What the Ofsted toolkit means for inclusion evidence in 2026

Under Ofsted's renewed framework, inclusion has its own evaluation area, and inspectors judge it by what pupils, staff and parents experience as much as by what a school's documents say. As of October 2026, an Ofsted inspection toolkit for inclusion rewards leaders who know their own picture, weaker areas included, and can show what they have done about it. A bundle of policies does not do that work.

This article explains how inclusion sits within the framework, which types of evidence carry weight, and how pupil, staff and parent voice can be gathered so that it serves improvement first and inspection second. It also covers the mistakes that most often leave a trust with plenty of data and no evidence of change.

How inclusion is framed in the current toolkit

Ofsted introduced its renewed inspection framework for state-funded schools in England from November 2025. Inspection outcomes are now reported through report cards that set out evaluations across a number of areas, and inclusion is one of those areas in its own right. The exact names of the evaluation areas, the grading scale and the detail of the toolkit text have been refined since launch, so check the current wording on gov.uk and the Ofsted pages before you build anything around it. What follows reflects the position as of October 2026.

In the toolkit's sense, inclusion asks whether all pupils can access the curriculum, take part in school life and achieve well. The test is whether this is true in practice for each group, and the toolkit does not treat it as a statement of intent.

Inclusion is not a SEND-only judgement

Pupils with SEND are central to the picture, and so are disadvantaged pupils. But inclusion covers any group at risk of being left behind. In one school that might be pupils who joined mid-year. In another it could be children in care, pupils with EAL, young carers, or a particular year group that has struggled since transition. A school that treats inclusion as the SENCO's area will find the evidence thin when an inspector asks about a group the SENCO does not lead on.

A description of what inspectors consider, not a document list

The toolkit tells inspectors what to consider and how to reach a judgement. It does not give schools a list of documents to produce. Leaders who respond by building a folder for each evaluation area often end up with volume and little substance. The more useful question is what you could show an inspector, in a single conversation, that demonstrates how well each group experiences school and what you are doing where it falls short.

What inspectors look for as evidence of inclusion

Evidence of inclusion falls into three broad types. Outcomes data covers attainment, progress, attendance, suspensions and exclusions, broken down by group. Observed practice is what inspectors see in lessons, at breaktimes and in how adaptations are made. The third type is the experience reported by pupils, staff and parents, gathered through conversations and through whatever survey data the school holds.

Inspectors triangulate. That means they test one source against another. A policy that says all pupils can take part in enrichment is checked against participation figures, and then against what pupils say when asked about clubs and trips. If the three agree, the claim stands. If they disagree, the gap becomes the line of enquiry.

An example of how a claim gets tested

Suppose a school states that its Pupil Premium pupils are fully included in school life. An inspector might look at whether those pupils attend at a similar rate to their peers, whether they hold positions of responsibility and take up trips and clubs in proportion, and how they describe the school when asked. If attendance is slightly lower, participation in after-school clubs is thin and pupils say they would rather not stay behind, the claim has not held. If the school already knows this and can show what it has changed, the conversation goes very differently.

That last point matters most. Leaders are better placed when they have found the weaker areas before anyone else does, and can explain the action they have taken and what they expect to see change. An inspector can work with a school that says "this group scores lower on feeling accepted, here is the owner, here is the plan." A school that offers only positive summaries leaves the inspector to find the gaps alone, and the inspector's version tends to carry more weight.

Where belonging and experience data fit alongside attendance and behaviour

Attendance, suspensions and exclusions are lagging indicators. They record something that has already happened. By the time a cohort's persistent absence rises, the disengagement behind it has usually been building for months. Perception data works as a leading indicator, because it describes how pupils feel about school before those feelings show up in the register.

Belonging questions can reveal what the MIS cannot. Your attendance figures will not tell you whether a pupil has an adult in school they trust, whether they feel able to be themselves, or whether they think people like them are treated fairly. Those are the things pupils tend to say in an inspection conversation, and they are also the things that predict whether a pupil keeps turning up.

Disaggregation matters here. A whole-school belonging score can look healthy while one group sits well below it. The value comes from comparing groups, and that means defining cohorts clearly: year group, SEND, Pupil Premium, EAL, and the combinations of these that your pupil population makes relevant.

The DfE's white paper, Every Child Achieving and Thriving, published in February 2026, sets out the government's direction on inclusion and SEND reform. Read the published text directly for what it says about belonging and inclusion in mainstream schools, and check how its proposals are being implemented, as the detail is still developing. It is a useful reference point for trust strategy, but your evidence should rest on your own pupils' experience and not on citing policy.

Perception data supplements what you already hold. It does not replace attendance analysis, behaviour logs, SEND provision mapping or your existing improvement plan. Its job is to explain the patterns those sources show and to flag the ones they have not shown yet.

A worked example: Year 7 belonging across a trust

The figures below are illustrative, not real results. Imagine a six-academy trust that runs the same belonging survey for pupils in October each term. At trust level, Year 7 looks healthy. On the statement "I feel accepted at my school", 78 per cent of Year 7 pupils agree or strongly agree, in line with the previous year.

The trust then segments by cohort. Five academies show little variation between groups. In the sixth, Year 7 pupils with SEND score 54 per cent on feeling accepted, against 81 per cent for their peers in the same year. The trust-level average had hidden a 27-point gap sitting in a single school and a single group.

The response

The trust's inclusion lead raises the finding with the academy's headteacher the same week. The headteacher names the SENCO as owner and asks Year 7 form tutors for context. Free-text comments point to unstructured times: pupils say they are unsure where to go at lunch and do not know which adult to approach.

The academy introduces two changes. Each Year 7 pupil with SEND is paired with a trained Year 9 transition buddy, and form tutors begin a short weekly check-in routine with a fixed set of questions. The owner agrees a review date, and the trust schedules a re-survey for the spring term using the same questions.

What the board can say

Suppose the spring results show the gap narrowing to 11 points, with attendance for the group holding steady and not slipping. The trust board can now tell an inspector what it found, what it did and what changed, with dates and a named owner behind each step. It can also say what has not yet changed, and what the next action is. That account is specific, checkable and honest, which is what makes it credible.

Building an evidence trail that stands up to scrutiny

An evidence trail is a record of what you asked, what you found, what you did and what happened next. It is built through a handful of repeatable habits.

1. Use the same questions at the same point in the term in every school. Comparable data from pupils, staff and families is stronger than headteacher summaries, and it lets the board see where to look.
2. Keep the survey short. Five to ten minutes is enough for pupils to answer carefully. Offer anonymity where the topic calls for it, and use named responses only where you have a reason and have told pupils why.
3. Make it accessible. Pupils with EAL should be able to read the questions in their home language and hear them read aloud, otherwise their silence will look like agreement.
4. Share results back. Tell staff and pupils what you found and what you plan to do. Pupils who see nothing change stop answering honestly.
5. Report response rates next to every result, and flag low-response groups as a caveat. A 40 per cent response from one cohort tells you to treat its score with care. Hiding it undermines the rest of the data.
6. Keep a simple log. For each finding, record the action, the owner, the date and the follow-up result. One page per term is enough.

Tools can take much of the manual work out of this. Perceptions, for example, builds cohort breakdowns from your MIS, compares results with previous surveys and with other schools in the trust, and analyses free-text comments by topic and sentiment. That leaves leadership time for the part no tool does, which is deciding what to do.

Common mistakes when gathering inclusion evidence

Four patterns account for most of the weak evidence trails.

Running a one-off survey

A single survey gives you a snapshot. It cannot show change, and change is what inspectors and trustees want to see. Without a second data point, you cannot say whether your action worked or whether the first result was a blip.

Collecting results without breaking them down

An overall score of 80 per cent feels reassuring and tells you very little. Gaps between groups are almost always where the leadership decisions sit, and they disappear in an average. If you cannot see results by SEND, Pupil Premium, EAL and year group, you have not yet reached the useful part of the data.

Over-preparing respondents

Coaching pupils on what to say, or briefing staff on the right answers, produces data that looks good and means nothing. It also fails the triangulation test, because an inspector's conversation will not match the scores. Explain the purpose, say that honest answers help, and leave it there.

Producing evidence for inspectors rather than for leaders

If the survey exists to fill a folder, it will be designed, timed and read accordingly. Evidence that leaders use to plan, assign owners and review progress is also the evidence that stands up when someone asks about it. The inspection value follows from the improvement value, not the other way round.

One step to take before half term

This term, schedule a short belonging survey with cohort breakdowns to run before half term ends. Choose questions you can repeat unchanged, and make sure results can be split by year group, SEND, Pupil Premium and EAL at minimum.

When the results arrive, name an owner for every group that scores noticeably lower than its peers, and log the action and review date. Then book the re-survey for the spring term now, while the diary is still open, so that the second data point exists before anyone asks for it.

Learn more about our services