What pupil voice evidence Ofsted inspectors look for

Inspectors form their view of pupil voice mainly from what pupils say and do on the day, so the question of what pupil voice evidence Ofsted inspectors look for is partly about conversations no school can script. The rest is what the school can show about how it listens: whether it hears from the whole cohort rather than a few confident pupils, whether it does so regularly, and whether anything changed as a result. A folder of positive quotes carries little weight. Representative, repeated, structured data tied to action carries a good deal more, because it matches what inspectors hear from pupils in corridors and classrooms. The sections below cover where pupil voice sits in inspection, which evidence is credible, why cohort breakdowns matter, and how to run a termly routine that stands up without becoming a performance.

Where pupil voice sits in the current inspection framework

Pupil voice is not a graded area in its own right. It is a source of evidence about experience, and inspectors use it to test what leaders say about areas such as inclusion, attendance and behaviour, and personal development. A school that says pupils feel they belong is making a claim, and what pupils tell inspectors is one of the main ways that claim is checked.

Inspectors gather pupils' views in several ways. They talk to pupils in lessons, at breaktimes and in planned conversations, and they observe how pupils behave and interact. They may also look at survey information a school chooses to share. Ofsted also runs its own questionnaires. Parent View is the best known, and inspectors can draw on Ofsted's pupil and staff questionnaires as well. How these are used, and how much notice is given of an inspection, has changed in recent years, so leaders should read the current wording rather than rely on memory or on what they heard at a previous inspection.

The renewed framework has been in use since November 2025, and its toolkit sets out the evaluation areas and what inspectors consider under each. As of November 2026, the Ofsted school inspection handbook and toolkit on GOV.UK are the documents to read for the exact language on pupils' views and which areas draw on them. It is worth noting the publication date on whichever version your team is working from, because guidance has been revised more than once and secondary summaries lag behind.

The Department for Education's white paper, Every Child Achieving and Thriving (February 2026), places inclusion and belonging at the centre of its expectations for schools. It gives a policy reason to treat pupil experience as a planning input year-round. The practical consequence is that evidence gathered only in the weeks before an inspection looks like exactly that. Evidence collected every term, as part of ordinary school improvement, tends to look like how the school runs.

What pupil voice evidence do Ofsted inspectors treat as credible?

It helps to separate three kinds of evidence, because schools often invest heavily in one and neglect the others. The first is what inspectors hear first-hand. Leaders cannot control this and should not try to, but they can know in advance what it is likely to say. The second is what the school can show it has collected systematically. The third is what it did as a result. Credible pupil voice work has all three, and each supports the others. If pupils tell inspectors that nobody asks what they think, a tidy survey archive will not help. If pupils describe real changes, a record of how those changes came about makes the picture convincing.

On the second kind, the strongest evidence comes from the whole cohort, repeated over time, using the same structured questions. A school council, a group of pupil ambassadors or a set of hand-picked pupils offers valuable perspectives, but it is a small and self-selecting sample. The pupils most likely to volunteer are often those who feel most comfortable in school. The pupils whose experience leaders most need to understand, such as quieter pupils, those who avoid unstructured time or those who feel on the margins, are far less likely to be in the room. Turning student voice into action starts with hearing from all of them.

Consider an illustrative case. A secondary school wants to show that Year 7 pupils feel they belong. One version of the evidence is a handful of warm comments from a transition event and a positive conversation with the head of year. Another is a short structured survey given to every Year 7 pupil each term, with statements such as "I have someone in school I can talk to" and "I know where to go at lunchtime", answered on an agree or disagree scale, and with results shown by cohort. The second version records the response rate, shows change from autumn to spring, and can say which groups scored lower. It makes a much stronger case, partly because it does not need to be flattering. A dip between terms, with a clear explanation of what the school did next, is more persuasive than a flat line of high scores.

Pupil voice here means the views and experiences of pupils, gathered in a way that lets leaders act on them. A structured survey uses the same closed questions for every respondent, so answers can be counted and compared. The response rate is the proportion of invited pupils who actually completed it, and it determines how far the results can be trusted. Schools can see what student voice surveys reveal when they are run this way.

Why groups matter more than whole-school averages

A high average settles less than it seems to. A whole-school or whole-year score blends the experience of every pupil, and the groups with the least positive experience are often a minority small enough to be hidden by the majority. Inspectors are interested in how well provision works for different groups of pupils, including those with SEND and those who are disadvantaged, so leaders are better placed when their own evidence already looks at those groups. The current toolkit is the place to check precisely how it frames this, and our guide to the Ofsted toolkit and inclusion evidence covers the detail.

Suppose, as an illustration, that 78% of Year 7 pupils agree with a statement about feeling part of the school. That reads as a healthy result. Now suppose the same statement shows markedly lower agreement among disadvantaged pupils, and lower again for pupils with SEND who are also disadvantaged. The overall figure has not changed, but the picture has. Leaders now know where to look first, and they know it before attendance dips or behaviour incidents rise, which is when many schools first notice a problem. The same logic applies to what pupils say about SEND provision.

Cohorts can be defined by year group, SEND, Pupil Premium or disadvantaged status, EAL, free school meals, children in care, and combinations of these. Combinations often tell the most, though they also produce the smallest groups. That is where caution is needed. A cohort of six pupils can swing wildly on the answers of one or two individuals, and reporting at that level can also make individual pupils identifiable, even in a survey described as anonymous.

Many schools and trusts deal with this by setting a minimum group size before any breakdown is reported. A threshold of around ten is a common starting point, though each trust should set its own and apply it consistently. Below the threshold, results are folded into a wider group or held back. Leaders can still look at small cohorts over several terms, treating the pattern as a prompt for conversation with the pupils' key adults rather than as a finding. The aim is to read the data for what it can support and no more, which also makes it easier to defend when a trustee or inspector asks how a figure was reached.

Running pupil voice so the evidence stands up

Consistency does most of the work. When the same questions are asked at the same point each term, results can be compared over time, and a drop or rise means something because the method has not changed. Where a trust asks every academy the same questions at the same time, the comparison extends across schools. Changing the wording, the timing or the year groups surveyed from one round to the next makes the results much harder to interpret, even when each individual survey was well designed.

Pupil surveys work best with structured, closed questions: agree or disagree statements, rating scales and multiple choice. Free-text boxes are not appropriate for pupils. They produce material that is hard to analyse fairly and can lead pupils to disclose things a survey is not the right channel for. Structured questions are quicker to answer, easier to compare and straightforward to report as scores and patterns, such as a finding that Year 7 pupils with SEND were less likely to agree that they have somewhere to go at lunch. Getting better data from belonging survey questions depends on this discipline. Open-text questions have a place in staff and parent surveys, where the context and the safeguards are different.

Length matters more than leaders sometimes expect. A survey that takes five to ten minutes tends to get a better response, and it is more likely to be completed carefully by younger pupils and by those who find reading harder. Schools often find that a few minutes of explanation from a form tutor, covering why the survey is happening, who will see the results and what will happen next, makes a visible difference to how seriously pupils take it. Pupils who have seen earlier results lead to something tend to answer more honestly.

Two decisions are worth making before the first survey goes out. One is whether responses are anonymous or named. Anonymous surveys encourage candour, while named surveys let a school follow up with individual pupils and build cohorts from MIS data. The other is who can see results, and at what level. Pupils are entitled to know the answers in advance, and the trust should control access centrally.

Response rate deserves to be recorded every time, alongside the results. A strong rate across the cohort means the findings reflect the experience of the whole year group. A small self-selecting group, perhaps the pupils who were in school on the day or who chose to take part, gives a much weaker basis for any claim, and leaders and inspectors will treat it that way.

Showing that listening led to action

Evidence of collection shows that a school asked. Evidence of impact shows that it listened, and the second carries more weight in a conversation with an inspector or a trustee. The most convincing pattern is simple. Each low-scoring cohort has a named owner, a dated action and a follow-up measure. Without an owner, findings are noted and left. Without a date, they drift. Without a follow-up measure, nobody knows whether the action worked.

For example, imagine that a Year 8 cohort scores noticeably lower than other year groups on statements about having an adult they can go to. The head of year and the SENCO agree an owner, a pastoral lead, and set up a small staff-led group that meets weekly for half a term, with a mentor assigned to each pupil in it. The action is logged with a start date. In the following term the same questions are asked of the same year group, and leaders look at whether agreement moved. If it rose, the school has a record of what worked. If it did not, that is also useful, and it points to a different response rather than a repeat of the first.

How leaders talk about results shapes whether people trust the process. Sharing scores and actions with staff in plain terms, and telling pupils what has changed because of the survey, treats measurement as a planning tool. If results are presented as a verdict on a particular teacher or year team, staff become wary and pupils sense it. Trustees tend to respond well when a headteacher presents low scores alongside the plan for them, because it shows the school understands its own position. Our piece on trustee questions on inclusion sets out what good evidence looks like at board level.

In a trust, this is where the same-method point becomes decisive. A board that receives headteacher summaries from each academy is comparing different judgements, written in different ways. A board that sees the same questions, answered by pupils at the same point in the term, can compare academies and cohorts directly, see where support is needed and ask better questions of its leaders. It can also see whether an intervention in one school is worth trying in another. This is the approach described in measuring belonging in a multi-academy trust.

How Satchel Pulse can help

Perceptions, from Satchel Pulse, is built around this kind of routine. It provides termly belonging surveys for pupils, staff and families, with structured, closed questions for pupils. Results are reported automatically, with breakdowns by year group, Pupil Premium and SEND, so leaders can see which cohorts score lower without building the analysis by hand. Comparison over time shows whether a term's actions moved the scores. For trusts, results from every academy can be viewed together, so boards work from the same questions and timing in each school.

Inspectors respond best to evidence that is representative, repeated and acted upon. A practical way to begin this term is to choose one year group and run a baseline survey, then name an owner for each of the lowest-scoring cohorts. That gives a school its first comparison point for the spring.

Learn more about our services