Behaviour in Scottish schools has become, in the run-up to the May 2026 Holyrood election, the subject everyone has an opinion about and almost nobody is discussing well. The political debate ranges from calls to restore classroom discipline and zero tolerance approaches to violence, through promises of expanded pupil support, to pledges on teacher workload. Each of these positions contains something defensible. None of them engages with the most important question, which is not what should we do about bad behaviour? but what do we actually know about why it happens?
The pattern is not uniquely Scottish — England's behaviour debate cycles through the same positions, from zero-tolerance academies to trauma-informed practice, without settling on what the evidence says.
This is not a call for sympathy over consequences, or for consequences over sympathy. It is a call for evidence — the kind of evidence that comes from looking carefully at what is happening in classrooms, rather than from political instincts about what schools ought to be like.
What the data says
The fifth wave of the Behaviour in Scottish Schools Research (BISSR), published in April 2024, is the most rigorous evidence base Scotland has on this question, drawing on responses from 3,754 staff across 525 schools. Its findings are serious. There has been a general worsening of pupil behaviour since the research began tracking it in 2006, with particularly sharp deterioration since 2016. In primary schools, the proportion of teachers experiencing physical violence at least once in the previous teaching week rose from 3% in 2016 to 12% in 2023 — a fourfold increase. Among primary headteachers and support staff, the figures rose from around 13–15% to 28–30%. In secondary schools, violent incidents have increased to more than four times pre-pandemic levels.
These figures are alarming and they are real. The experience of teachers working in classrooms where disruption is constant and escalation is unpredictable deserves to be taken seriously, not dismissed as anecdote or minimised by reference to what is happening in other countries. But the BISSR data, for all its value, tells us that behaviour has deteriorated. It does not tell us why — and the political conversation has rushed to fill that explanatory gap with ideology rather than evidence.
The false binary
The behaviour debate in Scotland has settled into a familiar and unproductive polarity. On one side: behaviour is a discipline problem. Schools have become too permissive, restorative practices have replaced meaningful consequences, and what is needed is clearer rules, firmer sanctions, and the removal of persistently disruptive pupils so that the majority can learn. On the other side: behaviour is a symptom of unmet need. Poverty, trauma, the cost-of-living crisis, the post-pandemic mental health landscape, and the explosion in additional support needs all mean that young people are arriving in classrooms carrying burdens that previous generations did not, and what is needed is more support, more understanding, and more investment in the pastoral infrastructure of schools.
Many of the young people at the centre of the behaviour debate are the same young people counted in the 43% ASN figure — and the supports they're entitled to are often missing.
Both of these accounts contain truth, and neither is sufficient. The discipline account is correct that schools need clear expectations, that persistent disruption causes real harm to the learning of other pupils, and that the absence of meaningful consequences — or the perception of that absence — erodes both teacher authority and pupil safety. The support account is correct that behaviour does not arise in a vacuum, that disadvantage and adversity are strong predictors of behavioural difficulty, and that punishment without understanding tends to produce compliance at best and exclusion at worst, neither of which constitutes education.
The problem is that both accounts treat the classroom itself as a black box. The discipline position assumes that behaviour is a function of character and consequence — that pupils choose to behave or misbehave, and that the school's role is to calibrate the incentives correctly. The support position assumes that behaviour is a function of need — that pupils who are dysregulated are communicating distress, and that the school's role is to meet the underlying need. Neither position spends much time asking what is happening inside the classroom that might be contributing to the patterns everyone agrees are getting worse.
The behaviour conversation can't happen in isolation from the pastoral-academic divide that structures most secondary schools — they're the same problem, viewed from different angles.
The missing evidence
The EEF's guidance report on Improving Behaviour in Schools (2019), drawing on a systematic review of the evidence, identifies six recommendations for schools — and the most striking feature of the report is how many of those recommendations relate to the quality of the classroom environment rather than to the behaviour management system. Two of the four proactive recommendations — Teach learning behaviours alongside managing misbehaviour and Use classroom management strategies to support good classroom behaviour — point to a body of evidence showing that what happens in the classroom — the quality of instruction, the clarity of routines, the degree to which pupils are cognitively engaged, the warmth of the teacher-pupil relationship — is at least as significant a predictor of behaviour as the school-wide policy on sanctions and rewards.
This should not be surprising. A pupil who does not understand what they have been asked to do is more likely to disengage. A pupil who is not cognitively engaged — who is sitting through a lesson in which the work is too easy, too hard, or insufficiently structured to require their active participation — is more likely to find alternative sources of stimulation. A pupil who does not feel known by their teacher, or who perceives that the teacher has already written them off, is more likely to behave in ways that confirm the low expectations they sense are held of them. None of this excuses poor behaviour. But it does suggest that the conversation about behaviour cannot be separated from the conversation about teaching, and that schools which address only the behaviour policy without attending to the classroom experience are treating symptoms while leaving causes intact.
Robert Balfanz's work at Johns Hopkins on early warning systems reinforces this point from a different angle. The ABC framework — tracking Attendance, Behaviour, and Course performance — treats behaviour not as a standalone problem but as one signal within a broader pattern of disengagement. A pupil whose attendance is declining, whose behaviour referrals are increasing, and whose attainment is dropping is not three separate problems requiring three separate interventions. They are one pupil, experiencing something that is going wrong across multiple dimensions, and the response needs to be equally integrated.
What observation evidence could tell us
Here is where the conversation Scotland isn't having comes into focus. If we want to understand why behaviour has deteriorated, we need to look inside classrooms — not to blame teachers, but to understand the conditions in which behaviour becomes difficult, the moments at which disruption escalates, and the practices that appear to prevent it. This is precisely what classroom observation, done systematically and analysed in aggregate, can provide.
Individual observation of an individual lesson is insufficient for this purpose. But observation evidence collected across a school — tracking, for instance, the proportion of lessons where pupils are cognitively engaged during independent work, or the consistency of routine-setting across departments, or the relationship between the clarity of task instruction and the frequency of low-level disruption — produces a picture that no other data source can offer. Behaviour referral data tells you where incidents happened. Observation evidence tells you something about why.
Consider a pattern that is common in many schools: behaviour referrals cluster in particular lessons, particular times of day, and particular year groups. The behaviour management system sees this as a data point: these pupils are behaving badly in these lessons. The observation-informed approach asks a different question: what is different about the lessons where behaviour is worst? Is it the point in the day? The transition between activities? The level of cognitive demand? The degree to which routines are established? The relationship between the teacher and the class? These are not questions that excuse the pupils or blame the teachers. They are questions that produce actionable insight — the kind of insight that allows a school to intervene upstream of the incident rather than downstream of it.
The BISSR report itself notes that the factors cited as root causes of disruptive behaviour — poverty, deprivation, trauma, challenges associated with home life — underline the need for schools to invest in understanding and addressing the conditions within which behaviour occurs. But 'investing in understanding' requires mechanisms for understanding, and the most direct mechanism a school has for understanding what is happening in classrooms is observation. Not observation as surveillance. Not observation as judgement. But observation as evidence — the disciplined practice of looking at what is happening in a classroom and asking what it tells us about the learning environment we have created.
Beyond discipline vs. support
The most effective approaches to behaviour I have encountered in schools do not fit neatly into either camp of the current debate. They are schools that have clear expectations and strong pastoral support, that enforce consequences and seek to understand the causes, that hold firm boundaries and maintain warm relationships. In HGIOS4 terms, they are schools that have genuinely connected QI 2.3 (Learning, teaching and assessment) with QI 3.1 (Ensuring wellbeing, equality and inclusion) — schools that understand that the quality of the learning experience and the wellbeing of the learner are not separate domains but two aspects of the same thing.
What these schools have in common is not a particular behaviour policy or a particular ideological commitment. What they have in common is evidence. They know what is happening in their classrooms — through observation, through data, through professional dialogue — and they use that evidence to make decisions that are specific to their context rather than derived from a national debate that has long since stopped being about what works and become about what we believe.
Scotland's behaviour debate will not be resolved by the Holyrood election, whichever party wins. It will be resolved, school by school, by leaders who are willing to look honestly at the evidence from their own classrooms and act on what they find. That requires a commitment not to a position but to a process — and the process begins with looking.
Tools that help schools aggregate what they're seeing in classrooms — without turning visits into a verdict machine — are part of that process; Learning Lens is available for Scottish schools — book a conversation from the home page.
References
Education Endowment Foundation (2019). Improving Behaviour in Schools: Guidance Report. London: EEF.
Scottish Government (2024). Behaviour in Scottish Schools Research 2023 (BISSR). Edinburgh: Scottish Government.
Balfanz, R., Herzog, L. & Mac Iver, D.J. (2007). 'Preventing Student Disengagement and Keeping Students on the Graduation Track in Urban Middle-Grades Schools.' Educational Psychologist, 42(4), 223–235.
Education Scotland (2015). How Good Is Our School? (4th edition). Livingston: Education Scotland.
Jamie Scobie writes from extensive experience in Scottish secondary education, including pastoral care, data for improvement, and school self-evaluation. This blog is an independent publication: he writes in a personal capacity as the creator of Learning Lens, writing about classroom observation, teaching evidence, and education policy. He speaks here only for himself and for Learning Lens, not for any employer or other organisation. He holds an MSt from Cambridge (Distinction) and a Masters from Stirling.