Bottom sets and the scourge of low-level disruption

//Bottom sets and the scourge of low-level disruption

In many English schools, low-level disruption is the norm. Children talking when expected to be silent, fiddling with equipment and each other, calling out, and generally not being ‘on task’ are all routinely accepted as just something with which teachers have to contend. In 2014, Ofsted published this report on low-level disruption in schools. It it, “around two-fifths of the 723 teachers in the survey who believed that disruptive ‘talking and chatting’ was a key problem said it occurred in almost every lesson.”


The entire concept of ‘behaviour management’ is predicated on the idea that teachers must manage students’ inevitable disruptive behaviour in order to teach effectively; good teachers are ones who can manage a roomful of potential miscreants. Until I visited King Solomon Academy two years ago I didn’t realise it was possible to run a school where low-level disruption could be eliminated.

But that’s not the full picture. In most schools good behaviour is also the norm, but only for certain students and certain classes. Teachers tend to have much higher expectations for bright students and top sets than they do for lower ability students and bottom set classes. All children learn what we communicate; if we signal that we will accept a certain amount of disruption, then that’s what becomes acceptable.

Recently, I was asked to observe some lessons in a school I visited. In one lesson students’ behaviour was perfect: they listened attentively, followed instructions, worked hard and got a lot out of the lesson. In the other two, a minority of children set out their stalls to be disruptive and effectively ruined the lesson for all the students in the class. The first lesson was a 6th Form class and the teacher obviously had incredibly high expectations of what her students could and would do. The other two classes were ‘lower ability’ students and, although the teachers had worked hard to plan interesting and productive activities, they expected students to be disruptive. The thing about our expectations is, people often meet them. If we don’t expect much, we’ll be proved right. If we expect everyone to rise to our high expectations then sometimes we’ll be disappointed but then nobody ever rose to a low expectation.

This is a big part of the problem with ability setting. On their Toolkit website, the EEF say:

Overall, setting or streaming appears to benefit higher attaining pupils and be detrimental to the learning of mid-range and lower attaining learners. On average, setting or streaming does not appear to be an effective strategy for raising the attainment of disadvantaged pupils, who are more likely to be assigned to lower groups.

… It appears likely that routine setting or streaming arrangements undermine low attainers’ confidence and discourage the belief that attainment can be improved through effort. Research also suggests that ability grouping can have a longer term negative effect on the attitudes and engagement of low attaining pupils.


Of course, it doesn’t have to be this way: it’s perfect possible for mixed ability teaching to be disastrous and ability setting can be made to work effectively.

It’s well-known that disadvantaged students are over represented in bottom sets. Teachers have an unconscious bias in favour of well-spoken, neatly dressed students who do what they’re told. The halo effect means that we see these traits as evidence of intelligence and their opposites as indicative of lower ability. Routinely then, children from disadvantaged backgrounds are packed into lower ability sets and everyone’s expectation of their behaviour and chances of academic success are low. If we don’t believe children can be successful they’ll usually prove us right. And if they feel unable to access an academic curriculum then there’s little incentive to get the most out of lessons.

But what if our standards for bottom sets were the same as that for top sets? What if our expectations for all students in all classes were that there was no disruption and that students would be polite, conscientious and respectful? And what if, in addition, we believed all children* could be academically successful and taught them with this in mind?

Doubtless there would be disappointment at times, but our expectations take on a critical mass. If students see a minority getting away with being disruptive it’s easy to join in. In fact it becomes increasingly difficult to commit to doing the right thing as behaving well when a majority are not can carry a social stigma. But when a majority conform to high standards of behaviour, it becomes increasingly difficult to stand apart. These social pressures mean that forgetting a pen becomes unacceptable, never mind mucking about or being disrespectful.

All this is easier said than done. Trying to manage persistent low-level disruption is draining. Teachers can only do so much in isolation and depend on schools to implement systems to encourage positive social norms. Schools need to support teachers by making it as easy as possible to use such systems so that the consequences for disruption are clear, consistent and certain.

Here are some suggestions for eliminating the scourge of low-level disruption:

  • Expect the same standards of behaviour from all students in every class. There may be reasons why children might choose to misbehave, but there is never an excuse.
  • Don’t be afraid to practise routines until they are automated. There’s little point in trying to cover content if some students are being disruptive.
  • Always use the schools’ behaviour system. Teachers who don’t follow school procedures undermine all their colleagues. School leaders who view using the system as a sign that teachers are struggling and in need of ‘support’ are dangerous idiots.
  • Poor behaviour can often be tied to poor self image. If students don’t believe they can be successful in your lessons then they might start to disrupt. By scaffolding the highest standards of work and then slowly, carefully removing this scaffolding we can help children establish clear mental representations of what high quality work looks like.
  • Never forget: what you permit you promote.

*When I say all children I’m not talking about those with severe special educational needs. These children need to be taught be specialists who understand and can provide for their complex needs.

2016-11-14T21:10:57+00:00November 14th, 2016|behaviour|


  1. Julie November 14, 2016 at 3:45 pm - Reply

    Hi David,

    A great article. You outline a key issue around expectations. I have just one point to add to it.

    In my experience low level behaviour has been diminished and often eradicated from the classroom altogether when I have shared the kind of data you refer to (see the ‘types of disruption’ chart above) with the children.

    Children as young as 5 have considered the issue of low level disruption, identified examples of it in class, discussed it, done the maths (with help of course) i.e. How many hours/weeks of learning are missed in a year if you waste 30 seconds during transitions times 10 times per day, 15 times per day, 20 times per day?

    By working through these calculations children have been amazed at the difference they can make to their learning, over time, if they were to cut the disruption and make transitions efficiently.

    Children have shown themselves to be very creative in finding ways to be vigilant of their own behaviour (to design their own scaffolding), to build their own high expectations of themselves and their peers and to coach each other to meet expectations, removing the scaffolding when they know they are ready. At times part of the scaffolding they have designed has been ‘me monitoring and feeding back on their behaviour’.

    I have found that it helps if behaviour is viewed in the same way as any other skill – not ‘good’ or ‘bad’ but on a continuum of development (i.e. we are all engaged in continually optimising our behaviour in reading, writing, communicating our needs, catching, throwing, learning, getting along with others etc.).

    An added benefit, for children being involved in designing their own solutions like this, is they build their capacity for self-efficacy, which can have positive knock on effects in other contexts of course.

    Glad to have your thoughts on this.


    • David Didau November 14, 2016 at 4:12 pm - Reply

      That sounds like a sensible idea. I’ll add it to my list 🙂

  2. Roz November 14, 2016 at 3:49 pm - Reply

    I currently have a bottom set and my expectations, I think, are even higher because with disruption in their lessons they get easily confused over simple points. Trying to explain to 13 students, a number of which are very new to English what Dramatic Irony is means I need no disruption. I have some but nowhere what I know could be going on..With the same expectations my set 2 year 8’s a different story..hmm.

  3. Tom Burkard November 14, 2016 at 4:39 pm - Reply

    The great majority of pupils I’ve taught have been identified as SEN, and a large minority have had statements. Of the latter, a much smaller minority have had severe (or ‘complex’) special needs which in a less inclusive age would have been met in a special school. I made it a point to never, ever look at statements, IEPs or psychologists’ reports, but simply concentrated on teaching the skills and knowledge that pupils needed. The only ‘special training’ I ever had was a methods of instruction course in the Army. As a matter of fact that’s the only ‘teacher training’ I’ve ever had, even though I now have a DPhil in Education. And I’m happy to say that I succeeded where any number of specialists had failed.

    We’d be far better off if we went back to the system that prevailed before Warnock, when the provision of support for special needs required a medical diagnosis. All we have accomplished is to create a major growth industry which creates pointless bureaucracy, drastically lowers teachers’ expectations and encourages the very opposite of a ‘growth mindset’. This is not to say that pupils don’t vary enormously, or that autism doesn’t exist. I’ve visited excellent special schools that cater for the small minority who have problems that are far too labour-intensive to be catered for in a mainstream school. But the vast majority of supposedly SEND pupils can thrive so long as they receive enough over-learning to thoroughly master decoding and encoding skills. And given good teaching materials, almost anyone can deliver this over-learning.

    • higgeldypig November 14, 2016 at 9:02 pm - Reply

      Hi Tom, as an NQT I would love to know more about “over-learning”. It’s not a phrase I’ve come across before. Could you point me in the direction of some of the good teaching materials you mention?

      • David Didau November 14, 2016 at 9:06 pm - Reply

        Overlearning involves teaching beyond the point where knowledge is acquired until it it is automated. Here’s a paper to start you off

      • Tom Burkard November 16, 2016 at 3:13 pm - Reply

        higgeldypig–I make it a point not to advertise anything I’ve written or published on blogs, but there aren’t many teaching materials that have a significant amount of overlearning built in. It’s generally most important when teaching basic skills where arbitrary associations (such as grapheme-phoneme correspondences or number bonds) must be learnt, and there is really no alternative to a lot of repetition. However, the more ways a given association can be presented, the less one need rely upon doing exactly the same thing over and over again. That is the tricky part! This is one reason why I always try to teach decoding and spelling together; the pupil is learning the same association, but in reverse order. This is generally a feature of beginning phonics programmes, but it’s not so easy with remedial programmes where decoding skills are normally in advance of encoding, or spelling.

    • David Didau November 14, 2016 at 9:05 pm - Reply

      I agree about the desirability of some sort of medical diagnosis for SEND. I included that caveat because of inevitable nonsense I get along the lines of, “Well, what about students with Down Syndrome? You see, you can’t just say all, can you?”

  4. Andy Tedd November 14, 2016 at 5:08 pm - Reply

    I notice that the only kind of disruption in the top table which parents think is more of an issue than teachers is ‘disturbing other children’.

    Are parents and parental expectations part of the problem? My experience of this is limited to governance (so very limited compared to teachers) but it seems often ‘more confident’ parents are in denial about the low level disruption caused by their little angels.

  5. George LILLEY November 17, 2016 at 10:59 pm - Reply

    I am going through Hattie’s evidence for his influence called ‘decreasing disruptive behaviour’. I was surprised to see he gives it a low effect size – 0.34. Hattie has often labelled focusing on influences less than 0.40 as ‘going backwards’ or ‘a disaster’. So his low effect size here seems to conflict with common teacher experience.

    Looking at his reasearch in detail, he used 3 meta-analysis – one giving a large negative effect size! Hattie’s interpretation – decreasing disruptive behaviour DECREASES achievement!

    Looking at that study I found the authors compared the achievement of students labelled with ’emotional/behavioral’ disturbance (EBD) with a ‘normative’ group. They used a range of measures to determine EBD, e.g., students who are currently in programs for severe behaviour problems e.g., psychiatric hospitals (p132).

    So the study Hattie uses is not looking at reducing disruptive behaviour at all, but comparing those students who are diagnosed with EBD with a normative group. So the large negative effect size means the EBD students are achieving much less than ‘normal’ students.

    If this study were removed from Hattie’s averaging – then reducing disruptive behaviour would shoot to the top 6 of Hattie’s rankings! This would then be consistent with teacher’s experience.

    For details see here –

  6. Chester Draws November 19, 2016 at 2:41 am - Reply

    My current school isolates out all the lowest ability students into a separate class. But

    1) give them stong teachers — ones with good control and able to teach well.

    2) we don’t confuse bad behaviour with low ability,

    Results are that they do very well.

    My current school broke both those rules. That’s when the bottom class is a problem.

    • Chester Draws November 19, 2016 at 2:49 am - Reply

      Sorry, My previous school broke both those rules. That’s when the bottom class is a problem.

  7. Francesca November 22, 2016 at 6:16 pm - Reply

    A really interesting blog post and some dynamic discussion happening in the comments also. I would be inclined to believe that a teacher’s expectations of a class does shape their behaviour. Positive feedback can be a really effective antidote to this. Rewarding everything even slightly positive that a disruptive child does is a simple way of shifting your focus to the positive instead of the negative. In addition, low ability classes can respond really well to alternative methods of teaching, for example using iPads to engage students in the topic. IT can be very powerful when it comes to engaging lower ability classes.

  8. […] 9 Bottom sets and the scourge of low-level disruption (November 2016) […]

  9. […] that they need special provision; none of these comment should be seen as applying to them. I wrote here about the scourge of low-level disruption in low ability classes; it’s scandalous that some […]

Constructive feedback is always appreciated

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: