Another education article in The Observer misses the mark
One British broadsheet has become the home of sensationalist education reporting
I’m not saying The Observer is the worst paper for education reporting in the UK, but it’s definitely in the bottom one.
The Observer’s education reporting has been shocking lately.
Several stories have been atrocious enough for me to blog about. There was that story about exclusions in October, which turned out to be about suspensions and prompted me to blog about it here:
Blog post: Another misleading news story about “exclusions”
Then there was a series of reports suggesting a DfE conspiracy to crush dissent, based largely (but not entirely) on the fact that Subject Access Requests revealed DfE employees have been reading and discussing education social media.
I wrote three blog posts examining the claims made and the evidence supporting them:
Blog Series: Is the DfE keeping files on “education experts” who criticise their policies?
Another Observer education story, which I didn’t blog about, is worth noting for its sensationalism:
Schools in England block lessons on Middle East over fears of bias claims
This story insinuated that there was something disreputable about the fact that few schools study the Middle East in GCSE history. This is something which I suspect does not seem remotely disquieting, biased or inexplicable to history teachers.
Today’s story
Another Observer report has just appeared, which again attempts to create a narrative out of very little good evidence.
Strictest academy schools in England suspend 30 times more pupils than the national average
The story is based on a cherry-picked statistic that was published in July. This is then used to rehash a lot of old education stories and arguments, many of which I’ve written about before. So be warned, I will be referring you back to a lot of old blog posts.
200% suspension rates at ‘zero tolerance’ trusts in England could affect highly vulnerable children, experts warn
The suspension rate at some of England’s strictest academy schools is 30 times the national average, with experts warning that this is affecting some of the country’s most vulnerable children.
Feel free to read why I think the term “zero tolerance” is unhelpful.
Blog series: Are there “zero-tolerance” or “no excuses” schools?
More importantly, the statistics here need a lot of attention. The national average used in the report is the national average for all schools in 2021/22 which is 6.9%. This is the number of suspensions divided by the number of pupils. This is a poor choice of base rate because suspensions are far more common in secondary schools than in primary schools. Secondary schools have an exclusion rate of 13.96% compared to 1.42% for primary schools. Therefore, when the national average is used as a base rate, the average secondary school has a suspension rate that is over twice the national figure.
School suspension rates in secondary schools also vary greatly; some schools have almost no suspensions and some have many. This has a lot to do with circumstances rather than policy. For example, there are many girls’ schools and grammar schools with very low suspension rates. Geography seems to be a significant factor. Secondary schools in the North East of England have a suspension rate of 23.63%, almost three times as high as that of Outer London. Secondary schools in Yorkshire and the Humber have an exclusion rate of 21.49%. Disadvantage is also key. Secondary pupils qualifying for Free School Meals (a demographic frequently used when analysing the effects of disadvantage) have a suspension rate of 34.20%. FSM pupils in the North East, or Yorkshire and the Humber, have a suspension rate of over 50%. Please bear this in mind when we look at the statistical claims that follow:
An Observer analysis of the latest Department for Education (DfE) annual data on suspensions has found that dozens of multi-academy trusts had suspension rates that far exceeded the national average in a number of their schools in 2021-22. Two of the country’s biggest trusts, Astrea and Outward Grange Academies Trust (OGAT), had schools with a suspension rate of about 200% of their pupil numbers – although some pupils were counted several times because they were suspended repeatedly – compared with a national average of 6.9%.
A suspension rate exceeding 200% is incredibly high. However, how many schools had a suspension rate that high? Well, if a newspaper is reporting that both of “the biggest trusts” can be said to have “schools” like that, you’d probably assume that there were at least a few in each trust. Except, there were not. There was just one secondary school like this in each of those two trusts, and no others over 200% in any trust in the whole of England. The Observer was reporting on just two schools.
In the case of the Astrea Trust, their one school was a clear outlier, with a suspension rate three times as high as the Astrea school with the next highest rate. In both cases, geography and disadvantage seem significant. The Astrea school was in Yorkshire and the Humber and had 47.3% of pupils eligible for Free School Meals. The Outwood Grange school with the highest suspension rate was in the North East, with 57.5% of pupils on Free School Meals. Both schools appear to be in exceptionally challenging circumstances. They both appear to have lost their leaders since the 2021-22 academic year, and both schools have been rated “inadequate” by Ofsted. The Observer article seems to ignore context and judge large Multi-Academy Trusts based on what is probably the most challenging school in each trust. Heaven forbid that MATs be given credit for taking on a difficult task. Maybe the MATs are to blame, but I don’t see anyone else lining up to take on schools like these.
Is the rest of the article any better?
The rest of the article seems to be a collection of the greatest hits of bad education commentary and journalism from the last few years. Many of the points I have addressed elsewhere. Some of the points will have teachers rolling their eyes. I’ll start with highly disputable claims, then move on to parts of the story that are calling out for more context.
Reasons for suspension for disruptive behaviour in some schools can include not bringing your pencil.
This is a staple of anti-discipline rhetoric, but no teacher I’ve ever met has encountered it. I challenge anyone to find a school where this appears in the behaviour policy.
Maureen McKenna, a former education director at Glasgow city council who has been hired by London’s Violence Reduction Unit to drive down exclusions, said: “The most common reason for suspensions and exclusions is persistent disruptive behaviour, but in some schools that starts with not bringing your pencil or refusing to take your jacket off.”
In an interview on The World Tonight, Maureen McKenna claimed that her work was “not particularly about reducing exclusions”.
Schools can only exclude the most serious matters; “persistent disruptive behaviour” should not be mistaken for minor incidents.
Anne Longfield, chair of the Commission on Young Lives and the government’s former children’s commissioner for England, warned that poorer children and those with special educational needs were more likely to be suspended or excluded, but “it also depends on the school you go to”.
She said: “Young people who end up being excluded are vulnerable to being exploited by criminal gangs, and their life chances are diminished.”
The evidence strongly suggests that the relationship between criminality and exclusion is more complex than this. This blog post covers many of the dubious claims from The Commission On Young Lives:
Blog post: A new report on exclusions
Here’s some context for other parts of the story
Exclusion rates were nearly back to pre-pandemic highs, with 3,104 pupils permanently excluded from school that term [autumn term 2022], according to analysis of DfE data by the Centre for Social Justice (CSJ) thinktank.
This is true, but termly rates do vary a lot. Here’s the missing context:
The CSJ also uncovered data using the Freedom of Information Act showing that pupil referral units (PRUs), where excluded pupils are often placed, are over-subscribed. This summer there was an excess of more than 800 pupils enrolled in PRUs than there were places.
Figures based on collating FOI requests often aren’t reliable. However, the important context here is that any shortfall in places is likely to be mainly a result of reduced places, rather than high levels of exclusions. Between January 2019 and January 2023, the number of PRUs fell from 224 to 177, and the number of pupils enrolled in PRUs fell from 10494 to 7470.
[Maureen] McKenna oversaw a drop of 90% in suspensions over 14 years in Glasgow, which coincided with violence halving in the city.
Violence decreased throughout the UK. Referrals for youth crime were reduced by relatively less in Glasgow compared to the rest of Scotland. See these blog posts:
Blog Series: Did Glasgow cut youth crime by ending suspensions and expulsions?
OGAT, which runs 41 schools in the north, has attracted criticism for its strict behaviour policies, including the use of isolation booths. Former teachers made allegations in 2019 about assemblies in which pupils were routinely humiliated and made to cry to enforce discipline. At the time OGAT said they did not “recognise the culture described”.
The precise allegations made about OGAT kept changing. Again, I blogged about it at the time.
Blog post: School Chain Shaming
Blog post: More on School Chain Shaming
I’m old enough to remember when The Observer was a newspaper
The Observer is the only Sunday paper I have ever bought regularly, but this was many years ago. Even without these education stories, I would have been put off by the time (earlier this year) when they reported on dozens of alleged kidnappings that, according to their sister paper the Guardian, were never actually reported to the police. But currently, what puts me off the paper the most is how I keep being disappointed when I click on an interesting-sounding education story only to discover it’s The Observer’s latest alarmist nonsense.