You may recall from this post that the Guardian misleads its readers about exclusions.
And today they have a new story on the topic:
‘It takes too long to get support’: alarm over rising primary school exclusions
It claims (in the sub-headline):
Parents and teachers call for action as figures show more young children in England being permanently excluded
The actual data for permanent exclusions in primary schools can be found here. The incredibly low level of permanent exclusions in primary schools means that the changes are unclear due to rounding, so let me calculate the exclusion rate for primary schools more precisely and plot it on a graph.
No, I don’t see that as a rise at the end either, even if we acknowledge that the most recent fall was due to lockdown. After an anecdote about a year 1 pupil being permanently excluded (not something that happens very often) we discover how the writer has concluded there has been a rise.
The number of permanent exclusions from primary schools in England has been rising steadily, from 610 a year in 2010 to 1,067 by 2018-19 (the last year of reliable figures).
The 2019/20 figures are not “unreliable”; they are just affected by lockdown and 1067 in 2018/19 is actually down from 1253 in 2016/17, so that’s not a steady rise. It’s also misleading to measure from 2010/11 which is a historic low point. This strategy (measuring change over time from an extreme value) is well known as a way to deceive, and you can read more about it here.
The article then points out:
In the autumn term of 2019, just before the Covid crisis, data recently published shows that permanent exclusions from primary schools rose by 20% to 455 compared with the same term the previous year.
Potentially this could be indicative of a rise. But the problem is, the term by term data was only introduced to make precisely this comparison and before autumn 2018 we don’t have it. This means we are comparing 2 data points for exclusions in the autmn term without any idea of how much the level of exclusions in the autumn term fluctuates from year to year. To put it in context, this is the graph the DfE published to show the termly data.
If that orange line isn’t causing you immediate worry it’s partly because the change (77 pupils) is tiny when you consider the total number of exclusions in a term across all schools, but it’s also that we have very few data points and termly exclusions seem fairly unstable. There are 16791 state funded primary schools in England (on last count). A change from 378 permanent exclusions in an autumn term to 455 is tiny compared with the number of schools involved, and is only noticeable because exclusions in primary are so rare. Anyone trying to spot a trend from this could be fooling themselves.
I don’t want to dissent from the suggestion that there may be a need to do more to support children with behaviour problems in primary schools. This might be true. But if you look at the data for primary schools in 2018/19 (the most recent non-lockdown year) we are looking at 1067 permanent exclusions among 4727089 children in 16784 schools. The average primary school could expect only 1 permanent exclusion in 15 years. Is this really a cause for alarm?
Scenes From The Battleground
The latest Guardian article on exclusions
September 18, 2021You may recall from this post that the Guardian misleads its readers about exclusions.
And today they have a new story on the topic:
‘It takes too long to get support’: alarm over rising primary school exclusions
It claims (in the sub-headline):
The actual data for permanent exclusions in primary schools can be found here. The incredibly low level of permanent exclusions in primary schools means that the changes are unclear due to rounding, so let me calculate the exclusion rate for primary schools more precisely and plot it on a graph.
No, I don’t see that as a rise at the end either, even if we acknowledge that the most recent fall was due to lockdown. After an anecdote about a year 1 pupil being permanently excluded (not something that happens very often) we discover how the writer has concluded there has been a rise.
The 2019/20 figures are not “unreliable”; they are just affected by lockdown and 1067 in 2018/19 is actually down from 1253 in 2016/17, so that’s not a steady rise. It’s also misleading to measure from 2010/11 which is a historic low point. This strategy (measuring change over time from an extreme value) is well known as a way to deceive, and you can read more about it here.
The article then points out:
Potentially this could be indicative of a rise. But the problem is, the term by term data was only introduced to make precisely this comparison and before autumn 2018 we don’t have it. This means we are comparing 2 data points for exclusions in the autmn term without any idea of how much the level of exclusions in the autumn term fluctuates from year to year. To put it in context, this is the graph the DfE published to show the termly data.
If that orange line isn’t causing you immediate worry it’s partly because the change (77 pupils) is tiny when you consider the total number of exclusions in a term across all schools, but it’s also that we have very few data points and termly exclusions seem fairly unstable. There are 16791 state funded primary schools in England (on last count). A change from 378 permanent exclusions in an autumn term to 455 is tiny compared with the number of schools involved, and is only noticeable because exclusions in primary are so rare. Anyone trying to spot a trend from this could be fooling themselves.
I don’t want to dissent from the suggestion that there may be a need to do more to support children with behaviour problems in primary schools. This might be true. But if you look at the data for primary schools in 2018/19 (the most recent non-lockdown year) we are looking at 1067 permanent exclusions among 4727089 children in 16784 schools. The average primary school could expect only 1 permanent exclusion in 15 years. Is this really a cause for alarm?
Share this:
Posted in Commentary | 1 Comment »