Freedom Of Information requests for more detail about some of the issues in the data.

 

Toilets

Every so often there will be an outbreak of school toilet mania on Edutwitter. Occasionally, it might even occur in the media. It largely consists of people who have no idea what it’s like to work in a challenging secondary school, being shocked that access to the toilets is in any way restricted.

The cause of these outbursts seems to be a complete lack of awareness of all the problems associated with secondary school toilets. This is combined with an equally thorough lack of awareness of the fact that secondary school-age pupils may actually get some pleasure from skipping lessons to hang out with their friends, play on their phones or just avoid hard work. Some are oblivious to this, and to the fact that this can be the case regardless of the merits of the subject, the teacher, or the lesson. If kids can arse about in the cinema where 100s of millions of dollars have been spent to grab their attention and engage them, they can definitely choose to arse about out of lessons rather than learn how to simplify fractions.

I wrote about this issue before in this post. The summary of problematic toilet-related behaviours stands the test of time:

There are several reasons why teachers can’t just let students answer the call of nature:

• There is often a problem of internal truancy. Students who should be in lessons stay in the corridors. Sometimes they play games or attempt to intimidate passers-by. Every so often they disrupt lessons by running in to classrooms, or reaching in and switching off the lights. Often they write graffiti, or look for things to break. Any student allowed out of lessons risks swelling their ranks.
• There are students who will ask to go to the loo every single lesson. This is not an exaggeration. At my school students have to ask at an office for a key to let them into the toilets. One of the women working in the office reported seeing the same girl five times during the average school day. (And, no, the girl didn’t have a medical complaint other than a severe allergy to school work).
• There are classes where up to half the students will ask to go to the loo. In some year groups asking to go to the toilet when they are presented with hard work has become an automatic response. Once one child has asked, then many others will also ask. Sometimes many will have notes from their parents claiming a medical condition.
• There is an ongoing problem of toilets being vandalised. I mentioned before that school toilets are unpleasant. Much of this is down to vandalism, or actions that have been taken to prevent vandalism (like removing all soap or paper towels).

Also, I stand by this concluding comment:

Nobody is arguing for kids never being let out, even if they are going to soil themselves, or have a medical condition. All we are saying is that staying in the room is the normal expectation, and exceptions should be exceptional.

Does anything need to be updated? Has anything changed since I wrote that in 2008? I would suggest there are two big developments.

Firstly, vaping. In my time in teaching, schools went from tolerating smoking as a fact of life to almost completely eliminating it. However, schools are now being plagued by the slightly more fragrant version of it known as vaping. One of the biggest problems now with letting kids out to the loo is that some will go out to vape. If toilets are not monitored, they will become the place to vape, and this will happen in and out of lesson time. This is not just bad because of the addiction and the violation of rules. It is also the case that any place where illicit activity is happening is a place that will feel unsafe to many pupils. If vaping isn’t prevented, pupils will fear that bullying and violence also won’t be prevented. Additionally, a safe place for pupils to have their nicotine hit may also become a safe place for pupils to take other drugs.

The other difference from when I was writing before is the much greater concern about mental health. A lot can be said about how schools have often chosen unwise ways to respond to the perception of worsening mental health among pupils. But one of these, which seems to have appeared without much thought, is the idea that a vacation to the toilets can be a suitable response to anxiety. Not only have schools been known to issue toilet passes for this reason, but this has been suggested as a reason why schools should cease restricting access to the toilets. This seems to me to be one of those situations where the extent of a problem is unknown and the effectiveness of the response to the problem is unknown, but schools do it anyway because “something must be done”. I would argue that this is a bad policy. Any intervention to address mental health problems should be evidence-based and officially recorded. It should not involve reduced supervision of the pupil, nor should it provide obvious incentives to other pupils to claim to be unwell. If you have significant numbers of pupils leaving lessons due to their mental health, this is not an appropriate provision, and not what toilets are for. It is likely to obscure the extent of those difficulties, and potentially encourage others to adopt unhealthy patterns of behaviour.

Finally, I will add a few words about an argument I keep seeing, but didn’t mention in the previous post. People love to object to restrictions on going to the toilets, or indeed any school rule, on the grounds that it is not preparing pupils for adult life and is not like what happens in “the workplace”. While there are obvious questions as to why we should expect schools to treat children like they were adults in the workplace, I have another objection to these arguments. They always seem to make huge assumptions about what workplaces are like. While there are no doubt workplaces where you can pop out to the loo whenever you like, it is simply not the case that this is a universal rule. As a teacher, I don’t have any freedom to go off to the toilet during lessons. Plenty of other jobs involve constant activity or constant supervision of people or property. You can’t just leave a factory production line, or a queue of customers in a shop, whenever you feel like it. I would be here forever listing jobs, from brain surgeon to bus driver, where people cannot always go to the loo when they want. I complain often enough about silly opinions about teaching from people who have never worked in a classroom, but it is even worse when they have apparently never worked anywhere but places (presumably mainly offices) where people can use the facilities whenever they like.

Teachers are not monsters inconveniencing children for the sake of it. Restrictions on access to the toilets are in place for reasons regarding children’s best interests, and if you don’t understand those reasons your opinion on the matter is worthless.

Prog and Trad are back

Greg Ashman’s latest blog post (apologies if you can’t read it due to the paywall) was a review of education stories, and he found that many of them, including some from England, centred on the age-old split between educational progressives and educational traditionalists (or “progs” and “trads” as some started calling the sides in the 2010s). This strikes me as a good excuse to revisit the dispute.

The words “progressive” and “traditional” are more fairly applied to ideas than people. They describe two schools of thought about education that have existed implicitly since at least the time of Plato, and explicitly since the nineteenth century. Full descriptions of how they differ can be found in this post. You should also read this by Alfie Kohn to see how a very partisan progressive would describe the same distinctions.

Here’s my short description of the two philosophies:

1. Traditionalism is characterised by the explicit teaching of identifiable bodies of valued knowledge by teachers, who are in a position of authority over their pupils.
2. Progressivism refers to those educational movements that oppose or devalue either explicit teaching, identifiable bodies of knowledge, or teacher authority.

However, please be aware that this short description consists only of what I personally think is essential and is utterly insufficient to explain even those details of the debate that I discuss in this blog post.

Historically, both sides were able to identify the two strands of thought. However, in situations where progressives hold disproportionate levels of power or influence, they have a tendency to deny there’s any debate at all. Their values are the only values, and any apparent debate is actually just about working out a few details of the pedagogy, and not a significant difference in philosophy. As a result, progressives will sometimes use rhetoric about avoiding “tired old debates”, or they will even find reasons to object to people defining the terms. These reasons rarely get beyond: “These words make our side look worse”; “These words hurt our feelings”; “These words are used to criticise us”, or “You can’t define words this way”. And, of course, broad bodies of ideas can be described, identified, or approached in different ways. No definition of a broad set of beliefs can be definitive, or contain all information about the ideology to which they refer. It is possible to get into endless, spectacularly pointless, semantic discussions, where people object to how you’ve defined the terms, without giving even one example of the terms being used in a way that is inconsistent with the definition they are objecting to.

Of course, the fact that words have come to refer to particular bodies of ideas doesn’t mean they are the best words to describe those ideas and often the debate about definitions is confused by people who cannot tell the difference between debating the usefulness of the words (as a way to categorising the ideas), and the accuracy of the definitions. It is certainly possible to debate whether the terms progressive and traditional are helpful or not.

There is one obvious difficulty in the concept of educational progressivism: a progressive might not object to all three of explicit teaching, identifiable bodies of knowledge, and teacher authority. They might only object to one or two of that list and still fall within the usual understanding of the word “progressive”. Therefore, you could imagine progressives who object to different aspects of traditionalism and have virtually nothing in common. A progressive could object only to teacher authority, and as long as the school had no rules chosen and enforced by adults, perhaps with attendance in lessons entirely optional, they wouldn’t mind if the curriculum was conventionally academic or the pedagogy based on explicit instruction. Equally, another progressive could have no progressive inclinations about discipline at all, but still believe that the curriculum should favour skills over knowledge, or that all teaching should be by discovery. As I said, this is a point about the usefulness of the term “progressive”, it’s not a problem with the definition. There’s no convention in language that a word cannot describe a broad range of ideas. Describing somebody’s educational philosophy as “progressive” is a model of clarity and precision compared with, say, describing their politics as “liberal”. And the word “Protestant” gives us an excellent example of a word that covers multiple (often contradictory) ideas that are, nevertheless, grouped together because of what they reject (i.e. Catholic beliefs).

But even if there is nothing wrong in principle with describing these differing ideas as “progressive”, the question is whether it’s helpful. Would we be better off using another schema to describe educational conflicts? Should we replace the word “progressive” with several different words for different types of progressivism? Possibly, but nobody has yet invented such a framework that’s come into wide usage. Whenever people attempt to come up with alternative terms to describe philosophical differences (e.g., “constructivist” or “behaviourist”) they end up being at least equally contested. Alternatively, they become appropriated by all (e.g. “knowledge-rich” or “child-centred”) and, therefore, become expressions of approval rather than ideological distinctions. After more than a century, it seems progressive and traditional are not easily replaced.

This, of course, leaves open the question as to why. Why is it that we don’t constantly end up tying ourselves in knots confusing the different types of progressivism? I have a hypothesis here, and it would take considerable evidence to demonstrate it one way or another, so let me propose it only tentatively. My hypothesis is: If you can identify somebody’s educational stance as “progressive” you will be better able to predict their position on educational issues than you will by assuming that their beliefs are consistent or logical.

So for instance, if we assumed consistency or logic applied to educational beliefs, we would expect somebody who has a very high opinion of the capacity of children to direct their own learning or discover knowledge for themselves, to also hold high expectations of children in other respects. We would expect them to believe children are able to bring a pen to lessons, or manage their toilet habits well enough to avoid having to leave lessons. In practice, however, you find the same progressives arguing both for children’s brilliance at self-directed learning, and for their complete helplessness when it comes to being ready to learn.

For another example, we can look at people who argue against the enforcement of rules by claiming that bad behaviour results from trauma. For instance, this would include some of those who promote the work of Bruce Perry. Logically, you would expect them to be the first to object to the idea that bad behaviour results from inherited conditions like ADHD or autism. You might even expect them to argue (like Bruce Perry does) against the scientific consensus on ADHD. In practice, progressives will switch from saying bad behaviour results from trauma to saying bad behaviour results from undiagnosed “neurodiversity”, without even noticing they have changed the subject.

If we place a high premium on logic and consistency we might be surprised by other apparent contradictions. We wouldn’t expect people to promote both the idea of children thinking for themselves, and the importance of indoctrinating them with the latest political ideologies. We wouldn’t expect the people who claim permanent exclusions are unnecessary, as sexual assaults are so rare; to be the same people who claim schools are systematically misogynist, as sexual assaults are so common. We wouldn’t expect the people who claim children with disabilities cannot help being violent and disobeying rules, to be the ones who insist that children with disabilities should almost always be in mainstream schools. You wouldn’t expect the people who deny there is a progressive/traditional split in education, to be the ones who constantly attack others for being outdated traditionalists.

My experience is that progressives typically deal with the inconsistency between different progressive beliefs by professing two (or more) contradictory beliefs at once. If I am correct it becomes irrelevant that progressivism should logically divide into distinct subgroups of beliefs. An educational progressive will be more concerned about departures from progressivism than departures from consistency. It’s not that every progressive believes the same thing, but as far as I can see it is rare for progressives, even those with clearly contradictory beliefs, to argue with each other about those beliefs. Until that changes, we can probably expect the term “progressive” to be extremely useful to describe the beliefs of those who oppose traditionalism.

Comments are open. I would be extremely grateful if anyone can argue against my central hypothesis. Please note, I already accept that I am biased and that I have not tried to provide rigorous evidence for my hypothesis.

Who would make excuses for sexual violence in schools? Part 3

I wrote two previous posts about those who are campaigning for greater tolerance of sexual assaults in schools.

• Who would make excuses for sexual violence in schools? Part 1
• Who would make excuses for sexual violence in schools? Part 2

In the second post, I showed that there are those who find it objectionable to even mention that sexual assault should be a major consideration when discussing whether schools should have the power to exclude.

Here are a couple of stories they should read, both BBC news stories from cases where schools chose not to exclude, even where a pupil was believed to have committed a serious sexual offence.

From BBC news in 2017:

‘My daughter had to share a classroom with her rapist’

Rachel’s daughter was raped by a boy at her school. He was arrested, bailed, and put back in his normal lessons, alongside his victim, the following day.

“Somebody who’s been raped is already in a terrible place, but to be expected to be back in the same space as the rapist is just terrible,” she told the Victoria Derbyshire programme. “It’s re-traumatising – it’s just a terrible thing to do to a rape victim.”

The government says it is writing interim guidelines for schools to prevent schoolchildren being forced to share classes with pupils who have raped or sexually assaulted them, but campaigners say it is taking too long.

Rachel – not her real name – said her daughter’s anonymity was compromised at an early stage – which made life especially difficult.

“Being in the same classroom as the person that’s raped you is difficult enough, but when people in that room know what’s happened and they’re watching how you cope being in the same room as the rapist – that’s just awful,” she explained.

“It’s a whole extra layer of stress, knowing that these people are watching you – it’s just vile. It’s voyeurism gone mad.”

I realise there is a possibility that the first sentence is badly phrased, and it could be incorrect in implying the rape took place in the school. But even if that were not the case, it should be clear that very bad decisions were made here and that exclusion, if legally possible, should be on the table in such a situation.

From BBC news in 2009:

Teacher raped by boy wins damages

A teacher who was raped by a pupil will receive an undisclosed payment from her London school in compensation for the “truly horrifying” attack.

The teacher at Westminster City School was attacked by powerfully-built Dwayne Best, then 15, in September 2004.

At London’s High Court she argued that Best, the main suspect in a previous sexual assault at the school, should have been excluded prior to the rape.

The school insisted the rape could not have been foreseen or prevented.

The teacher, who cannot be named, was raped within a week of starting her job as she worked alone in a classroom after school hours.

On Monday, her counsel Neil Block QC told the court it was a “sustained and terrifying sexual assault” which left her unable to work as a teacher.

The court was told the teacher was unaware that a cleaner had been sexually assaulted three months previously – an attack linked to Best.

Best had a “most appalling” behavioural record in and out of school, and was said to be belligerent and aggressive.

Mr Block QC added: “Best should have been excluded much sooner than happened, which was after this incident.”

This case is the most compelling proof of the need for exclusions. At the time of the incident, schools were under tremendous pressure to reduce exclusions. It is likely that anyone involved in the decision not to exclude Dwayne Best after his first sexual assault would have been influenced by that pressure. Those who call for exclusions to end, or for even more pressure on schools to reduce exclusions, risk creating another case like this.

In my final post on this topic I will look at a case where a school did the correct thing, and excluded to protect the alleged victim of sexual assault, only to face legal challenges from those who are campaigning against exclusions.

Exclusions, ethnicity and disadvantage

Last year I looked in detail at the data on exclusions and ethnicity.

• Why do some ethnic groups have higher exclusion rates? Part 1
• Why do some ethnic groups have higher exclusion rates? Part 2

I found that while there was a time when ethnic minority pupils were more likely to be excluded than white British pupils, ethnic minority pupils have had a noticeably lower exclusion rate since 2014/15. At this point, a gap continued to exist between white British pupils and black pupils and then, after years of falling exclusions for black pupils, this closed in 2019/20. The media, however, continued to claim otherwise.

If you slice the data up more finely, you can find ethnic groups with a consistently higher exclusion rate than white British pupils. But far from being representative of ethnic minorities, only 11% of ethnic minority pupils come from those groups. Combined with the fact that very few pupils are excluded, this means that we should be very careful assuming that the factors that cause the disparities are relevant to all ethnic minority pupils. Nor should we assume that the factors affecting the exclusion rate for a specific ethnic group are relevant to all pupils in that ethnic group. Nor should we assume that the effects of those factors are system-wide.

The groups with a risk of exclusion that is consistently higher than that of white British pupils were:

• Any other black background (i.e. any black background other than black Caribbean or black African)
• Black Caribbean
• Gypsy Roma
• Traveller of Irish heritage
• Mixed – White and Black Caribbean

Despite much of the media coverage of exclusions focusing on differences between black and white pupils, there is some variety here. Two of these groups are classified as white; two are classified as black, and one as mixed race. But what they do have in common is that they are relatively small groups (ranging from 6,140 Irish traveller pupils to 129, 123 mixed white and black Caribbean pupils) and they are more disadvantaged. Whereas 20% of white British pupils qualify for Free School Meals, these 5 groups range from 32.8% on FSM (any other black background) to 63.5% on FSM (Traveller of Irish heritage). They have the 5 highest FSM percentages out of the 18 ethnic groups. This may also be part of the reason these groups are small; one of the reasons to identify a particular group in the figures is because they experience more disadvantage than other groups they might otherwise be included with.

The fact that these five groups have very high FSM rates is strongly indicative of the possibility that racial disparities in exclusion rates are actually disparities based on poverty. It is here that we face the limitations of the data. We do not know the levels of income or wealth of parents. We have only two measures of disadvantage. One is FSM and that is generally regarded as effective at an aggregate level (i.e. identifying schools or demographics with a high level of disadvantage) but less accurate at an individual level. When looking at ethnic groups there are no guarantees that, for a given level of poverty, all ethnic groups are equally likely to claim Free School Meals. And even if we could be sure that FSM pupils from different ethnic groups had similar levels of disadvantage, it is extremely unlikely that non-FSM pupils are all comparable. At the time of the 2020/21 exclusion figures, 79% of pupils were not eligible for FSM. We would expect the non-FSM pupils in some ethnic groups to be from far wealthier backgrounds and to come from higher-earning households compared with the non-FSM pupils in some other ethnic groups.

The other method of identifying disadvantage is through data identifying pupils from poorer neighbourhoods by their postcodes. I am skeptical as to whether this data will cast any light on the issue. Partly this is because I suspect the numbers in these figures would be very small. More importantly, though, I would assume that because of the different geographical dispersions of ethnic minority groups, the relationship between postcodes and disadvantage may vary between different ethnic minorities.

Given these reservations, the one set of figures I suspected might be most useful in taking account of disadvantage when looking at the exclusion rates for different ethnic groups, is the figures for FSM pupils. Do the disparities in exclusion rates between ethnic groups persist when we look only at these pupils?

Unfortunately, these figures are not published, but I made a Freedom Of Information request for them. Using this data I compared white British pupils on FSM and the 5 groups that had consistently had higher exclusion rates than white British pupils. Only Gypsy Roma pupils were at greater risk of exclusion than white British pupils on FSM.

There are 1,075,766 white British pupils on FSM, whereas there are only 25,836 gypsy Roma pupils (and only 307,959 pupils in all 5 “high-risk” ethnic groups combined). If the reason for looking at exclusion rates and ethnicity was to identify pupils at the greatest risk of exclusion, then looking at white British FSM pupils would clearly be more useful than looking at ethnicity.

Of course, if the reason for looking at exclusion rates and ethnicity is to identify evidence of unfairness, or racism, this would not be the appropriate comparison. It would be more appropriate to look only at FSM pupils in each of the 5 high-risk groups. In the 2020/21 figures for FSM pupils, 3 of those groups (Black Caribbean FSM pupils, traveller of Irish heritage FSM pupils, and Any other Black Background FSM pupils) had a lower rate of exclusions than white British FSM pupils.

Some of this might be the greater variability you get when looking at smaller numbers, but it is noticeable that among FSM pupils membership of one of the black ethnic groups in the statistics is never a risk factor for exclusion. This is despite the great emphasis on black pupils in campaigning about exclusions.

For some, race and ethnicity will always be the most important factor when considering exclusions. There are clearly still questions here about how gypsy Roma and mixed white and black Caribbean pupils have such high exclusion rates. There are also questions about why different ethnic groups differ so much in their FSM rates. But differences in exclusion rates due to ethnicity increase the risk of exclusion for remarkably few ethnic minority pupils. The big question should be why, on average, being from an ethnic minority lowers the risk of exclusion. Beyond that, focusing on race and ethnicity may have distracted us from the much bigger influence of disadvantage.