Inclusion and the Special Needs Racket

February 27, 2010

Some months back, I met the education spokesperson from the opposition group of my local council. Naturally, I mentioned that, of all the policies pursued locally, inclusion was the most disastrous.

“No,” she assured me, “All the evidence shows that students with special needs are better off in mainstream schools.” Unfortunately I didn’t get a chance to enquire what evidence.

Now, every so often I get into a discussion online about inclusion. Examples include here, here and (now) here . (Please note the calmness, good humour and respect for the opinions of teachers shown by the pro-inclusion side). I have also talked about this issue in a number of blog posts such as here , here and here.

To sum up: the disaster that is inclusion has seen many children who are unwilling or unable to learn forced onto our schools. The main provision for this extra responsibility has been the special needs racket, the process in which a huge amount of paperwork is generated and children are labelled with medical or psychological conditions they usually don’t have by people with no medical or psychological training. Students are routinely selected for labelling if they are badly behaved, academically weak or have pushy parents. Sometimes (for instance in the case of ADHD) we can’t even be certain the conditions exist in any real way, rather than as a list of behaviours which may be caused by a multitude of different factors including deliberate choice by a student to act in a particular way.

Once the system is in operation most effort appears to go into maintaining mountains of paperwork. However, SEN staff also organise and teach withdrawal lessons, support students in lessons, interfere with teachers (particularly over discipline) and generally find ways to lavish attention on students, particularly the badly behaved. In my experience it doesn’t seem to help students with genuine needs and in many respects it does amount to spoiling and encouraging the badly behaved. Every teacher I know has a story of horrific stupidity by an interfering SENCO or SEN teacher, often involving appalling behaviour being excused.

Anyway, the reason I am returning to this subject is that I have recently become aware of some of the research into the issue. It had been known for quite a long time that the more inclusive a school was (i.e. the more SEN students, statemented or unstatemented, a school had) the worse its GCSE results would be (Lunt, 1999). However this correlation could always be explained away as it did not look at the results of SEN students in particular or consider value-added results. After several years the government actually commissioned some research (Dyson, 2004) which considered more detailed data. A group of academics used data from the National Pupil Database from 2002 and used advanced statistical techniques (multi-variable regression analyses and multi-level modelling, for all you geeks) to identify the effects of inclusion on attainment.  They looked for any effect “inclusivity” (i.e. the proportion of students with SEN) had on Average Point Scores in the various key stage exams. They considered three categories of students. Those without SEN, those with SEN but not statemented for it, and those statemented for SEN.

They found that for Value Added (in this case calculated as the difference between a student’s APS score and what would be expected from a student with their prior attainment):

“At KS 2 the analysis suggests that, as school inclusivity increases one unit (in percentages), a school’s average pupil VA score decreases by 0.018.
At KS 3 the analysis suggests that, as school inclusivity increases one unit (in percentages), a school’s average pupil VA score decrease by 0.111.
At KS 4 the analysis suggests that as, school inclusivity increases one unit (in percentages), a school’s average pupil VA scores decrease by 0.026.” page 137

And for Average Point Scores:

At Key Stage 1:

“It appears that the effect of increasing a school’s inclusivity has a negative effect on APS scores and this effect is more pronounced for pupils with statements.” page 143

At Key Stage 2:

“there is a strong relationship between a school’s inclusivity measure and its APS score…It appears that the effect of increasing a school’s inclusivity has a negative effect on APS scores and this effect is slightly more pronounced for pupils with statements.” pages 145-6

At Key Stage 3:

“there is a strong relationship between a school’s inclusivity measure and its APS score…It appears that the effect of increasing a school’s inclusivity has a detrimental effect on APS scores and this effect is particularly pronounced for pupils with no special needs.”  pages 149-151

At Key Stage 4:

“There is a strong relationship between a school’s inclusivity measure and its GCSE & GNVQ point score. As school inclusivity increases, GCSE & GNVQ point score decreases… It appears that the effect of increasing a school’s inclusivity has a detrimental effect on GCSE & GNVQ particularly for the non-special needs group.” pages 152-155.

It would be nice to think that this would have been written up as definite proof that that inclusion was harmful for all concerned and politicians would have rushed to reverse the policy. But, of course, this was a government report and it is not a good idea to write a report that shows a complete failure of government policy, and so by the time the authors had taken into account case studies, anecdotal evidence, the size (rather than the direction and probability) of the effect and the reasons for inclusion other than academic benefit, they claimed in the Executive Summary that:

“There is nothing in our findings to suggest that the current national commitment to maintain pupils in mainstream schools wherever possible is likely to have a significant impact on overall levels of attainment at national and LEA level, nor that it need have an impact at school level.” page 13.

I found the statements I have quoted above by sifting through Appendices III and IV (the graphs in Appendix IV make things particularly clear). I wonder if any of the education ministers at the time bothered to do that? I only recently discovered the report existed (when it was used as evidence in favour of inclusion) so it clearly hadn’t been shouted about from the rooftops.

The second piece of research I want to bring to your attention is a more recent study. Another government commissioned report looked into the use of teaching assistants. In two waves, data on thousands of students along with observations and surveys were collected and used to see what effect teaching assistants had. On the whole it confirmed what most teachers already knew, they can be helpful to teachers, and improve discipline and atmosphere within a class. However, if we recall that in most schools they are deployed specifically to assist students with SEN then, the effect they have on the achievement of the students they directly support, would tell us a lot about how SEN provision in our schools works. The research found:

“• At Wave 1 and 2 there was there was a consistent negative relationship between staff ratings of the amount of support a pupil received and the progress they made in English and mathematics, and at Wave 2 in science. The more support pupils received, the less progress they made, even after controlling for other factors that might be expected to explain the relationship such as pupils’ prior attainment, SEN status and income deprivation. A similar though less marked trend was found with measures of the amount of support taken from the systematic observation data.

• Further analyses showed that the negative effect of support was not attributable to pupils who were making less attainment progress being allocated more support over the year, and results were not attributable to any bias resulting from missing data.

• There was evidence that unsupported pupils in year 9 made less progress in those classes that had a higher proportion of pupils receiving support.”

Now, there is plenty of other evidence in the report about all the good teaching assistants do. They only fail in one important respect: improving the academic achievement of the students they support.

So, with regards to the policy of inclusion and the special needs racket. We now have large scale statistical evidence that including kids with SEN in mainstream schools has a negative effect on their attainment and a negative effect on the attainment of others without SEN. We also know that the most expensive part of inclusion, the employment of large numbers of teaching assistants (116,000 more since 1997 according to the party of government) to support students in lessons actually only harms the achievement of SEN pupils. Now these are the government’s own studies, coming up with hard data that shows that the policy of inclusion and the special needs racket are harming the achievement of the very children it was meant to be helping. This is not a surprise to those of us who have seen it causing this harm right in front of us. We can see that more and more kids with SEN in the mainstream lowers expectations. We can see that some of the students who have received the most support through the deployment of teaching assistants have become helpless to the point where they won’t pick up a pen without asking for help first. The question is: will any politician, of any party, have the courage to say this disastrous policy isn’t working and bring it to an end?


Blatchford, P., Bassett, P., Brown, P., Koutsoubou, M., Martin, C., Russell, A. and Webster, R. with Rubie-Davies, C.. Deployment and Impact of Support Staff in Schools: the Impact of Support Staff in Schools (Results from Strand 2, Wave 2). 2009

Dyson, A., Farrell, P., Polat, F., Hutcheson, G., & Gallannaugh, F. Inclusion and pupil achievement (Research Report RR578), 2004

Lunt, I., & Norwich, B. Can effective schools be inclusive schools?, 1999.


  1. OldAndrew,
    There is an issue around the concept of inclusion at all levels. The question to be asked is what level of inclusion is acceptable?
    Are we allowing any and all students of a common age into the same educational grouping?
    Do we preclude any student who is not exactly of the same educational attainment from the group?
    What position along the spectrum of inclusion is valid?
    Is there a universal answer or is it dependent on the environment/teacher/support network?

    I think every teacher comes to their own answer about these questions (perhaps all are not equally valid, or are driven by a specific ideology). My own opinion is that students who are in an instructional group when they have not achieved functional mastery (or at least working familiarity) of pre-requisite skills is unfair to the individual student and other members of the group. If an individual is unable to interact with the material, they will require additional support, support which might have been provided to students at the level to develop their mastery.

    As for including students who are unable to attain the level yet included due to a “need to be with their social/ age/ peer group” the less said the better.

    Finally, if you ADHD (or any of its variants) to be invalid, you’ll love ODD (Oppositional Defiance Disorder).

  2. What about the money?

    I realise these reports are about specific research. I also expect that the terms of reference were narrow enough that they were not expected to investigate how this money could be spent for maximum effectiveness.

    Surely, doesn’t everybody know by now that the same amount of money spent on early intervention would actually produce real results? By early, I mean early. Toddler, pre-school, and junior primary years. No matter whether it’s for real, obvious disability like autism or Down’s or for social and behavioural issues including lack of family support, the earlier the better.

    The 30 million word gap research was written eons ago. (I’m not fussed that some followup indicates the gap may be less. If the average deficit is actually 10 million words at 3 years old it’s still massive.)

    Intervening at 2-3 years old could not possibly cost more. And the benefits of narrowing such gaps would flow through to all schooling. All these effects, both negative and positive are cumulative.

  3. Your final question is obviously rhetorical.
    A father and teacher in the US who almost had his children seized by the state, wrote the following: The state receives about $100,000 from the federal government for each child they seize – which then goes to pay the people who seize them and their various associates who “care” for the children.
    Can anyone be surprised when the state becomes ever more eager to seize people’s children?
    How much “free” (i.e. taxpayers’) money is going for each SEN child?
    Neuroskeptic writes: In the case of mental illness, those who research mental illness know that their funding depends on the idea that it’s a widespread problem. The more common people think it is, the more important studying it seems. Meanwhile, charities representing the interests of the mentally ill like high statistics because they make mental illness seem more “normal”, thus destigmatizing it. It can’t hurt their donation rates either.

    • Oh for pity’s sake.

      Let’s skip the libertarian rhetoric. The state does not have a single unified motive. You cannot analyse the policies and practices of politicians and bureaucrats by suggesting all the state ever wants to do is spend money. Some of the governments involved in promoting the move towards inclusion were rabid rightwingers who were cutting education funding at the same time (the basic system was set up in the early 1980s). Inclusion has often been promoted as a way to save money on special schools.

      The trouble with the SEN racket is that governments of several different ideological hues have found reasons to support it.

      If you can get hold of it this journal article is worth a look:


      It looks at the issue of medicalising behaviour from a variety of ideological perspectives. It isn’t a left/right issue (nor a libertarian/statist issue).

  4. Even with the findings available, there are potential problems with the interpretation that greater inclusiveness leads to poorer performance. Multiple Regression is powerful because it allows the researcher to look at co-variance between measures in isolation by accounting for other effects. So in this study, a small percentage of the variance in Value Added is statistically accounted for by variance in the number of children on the SEN register of a school when other demographic measures are controlled in the equation. The problem with this is that the researchers have defined inclusiveness by number of children on the SEN register which, I would argue are related but different. One is a concrete measure and the other is a philosphical concept. It’s a bit like measuring justice by the number of people you have in prison.

    I think there is substantial merit in your idea that educational standards are affected by a general culture of low expectations. And intuitively, if a school declares that many students are disabled in some way they low expectations of learning and behaviour may well follow. However, in my experience, some schools will have a greater ratio of children on the the SEN register (yes, even with statements) because they struggle to cope with the resources and systems they have in place as part of ordinary provision. So it is possible that well run schools, may have less children at SA or SA+ in comparison to those which are less effective at managing with a similar demographic intake because they are coping without the additional funding. So some of “inclusivity” could be accounted for by a schools inability to cope with the students they’ve got. Off hand, however, I am not aware of research which would support this idea – but it makes sense intuitively.

    • I realise they are measuring the number of children on the SEN register rather than the very concept of “inclusivity”. But it is the former, rather than the latter, which I am interested in.

      I’m not sure how you can tell that the effects are a small percentage of the variance of the VA. Do they actually give any figures for the variance of the VA?

      • Firstly I should point out that, although I know a fair bit about stats, I am not a statistician and secondly I have not analysed this in enough depth to draw firm conclusions. I took the researchers on their word in this matter.

        “First, the effect is small in relation to the range of points available at each key stage. For instance, the 0.03 points ‘effect’ of inclusivity at KS1 has to be set in the context of the 25 points available and the national mean APS in 2002 of 15.5. Likewise, the 0.53
        ‘effect’ at KS4 has to be set against the 64 points available (when only the pupil’s best 8 results are taken into account) and the national mean APS of 35.63…

        Second, the effect is also small in relation to the size of schools’ SEN populations. For instance, at key stage 1, every additional 10% of pupils with high levels of SEN in a school population is associated with average attainment scores which are 0.3 points lower. In fact 10% is a large difference in most primary schools. In schools with 300 pupils, for instance, it represents a difference in the population with SEN at School Action Plus or above of 30 pupils. If the difference were a more modest 2% – equating to 6 pupils at School Action Plus or above – the associated difference in attainment would be
        0.06 points”

        (Dyson et. al. 2004)

        After a scan of the Beta values in the appendix I would concur that, for those values provided, there is a small but significant effect. I.e. “inclusivity” accounts for a small amount of the variance on the stated measures.

        • Sorry, are you not talking about the VA any more?

          On the more general point of the harm done, I would suggest that the amount of harm to results any policy does is going to be limited. However, schools and governments following policies that harm results create a climate in schools, particularly when it is not explicitly admitted and the opposite is widely claimed.

          • Irrespective of whether I’m talking about the VA or the APS, ultimately, the effect of the number of kids at SEN or with statements on overall performance, statistically speaking, is small. You are welcome to interpret that any way you choose. Unfortunately, the statistical quagmire does not lend itself to easy answers. I stand by my personal experience which tells me that ineffective schools are more likely to struggle with difficult children.

            • The judgement that it is small seemed to be based on the idea that the only concern here is a few points on the inclusivity measure either way. However, the number of students on the SEN register is out of all proportion to what was intended when the policy started. The policy of inclusion is probably putting about 20% onto the measure of “inclusivity” in some schools. If you take that into account, plus the fact that for individual students the effect is cumulative over 4 key stages, and the general effect following policies that harm results has on the climate in schools, I don’t think it can be ignored. That said it is only one policy among many that harm students outcomes.

            • My judgement as to the size of the relationship between the stated measures is based on the statistics which, in the case of attainment,are derived from on a “model” of how inclusivity relates to attainment extrapolated from the data. Statistically speaking, that relationship represents a small but significant figure. However, your professional experience says that the impact of inclusion of children with SEBD in mainstream is substantial and detrimental. However, given that you argue that schools you have worked in have ineffective behaviour management policies, then inclusion of children with behaviour problems is likely to have a greater impact than in other schools which follow through, support teachers and children or maintain high expectations of behaviour.

  5. Can I preempt the pithy comment about referring to Value Added and then discussing Attainment by saying whoops. The regression analysis was performed on attainment a multi-level analysis was performed on Value Added. I am not familiar with the latter technique but I understand that it is used for predictive modelling of outcomes based on statistical data.

    “Similar findings are obtained with the multi-level analysis using value added scores. For every one percent of pupils in a school with SEN at School Action Plus or with a statement, we can typically expect average pupil value added scores in the school to be lower by the following amounts:

    KS1- 2 0.02 points
    KS2- 3 0.11 points
    KS3- 4 0.03 point” (Dyson et. al. 2004)

    However it is clear that that the level of variance based on “inclusivity” was small for Value Added also.

    • The effect on VA is compared with the number of points available at the key stage and with the mean APS, not with the mean and variance of the VA which would have made far more sense.

  6. I had another read through this.

    “We can see that some of the students who have received the most support through the deployment of teaching assistants have become helpless to the point where they won’t pick up a pen without asking for help first.”

    Aren’t the staff trained in how to help one to one? The most important aspect of what I do is teach the student to do as much as they can. The skill is in choosing the task to be within their capacities, and that should have been done by the teacher or some other qualified person.

    If they’re unwilling to try then you use a whole range of approaches to get them going. None of these allows the student to avoid doing a task themselves or to avoid by talking about it instead of doing it.

  7. But in a whole-classroom situation, you can’t have the TA wheedling some cooperation and effort out of a pupil without talking over the teacher and the pupil’s attention being divided.

    As of yesterday’s TES discussion: When I started in SEN in 1993, the Special Schools in our LA hadn’t been closed long. There was adequate funding to ensure a level of support, even if shared, for children who would previously have been allocated a Special School to make it in mainstream. If they still couldn’t, they were transferred back. SEN was the largest department in the inner-city school, and it waa staffed for many years with people who used to work in the Special Schools: a positive choice, experienced. Full statements were available to any child who simply couldn’t hack it withour f/t support.

    I’m not aware of a single problem with a child whose most substantial problem was a learning difficulty. In the intervening years however, the new code of practice ensured that short of being on a life-support machine, you weren’t getting a statement. There was a hockey-stick rise in the number of badly-behaved children being awarded descriptive-disorder acronyms when it is obvious to all concerned that it was merely a symptom of the breakdown in parental and school authority and discipline. They all however have to receive their chunk of the funding and that’s why LAs can no longer afford to fund at a rate that makes any damn difference.

    I suspect that reports deal mainly with the LD kids and is choosing to sideline those with EBD, who can be neatly passed to the BSW, mentoring and pastoral teams at no extra SEN expense. They are coincidentally the kids who cause the most disruption to learning, and I can’t understand why SMT aren’t itching to rid themselves of them. The single main cause, after arriving in the middle of Yr10 speaking no English, for poor results at our school anyway, is teaching time wasted dealing with ass*holes. The top set results are very good because kids with EBD (whether real or imagined) tend to achieve poorly and aren’t represented in top sets. The behaviour problem kids are taken out of lessons constantly for mentoring, trips, alternative provision, counselling, exclusions etc and miss a great deal of content.

  8. Sorry Lil, my imagination ran away with me. I thought that the LD students actually got some real, individual out of classroom help.

    So, now I really understand what you and OA say. That the *only* ones who get a bit of time away from the classroom because of their educational needs – are the EBD students.

    Excellent! I might write a play.

  9. I’ve found this very interesting. There are two, almost contradictory, arguments in this thread. One is that the inclusion of children with special needs and, more specifically, Behavioural, Emotional and Social Difficulties (BESD) is detrimental to mainstream education. The second is that such children are erroneously labelled in way that mitigates responsibility for their actions. Additionally, I cannot believe that all of the vast number of children who are on the SEN registers in struggling schools would have gone to special schools had they been at school pre-inclusion or even been permanently excluded. So what changed?

    • I don’t think there are two points here. There is just one: students who cannot benefit from, and will actually harm, the educational process are allowed to be in schools and allowed to cause harm.

      The two arguments are simply the two alternatives to this state of affairs for any such child: removing them from the mainstream or allowing them to remain but make them behave.

      Yes, these are two distinct options and for any given child it might be a different one of these two options that is best, but either would be a better approach than what is happening now.

  10. I agree. Disastrous SEN policies have soaked up so much time from primary class teachers for no discernible gain.

    Currently, I have several SEN children in the class and they are actually making no progress on the intervention programmes that the SENCO has decreed. However, in class, where I have control, I can motivate them to work: to write to an extent where I have something to work with, to take pride in contributing to whole class work, to hear ideas that are beyond the abysmally patronisingly low level of SEN reading materials, to work with different abilities of children in English, maths and science etc.

    I have seen this time and again, intervention rarely works, when it does then that’s great but does the occasional payoff justify the huge expenditure of resources. A child on a statement brings an additional £2500 to our school and we would be lucky to make a sub-level progress in maths and english for most in a year. Added to which, these children suck up so much teacher time when one compares them to the rest of the class and I believe this is the great scandal, the great lie, in waiting. Why should others in the class suffer from a lack of time with the teacher because of these SEN children?

    Before anyone suggests it we are not desperately bad teachers in our school.

    I feel that too many children are dumped into mainstream education which they are simply not suited for, especially those with behaviour issues. However, in our local authority there is nowhere else for them to go.

    With regard to behaviour, we see in the foundation stage which children are going to be a problem. Poor behaviour is not always a product of some appaling domestic situation, but it is often a product of pretty weak parenting. These children are frequently not ready for school – they need much longer in a playgroup environment where they can learn how to socialise with others first. They often only have one or two terms in the foundation stage before they move on to more academic pastures in KS1 which is where the behaviour gets progressively worse as their academic performence invariably tails off because of their inability to cope with a classroom environment.

    The older they get the more they need to be away from other children, and the classroom. These people are not a point in their life where school is right for them. They need somewhere else to learn where their experience with others and authority is not going to be soured by their behaviour.

    I often find that with these SEN children academic and social performence improves in spite of work done on these programmes. I can think of one child whose behaviour stopped being the barrier to learning because his mum started getting help for her mental health issues. That was when the class TA actually could motivate him to work, in class, to do what the other children were doing. That was when school was right for him, not the previous four years (that we’d had him). Her work was funded by his statement but she was originally best deployed in the rest of the classroom to mitigate the effects of his behaviour on others. The bonkers thing is that if he did not have a statement I would not have had a TA. We have others in our school with similar behaviour issues that don’t have that statement and we’re pretty good at behaviour management, and motivating children to work. Despite that those children are still not learning.

    Our lords and masters do not want to know about how effective these policies have been. Everyone who seems to be directing education seems to have their heads up, sorry, in the sand when actually listening to experiences from classroom teachers. One positive thing that they could do would be to actually ring-fence funding for TAs in primary classrooms – one per class. My God, since they’re not going to make proper provision for these children in alternative settings than school, then they could actually trust the school, and the teacher, to work with these children as best as they can.

  11. Interesting to see how the TDA have interpreted the research on Teaching Assistants:


    • Yes, but I bet it is not the SEN children that are making the progress.

      They will be removed from the classroom, leaving the teacher with a smaller group to work with.

      The question still needs to be asked: Should they even be in a school environment?

      • Yes, they should be in the school environment. If the SEN children are not coping in the mainstream schools, they should be transferred to a special school. Every child has the right to an education. Teachers should assess which school they would cope better in and discuss with the child’s parents about the decision. There is an issue of equality and equity here. Children with SEN must have the opportunities which are appropriate for them just as mainstream schools offer opportunities which are appropriate for non-SEN children.

  12. I have read through this article with great interest.
    As a parent of a child who has been placed on the SEN register due to falling behind in her classes in a mainstream school, I have found myself to become increasingly frustrated by the lack of progress she appears to have made.
    By placing her this way it was meant to have the effect of boosting her attainment levels and thus enabling her to catch up with her peers.
    Im afraid that I find myself agreeing with suggestion that SEN lowers expectations and disempowers the child, thus creating a situation whereby the child becomes increasingly dependent on the help provided. This is not what we wanted for our child, we currently send our child once a week to a private tutor and I honestly feel that she learns,achieves and takes more pride and ownership in her work in that one hour, than she does in a whole week at school with the extra SEN support thrown in.
    This is worrying to me, we have been given no other option by the school to help our child other than this SEN route and now we feel that it has done more harm than good.
    We have seen no improvement in her work and even more worringly her self esteem is so poor as she now feels that she is in her words stupid or cant do it.
    What is the education system doing to my child?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: