Behaviour Management in the Classroom: What Helps

Behaviour is not the same problem as classroom organisation

There’s a lot of good advice out there about seating plans, routines, and how to structure a lesson so pupils know what’s coming next. That’s classroom management, and it matters. This article is about something narrower and, for a lot of teachers, more draining: what actually happens when a pupil is shouting, refusing, walking out, or grinding a lesson to a halt, and what you do in the sixty seconds after that starts.

The research base on this has grown a lot in the past decade. The Department for Education published fresh guidance on behaviour in February 2024, the Education Endowment Foundation (EEF) ran a full evidence review specifically on behaviour, and Tom Bennett’s independent review for the DfE, Creating a Culture, spent a year visiting schools that had turned their behaviour around. None of it pretends there’s a script that works on every child every time. But it does converge on a few things that are worth knowing, especially when you’re dealing with a pupil whose behaviour keeps derailing your lesson and nobody’s quite sure why.

Behaviour is communication before it’s defiance

The EEF’s guidance report puts it plainly: pupil behaviour has multiple influences, and understanding a child’s context is what allows you to respond effectively rather than just react. It gives a simple example worth sitting with. Two children are disruptive in the same lesson. One is after attention, even negative attention. The other wants to escape because she’s bored or overwhelmed. If your default response is to send both children out, you’ve accidentally rewarded the second child’s behaviour while doing very little for the first. The behaviour looks identical from the front of the room. The function underneath it is completely different.

The DfE’s own guidance says the same thing from a different angle. Where appropriate, staff should take account of contributing factors identified after an incident: bereavement, abuse or neglect, mental health needs, bullying, undiagnosed SEND, criminal exploitation, or serious difficulty at home. That’s not an excuse-making exercise. It’s basic diagnostic thinking. A sanction aimed at a child who’s testing boundaries does a different job to one aimed at a child who’s dysregulated because something has gone badly wrong outside school.

This matters more for pupils with social, emotional and mental health (SEMH) needs. The DfE’s separate advice on mental health and behaviour is explicit that disruptive, withdrawn, anxious or depressed behaviour can all be signs of an unmet mental health need, and that schools shouldn’t wait for a formal diagnosis before putting support in place. The advice lists risk factors that stack up: bullying, poor relationships with staff, family breakdown, bereavement, socio-economic disadvantage. Where several risk factors are present at once, the research it cites found the likelihood of a child developing a conduct disorder rises sharply. Behaviour rarely comes out of nowhere. It’s usually the visible end of something that’s been building for a while.

When it might be undiagnosed SEN

The DfE’s behaviour guidance is careful on this point: some behaviours are more likely to be associated with particular types of SEND, such as a pupil with speech, language and communication needs who genuinely hasn’t understood an instruction rather than chosen to ignore it. But it doesn’t follow that every incident of misbehaviour is connected to a pupil’s SEND, and a school shouldn’t assume it must be just because a diagnosis exists. What it does say is that schools should try to anticipate likely triggers and build in prevention: short movement breaks for a pupil who struggles to sit still, seating a child with a hearing impairment in sight of the teacher, adjusting uniform for sensory issues, training staff in conditions like autism so they can recognise what they’re looking at.

Where there’s no diagnosis at all, the same logic still applies. A child who suddenly can’t cope with instructions, who masks all lesson and falls apart at breaktime, or whose behaviour has changed sharply from their normal pattern, is telling you something even if nobody has named it yet. That’s exactly the kind of pattern a SENCo referral exists for, which is covered further down.

Whole-school policy is not the same job as what you do at 9:47am on a Tuesday

It’s worth being clear-eyed about where the school’s behaviour policy ends and your own judgement begins, because the two get conflated constantly. The DfE’s guidance sets out national minimum expectations for the whole school: consistent standards, proportionate action when things go wrong, support for pupils who need it to meet expectations. That’s a leadership job. Tom Bennett’s review found the same pattern in every effective school he visited: a headteacher with a clear, detailed vision, communicated relentlessly, backed by a leadership team that turns up in corridors and at the school gate rather than staying in the office.

None of that is what you’re doing when a pupil throws a chair. In the moment, your job is narrower: keep everyone safe, de-escalate, and get back to a calm room as fast as you reasonably can. The DfE guidance is direct about this. When staff become aware of misbehaviour, the first priority is the safety of pupils and staff and restoring a calm environment, not working out the perfect proportionate consequence on the spot. De-escalation techniques and pre-agreed scripts exist precisely so you’re not inventing a response from scratch while adrenaline is running high.

The gap between policy and practice is also where things most often go wrong, and it’s not really about having the wrong policy. The EEF’s report flags that in an Ofsted review of 95 schools, inconsistency in how behaviour was handled across different classes came up in over a third of reports, more than any other single issue. Tom Bennett’s review found something similar from survey data: only around a quarter of secondary teachers and half of primary teachers agreed that their school’s behaviour policy was actually applied consistently by all staff. A well-written policy that three different teachers enforce three different ways is, from a pupil’s point of view, not really a policy at all.

De-escalation: what actually works in the moment

The single most useful shift most de-escalation approaches share is that the goal isn’t control, it’s lowering arousal so a conversation becomes possible again. The Low Arousal Approach, developed by Dr Andrew McDonnell, starts from the idea that an adult’s own behaviour is the one variable they can actually control in a heated moment. A distressed pupil in fight-or-flight doesn’t respond well to an adult closing the distance, raising their voice to match, or demanding instant compliance. All of that reads as threat, and threat escalates.

Some concrete things that tend to help:

• Give physical space. Stepping back and dropping the sense that the pupil is being cornered, literally and socially, removes the audience effect and gives their feelings somewhere safer to land than in front of thirty classmates.
• Watch your own tone and pace before theirs. A flat, unhurried voice signals safety in a way that a raised or clipped one doesn’t. If you need a second to steady yourself, take it visibly. A pupil watching you breathe and think rather than snap is learning something too.
• Use take-up time. Give an instruction, then step back and allow a genuine pause before expecting compliance. Pupils who are already overwhelmed often need a few seconds to process before they can act, and crowding that gap tends to provoke refusal rather than speed things up.
• Offer a narrow choice rather than an ultimatum. “You can sit here or by the door, your call” keeps a boundary in place while handing back a scrap of autonomy, which is often what a pupil is actually chasing.
• Save the full conversation for later. The DfE’s own guidance notes that schools may use pre-agreed scripts and phrases to help restore calm in the moment. The point isn’t to resolve everything there and then. It’s to get the room calm and safe, then deal with the substance once everyone, including you, is regulated.

Avoiding the power struggle

A lot of escalation isn’t really about the original instruction at all, it’s about what happens once a pupil digs in publicly and an adult feels they can’t back down without losing face. It takes two people to sustain a power struggle, and a teacher can choose not to be the second one. That doesn’t mean dropping the boundary. It means separating the boundary from the argument about it: state the expectation once, calmly, then disengage from the debate rather than repeating yourself louder. Offering choices within the boundary rather than issuing a flat command tends to defuse the sense that this has become a contest either side has to win.

This is also where the EEF’s evidence on relationships pays off before you’re ever in this situation. Teachers who’ve built genuine rapport with a pupil, even just a couple of minutes of non-academic conversation a week, have more credit in the account when things get tense, and pupils are less likely to feel they need to prove something in front of an adult who already knows them.

Working with a TA on a behaviour plan

If a pupil has an individual behaviour plan and a TA supporting them, the plan is only as good as the consistency between the adults running it. That means the TA needs to know the plan’s actual content, not just be told “keep an eye on him.” What are the early warning signs the plan expects them to watch for? What’s the agreed first response, and at what point does it hand over to you or to another named adult? What language are you both using for praise and for redirection, so the pupil isn’t getting two different sets of signals from the two adults in the room?

A TA supporting a pupil with behavioural needs is often the adult best placed to spot a pattern building well before it reaches you, precisely because they’re closer to that pupil more consistently. That’s genuinely useful information, but only if there’s a quick, reliable way for it to reach you and get folded into the plan rather than sitting in the TA’s head or a notebook nobody reads. Regular short check-ins, even five minutes at the end of a week, do more for consistency than an annual review meeting.

It’s also worth being honest with a TA about what they’re not expected to carry alone. A behaviour plan should never quietly become “the TA’s job” while the class teacher steps back from the relationship. The pupil still needs to experience you as someone who knows them and is invested in how their day goes, not just the person who signs off the plan someone else implements.

When to bring in the SENCo or pastoral team

The DfE’s guidance points to what it calls the graduated response, borrowed from the SEND Code of Practice: assess, plan, do, review. In practice that means if a pupil’s behaviour is persistent, not improving after low-level sanctions, or represents a sudden change from how they usually present, that’s the trigger to loop in the SENCo or pastoral lead rather than keep managing it solo. A sudden change in particular is worth flagging quickly. A previously settled pupil who starts refusing work or becoming aggressive is very often telling you something has changed at home, with their mental health, or with a need that hasn’t been identified yet.

The EEF’s report describes a more structured version of this called functional behavioural assessment: working with the SENCo or another colleague to observe what actually happens before a behaviour occurs, what the behaviour seems to achieve for the pupil, and what’s been tried already, before agreeing a plan with clear goals and a review point. It’s more work up front than reacting incident by incident, but it tends to be far more effective for a pupil whose behaviour hasn’t shifted despite consistent sanctions, because it treats the behaviour as a symptom worth investigating rather than a rule-break to be punished harder.

Where there’s any suggestion the behaviour might connect to a safeguarding concern, bereavement, abuse, neglect, exploitation, or a child protection issue, that’s a conversation with the designated safeguarding lead, not just the SENCo. The two roles often need to work together on the same pupil, and it’s worth knowing which door you’re knocking on for which concern.

Repairing the relationship afterwards

What happens after a difficult incident often matters as much as what happened during it, and it’s the part that gets skipped most often because everyone, understandably, wants to move on. The EEF’s report describes a straightforward framework for this called Establish–Maintain–Restore. The “restore” part is specifically about what happens after a negative interaction: taking ownership of your own part in how it went, offering a genuine acknowledgement of how the pupil was feeling, letting go of the incident rather than carrying it into the next lesson, and being explicit that you still want that pupil in your class. Small-scale evidence behind the approach suggests it can be done in a matter of minutes and doesn’t require a formal meeting, just a conscious decision not to let the relationship stay damaged by default.

The DfE’s guidance points in the same direction from a slightly different angle, describing a restorative conversation as part of reintegration after a pupil has been removed from a lesson or spent time in a support unit: facilitating the pupil’s own reflection on what happened and what they can do differently, rather than simply reinstating them and hoping it doesn’t happen again. A conversation that asks a pupil what happened from their side, who else was affected, and what could put things right does more for the relationship than a lecture, and it models the kind of accountability you’re actually trying to teach.

This is also where the school-wide culture and your individual practice meet again. A school can have a first-rate behaviour policy on paper, but if incidents are handled well in the moment and then never followed up, pupils learn that repair isn’t part of how the school works, only punishment is. Teachers who build the repair step into their normal practice, even briefly, tend to see fewer repeat incidents with the same pupil, because the relationship survives the bad day instead of accumulating damage from it.

None of this is a substitute for the basics

Nothing here replaces a clear, consistently applied whole-school policy, and nothing here works if a pupil doesn’t already know, in calm moments, what’s expected of them. But the moment-to-moment part of behaviour, reading what’s underneath it, staying regulated yourself, knowing when to hand something up the chain, and repairing things afterwards, is a real skill set of its own, distinct from classroom organisation, and it’s one that gets easier with deliberate practice rather than instinct alone.

If you’re tracking a pupil’s progress and resource allocation across subjects, it’s worth checking whether the system you’re using lets you flag SEN-aware needs at the point you assign work, so that behaviour-related adjustments don’t have to be re-explained to every colleague from scratch. Learnaroo Hub is built with that kind of flexibility in mind, though the day-to-day judgement calls covered above will always sit with the adult in the room.