How schools can measure staff wellbeing: a practical guide for UK school leaders

Measuring staff wellbeing is one of those things most schools intend to do, many start, and fewer finish properly. This guide sets out what to measure, which validated instruments to use, how often to ask, how to keep it genuinely anonymous, and what to do with the answers once you have them.

Why should schools measure staff wellbeing?

Three reasons: it is the earliest reliable signal of a retention problem, it is a commitment schools have made under the Education Staff Wellbeing Charter, and Ofsted inspectors now consider evidence of it under Leadership and Governance.

The retention reason is the one that matters most. A resignation is not the moment a teacher decides to leave, it is the moment they finish the paperwork. Measuring wellbeing regularly is the only practical way a leadership team finds out about a problem while it is still a problem rather than a vacancy.

The compliance reasons are real but secondary. The Education Staff Wellbeing Charter, published by the DfE in 2021, includes a commitment for schools to hold themselves accountable, including by measuring staff wellbeing. Ofsted's renewed framework places staff workload and wellbeing explicitly within Leadership and Governance, and inspectors look for evidence including staff survey results and wellbeing action plans.

A note on that second point. A survey run the week before an inspection proves what is happening that week. It does not prove anything is typical. What carries weight is a history of measurement with visible action attached to it, which is an argument for starting now rather than starting later.

What should schools actually measure?

Six things, because these are the areas UK research consistently links to whether teachers stay or leave: workload, mental wellbeing, leadership and voice, culture and recognition, pupil behaviour, and growth and autonomy.

Two additions are worth making. Ask a direct retention question, something like whether the person expects to still be working at the school in two years if circumstances stayed broadly the same. It is the single most predictive item you can include. And measure early career teachers separately, because a handful of struggling ECTs will never move a whole-school average.

What to avoid is a survey built entirely around satisfaction. "Are you happy here?" produces a number that goes up and down and tells you nothing you can act on. The six domains above are actionable because each one points at something a leadership team controls.

Should schools write their own wellbeing questions?

No, or at least not for the core measurement. Homemade wellbeing questions are effectively guesses dressed up as measurement. Validated instruments have been through peer review and psychometric testing, which means your results are comparable to national data and defensible if a trustee or inspector asks how you know the figures are accurate.

You can add your own questions around a validated core. That is often the right approach: use a tested instrument for the wellbeing measurement, then add school-specific items about the marking policy or the new timetable. What you should not do is invent the whole thing and treat the output as data.

Which validated instruments can schools use?

Four are realistic options for UK schools, and they differ substantially in cost and licensing.

The HSE Management Standards Indicator Tool. Free, publicly available, 35 items covering six workplace stressor areas: demands, control, support, relationships, role and change. Developed by the Health and Safety Executive, independently evaluated, and it comes with cross-sector benchmarks. This is the strongest free option available to any school in the country and it is under-used. Its limitation is that it measures workplace stressors rather than mental wellbeing directly, and the language is generic rather than education-specific.

The Short Warwick-Edinburgh Mental Wellbeing Scale (SWEMWBS). A seven-item measure developed at Warwick and Edinburgh medical schools, used by the NHS and by Education Support in the Teacher Wellbeing Index. It gives you a wellbeing score directly comparable with national population data. It is free for schools to use with registration, but commercial use requires a paid licence from Warwick Innovations, which is why not every platform includes it.

The Copenhagen Burnout Inventory. Free to use, covers personal and work-related burnout specifically. Useful if burnout is the thing you are most worried about, less useful as a general wellbeing measure.

The DfE Working Lives items. Not a formal instrument, but the questions used in the national Working Lives of Teachers and Leaders survey are published, and using their wording lets you compare your school against the national picture directly. This is a genuinely underrated option.

Whichever you choose, use it verbatim. Rewording a validated instrument to sound more like your school invalidates the comparison you were using it for.

How often should schools measure staff wellbeing?

Termly, for most schools. Three times a year is frequent enough to produce a trend line and infrequent enough to avoid fatigue, provided every round is followed by a visible response.

Annual measurement is too slow to be useful. If a teacher's workload became unmanageable in October, a survey in June finds out eight months late, and resignations for September posts are typically in by the end of May. You are measuring after the decision window has closed.

Monthly is usually too often unless the survey is very short and you have the capacity to respond every time. The fastest way to kill a feedback culture is to ask more often than you act.

One distinction worth holding onto: how often you ask and how often you write a new action plan are two separate decisions. Some terms the right response is to check whether the plan you already have is working, and tell staff that is what you are doing.

How do you keep a staff wellbeing survey anonymous?

Properly, or not at all. That means no names attached to responses, a suppression floor so no result is shown for fewer than five respondents, and filter combinations that cannot be narrowed to a group small enough to identify someone.

The suppression floor is the part schools most often get wrong. If your dashboard lets you filter to "Year 6 teachers who have been here under two years", and there are three of them, you have just broken anonymity while technically collecting anonymous data. Apply the floor to every segment, not just to the school as a whole.

5 respondents

This is not a legal box-ticking exercise. It is the thing that determines whether your data is worth having. If staff suspect their answers can be traced, they give you the answers they think are safe, and safe answers tell you nothing. A survey people do not trust is worse than no survey, because it produces a false sense that you have checked.

Practical points: use single-use anonymous links rather than named invitations, show a privacy notice before submission, and be able to delete a whole round of responses if someone asks. Survey responses count as personal data under GDPR even when anonymised.

What response rate do you need?

Aim for 60% or above. Below 50%, treat the results as indicative rather than representative, and say so openly when you share them.

60%

Response rate is itself a signal. A rate that falls between rounds usually means staff have concluded that nothing happens as a result, which is more useful information than anything in the survey. Report the rate alongside the results every time.

What are the options for measuring staff wellbeing?

Roughly five, in ascending order of cost.

Do it yourself, free. Use the HSE Indicator Tool through Microsoft or Google Forms. Costs nothing. The real costs are your time analysing it, the absence of benchmarks beyond the HSE's own, and the risk of accidentally breaking anonymity through your own spreadsheet filtering. Genuinely viable for a small single school with someone confident in spreadsheets.

CORC Wellbeing Measurement for Schools. A staff survey using validated tools, delivered with an analysed report you can share with governors. £300 plus VAT for primaries and £400 plus VAT for secondary and all-through schools. Strong on research credibility, being run by the Child Outcomes Research Consortium and Anna Freud. It is a measurement service rather than an ongoing platform.

Welbee. Built on the HSE Indicator Tool, with benchmarking against other schools and cross-sector organisations, heatmaps by job role, recommended actions and an anonymous follow-up conversation feature. Subscriptions start from around £397 plus VAT per school. Well established, multiple sector awards, and the HSE foundation means the evidence base is solid.

Edurio. The largest platform in the space, used by more than 2,000 education organisations, with 18 survey modules covering staff, pupils and parents, national benchmarking, and detailed analytics. Strongest choice if what you want is breadth of measurement and the ability to compare yourself against a large national dataset. Priced at trust scale, so generally a bigger commitment than the options above.

Echo. What I build. A short research-grounded staff check-in each term, which produces an action plan rather than a report, plus a "you said, we did" update to close the loop with staff, and a trust-level view for MATs. Priced at £150 a year for primaries, £500 for secondaries, and £150 per trust for MAT analytics. The honest limitation: Echo is newer than the others, so the benchmarking dataset is smaller. If national benchmarking is your main requirement, Edurio has more data than I do.

How do you choose between them?

Ask what you are going to do with the results, then buy the thing that supports that.

If you need a defensible number for governors once a year, CORC does that cheaply and credibly. If you want workplace stressor analysis with strong benchmarks, Welbee. If you are a large trust wanting to measure staff, pupils and parents against national data, Edurio. If your problem is that you already have data and nothing happens as a result of it, that is the gap Echo was built for.

What should schools do with staff wellbeing results?

Four things, in this order, and the third one is where most schools fall down.

Pick one or two things, not ten. An action plan with nine priorities is a document, not a plan.

Make them specific and dated. "Reviewing the marking policy" is not an action. "From January, written marking is expected fortnightly rather than weekly, with the revised policy going to the January INSET" is.

Tell staff what you heard and what you are doing. Including the uncomfortable findings, and including what you are not doing and why. An honest no earns more trust than silence, because silence reads as indifference. This step costs a paragraph in the staff briefing and it does more for your next response rate than anything else on this list.

Report what you did and what the scores show, without claiming one caused the other. A term contains a hundred variables. If you claim a change caused a score to move and staff did not experience it that way, you damage the credibility you were trying to build. Put the action and the number next to each other and let people draw their own conclusions.

The five most common mistakes

Measuring annually, which is too late to act on anything.

Writing your own wellbeing questions instead of using a validated instrument.

Allowing filters that break anonymity for small groups.

Collecting data and never telling staff what happened.

Treating a pre-inspection survey as evidence, when what carries weight is a history of measurement.