Student voice in schools: how to turn it into action

Most schools already listen to pupils through councils, surveys and learning walks. Fewer can show what changed as a result. Student voice only counts as evidence of improvement when a school can trace a line from what pupils said, to a decision, to a later change in what pupils report.

This article sets out what student voice covers, which methods suit which questions, and why cohort breakdowns matter more than averages. It then covers how to tell pupils what happened, and how to build a termly cycle that SLT and trustees can follow.

What student voice covers, and what it does not

A working definition: student voice is the practice of gathering pupils' views on their learning and school experience, taking them seriously, and using them to inform decisions. The last clause is the one that separates it from a suggestion box. If views never reach a decision, the school has collected opinions, not practised student voice.

Roger Hart's ladder of participation (1992) is a useful way to see the range. At the lower end, pupils are informed or consulted: adults ask, then decide. Higher up, pupils take part in shaping decisions, and at the top they share decision-making with adults. Most school-wide work sits at consultation, and that is a legitimate place to be. The useful question is whether you have chosen that level deliberately for each topic. Asking pupils what makes the corridors feel unsafe is consultation. Inviting a pupil panel to help redesign the tutor-time programme is participation.

The topics are wider than many schools assume. Typical ones include:

• belonging and whether pupils have an adult they can go to
• teaching and learning, including feedback and challenge
• behaviour climate and how consistently rules are applied
• how safe pupils feel on corridors, at break and on transport
• transition from Year 6 to Year 7, and in-year arrival
• inclusion, and whether pupils from different groups feel equally welcome

The most common misconception is that student voice means the school council. A council gives depth and ownership to a small group of elected pupils. It does not tell you what the quiet pupil, the persistent absentee or the new arrival thinks, and those are often the views a leadership team most needs.

The rights basis is Article 12 of the UN Convention on the Rights of the Child (1989). It says children who are capable of forming their own views have the right to express them freely in all matters affecting them, and that those views should be given due weight according to age and maturity. That framing helps when a governor asks why the school does this. It is a right, and it also produces better information.

Choosing the right method for the question you are asking

No single method answers every question. Each does one thing well, and problems start when a school uses the wrong one for the decision in front of it.

School councils and pupil panels give depth and ownership. They suit questions where pupils can help design a solution, such as how a new break-time arrangement should work. Focus groups explain the reasons behind a pattern. Learning walks with short pupil conversations show what classroom practice feels like from the other side of the desk. Whole-cohort surveys give breadth and comparability, so you can see how many pupils feel a certain way and whether that differs by group.

The simplest rule is to match method to decision. A survey tells you where an issue sits. A focus group tells you what is behind it. If a Year 8 survey shows lower belonging in one tutor group cluster, the next step is a conversation with a sample of those pupils, not another survey.

Practical choices that change the quality of answers

Length matters. For secondary pupils, five to ten minutes is a sensible ceiling for a termly survey. For younger primary pupils, fewer questions with simple response scales, and adult support read-aloud where needed, produce more reliable answers than a long form completed by tired seven-year-olds.

Wording should differ by phase. Primary questions should be concrete ("Is there a grown-up in school you can talk to?"). Secondary pupils can handle statements on a scale, but avoid double-barrelled items such as "teachers are fair and helpful", where a pupil may agree with one half only.

Anonymous or named is a real trade-off. Anonymous surveys tend to encourage candour. Named surveys let you follow up with individuals and link results to attendance or other records. Anonymous responses can still be broken down by cohort, provided the platform tags year group or group membership without exposing identity.

Access needs to be built in. EAL pupils respond more fully with translated questions and audio read-aloud, and pupils with SEND may need a reader, extra time or a simplified version. If those pupils cannot complete the survey, your results will quietly leave them out.

Finally, ask only what you can act on. If the school cannot change the length of the lunch break, do not ask about it. Each unanswerable question teaches pupils that the exercise is decorative.

Why the pupils who are not speaking matter most

An average can look healthy while hiding a gap. Suppose a Year 9 cohort scores "fine" on belonging overall. That score can sit comfortably alongside much lower scores among pupils with SEND, pupils eligible for Pupil Premium, or those who arrived recently. Leaders who only see the headline never see the difference.

A cohort breakdown splits results by group: year group, SEND, disadvantage, EAL, free school meals, children in care, and combinations of these. The combinations matter. A pupil who is both disadvantaged and has SEND may have a different experience from either group alone. When cohorts come from your MIS rather than being typed in by hand, the breakdown is quicker and less error-prone, and you can repeat it each term on the same definitions.

Small groups need protection. Set a minimum group size below which results are suppressed, so no pupil can be identified from a cell of three or four. Many schools use a threshold of around ten, though the right number depends on your context and data protection advice.

The performance link

There is a reason leaders treat this as more than a pastoral matter. Belonging and attendance are linked in the research literature. A frequently cited review, Allen and colleagues' meta-analysis of school belonging and academic outcomes (Educational Psychology Review, 2018), found that school belonging is associated with academic outcomes and engagement. That is an association across many studies, not proof that raising belonging alone will lift attendance or attainment. Treat it as a reason to monitor belonging alongside your other data, not as a promise.

An illustrative example

Imagine a secondary school where Year 7 belonging looks healthy overall. Split by cohort, pupils who joined in-year and pupils with SEND score noticeably lower on "I have an adult I can go to". The whole-year average hid that. The school now has a specific, answerable problem: how do new arrivals and pupils with SEND get a trusted adult in the first half term? That is a far better starting point than a general plan to improve Year 7 belonging.

From results to action: closing the loop with pupils

Collecting views is the easy part. The sequence after results arrive decides whether pupils believe the exercise was worth doing. A workable order:

1. Share the headline results with staff first, so nobody learns about them from a pupil.
2. Choose one or two priorities. A list of eight becomes a list of none.
3. Name an owner for each priority, with a person's name rather than a department.
4. Set a review date, ideally the next survey window.
5. Tell pupils what is changing.

The last step is the one most often skipped. "You said, we did" messages work best when they are specific and visible: a slide in assembly, a few minutes in tutor time, a display near the canteen. Include what will not change and why. If pupils asked for later finishing times and the school cannot move the timetable, say so plainly, with the reason. Pupils generally accept a clear no more readily than silence.

An illustrative trust example

Suppose a trust of five academies runs the same Year 7 belonging questions in the same fortnight each term. One academy scores lowest on feeling part of the school. Its leaders do not guess at causes. They run pupil focus groups with a mix of Year 7 voices, and the pupils describe tutor time as rushed and say the transition buddy they were given in September stopped checking in by half term.

The academy changes two things. It restructures tutor time to include a short weekly check-in with a named tutor, and it reworks the buddy scheme so older pupils have a set schedule with their Year 7 partner for the first term. The head of Year 7 owns both actions. The academy then tells Year 7 what it heard and what it changed, and re-surveys the next term. Because every academy asked the same questions at the same time, the trust can see whether the gap narrowed.

The main pitfall is survey fatigue. Pupils who complete a survey three times and see nothing happen will stop answering honestly, or at all. A falling response rate in one cohort is often the first sign that the loop has broken.

Building a termly student voice cycle that leaders can evidence

A cycle gives student voice a rhythm that fits the school calendar. A termly or twice-yearly survey provides the quantitative spine. Between survey windows, qualitative methods such as focus groups, pupil panels and learning walk conversations explain what the numbers show. Results then go to SLT and trustee meetings in the same pack as attendance and behaviour data, so the three are read together.

Track a small number of measures over time:

• response rate by cohort, since a low rate in one group is a finding in itself
• change in your key belonging measures, term on term
• attendance or suspension patterns for the same cohorts

Be careful with the third. If belonging rises and attendance improves for the same group, that is encouraging, but correlation is not proof of cause. Attendance may also move because of a new family support arrangement or a change in staffing. Say what you can and cannot claim, and trustees will trust the rest of your reporting more.

At trust level, comparable data matters more than polished summaries. When every academy uses the same questions, method and timing, a board can see differences directly instead of reading each headteacher's account. The most useful governor question is simple: what did we do about it? That keeps the focus on action, and it gives headteachers a clear expectation to prepare for.

A platform can take on the manual work. Perceptions offers ready-made belonging questions, cohort breakdowns drawn from your MIS, automatic reporting, and AI analysis of free-text comments by sentiment and topic, so leaders are not hand-coding hundreds of responses.

On policy, as of 2026 the Every Child Achieving and Thriving white paper (February 2026) and Ofsted's inspection toolkit both place weight on inclusion and pupil experience. Check the current wording before quoting either in board papers, because guidance can be updated. A termly cycle with named owners and dated reviews gives you the evidence whatever the precise framing.

One cohort, one question, one owner this term

Start smaller than a full programme. Pick one cohort or year group, such as Year 7 or pupils who joined in-year. Choose a single question you can act on, for instance whether pupils have an adult they can go to, and gather views through a short survey or a focus group.

Before results arrive, name the person who will own the response and write the first draft of the "you said, we did" message. Deciding what you will tell pupils in advance forces you to commit to acting, and it makes the next survey far easier to run.

Learn more about our services