What your head sees
People answer honestly only if they trust that nobody can work out what they said. This page shows the rules working on a fictional school, then sets them out in plain words. The same rules apply to every survey, and the school cannot switch them off.
Been sent a survey by your school? Got a survey from your school? is written for you.
Fig. 1 A fictional school
aResults by role
Question
I feel valued by my school.
Percentage who agree or strongly agree.
41 answered. Shown: whole staff, teachers. Not shown: teaching assistants, senior leaders, office team.
| Group | Result |
|---|---|
| Whole staff41 answered | 68% |
| Teachers22 answered | 68% |
| Teaching assistants9 answered | Not shown. Fewer than 10 answered. |
| Senior leaders6 answered | Not shown. Fewer than 10 answered. |
| Office team4 answered | Not shown. Fewer than 10 answered. |
bAnswers chosen by few people
Any answer chosen by fewer than 5 people is hidden. Where hiding one would still let it be worked out from the others, the next smallest is pooled with it, so neither can be.
Question, whole staff, 41 answered
My workload is manageable.
| Answer | People | Share |
|---|---|---|
| Strongly agree | 7 | 17% |
| Agree | 18 | 44% |
| Neither agree nor disagree | 9 | 22% |
| Disagree and strongly disagreeToo few to show. Each was chosen by fewer than 5 people. | Too few to show | |
cNames in comments
Names come out of comments automatically before the school sees them: first names, names with a title, class codes, email addresses and phone numbers. Try your own: nothing you type leaves this page.
As your head sees it
The check uses 6,447 first names from ONS baby names data. A leader can also remove any comment before the report is shared, and comments only appear once 10 people have answered.
The rules in plain words
11 rules, each one enforced in code rather than left to the school. A school cannot turn any of them off.
- Nothing until 10 have answered
- A survey shows no results at all until 10 people have answered. The staff leavers survey, which collects a few answers at a time all year, needs 5.
- Small groups are hidden
- Results by role or year group appear only for groups of 10 or more staff, or 5 or more parents or pupils. If the hidden groups together could still be worked out by taking the others away from the total, the next smallest group is hidden too.
- Rare answers are pooled
- Any answer chosen by fewer than 5 people is hidden, together with enough others that it cannot be worked out by subtraction. The report says “too few to show”.
- Lopsided groups lose their percentage
- Where a group is big enough to show but fewer than 5 people in it gave one of the answers, its percentage is hidden. “90% agree” from ten people would point at the one who did not.
- Unanimous groups are hidden
- If everyone in a group gave the same answer, its percentage and its answers are both hidden. “100% agree” from a named group would tell everyone how each person in it answered.
- Comments need numbers too
- Comments are only shown once at least 10 people have answered (5 for the staff leavers survey), and a question with fewer than 5 comments is withheld. Comments appear in a fixed shuffled order, never the order they arrived.
- Names come out
- First names, names with titles, class codes, email addresses and phone numbers are removed from comments automatically. Leaders can remove any comment before sharing the report.
- No name, device or time
- A response stores no name, email, account, device or time of day. The date is kept, so a school can see how many answered over the days a survey was open, but never when.
- Quiet days are merged
- On the chart of responses by day, quiet days are merged with their neighbours so that no bar holds fewer than 5 responses. A school cannot see that one person answered on a given day.
- Personal links are not stored with answers
- A personal link stops anyone answering twice, but the link is never stored with the answers. Each email address used to send a link is deleted as soon as its link is used, 30 days after it was sent, or when the survey closes, whichever comes first.
- Too many answers, no breakdown
- If a survey gets more responses than the number of people it was sent to, someone may have answered twice to see a small group, so results by role are withheld.
Safeguarding
Anonymity has one deliberate exception. A comment that suggests someone may be at risk of harm goes to the school’s safeguarding leads, still without the writer’s name, because nobody can know it.
Fig. 2 The path of a worrying comment
Before the survey opens
A safeguarding lead is named
A survey cannot open until the school has named at least one safeguarding lead. We suggest two: the DSL and a deputy.
As each comment arrives
The comment is checked
Every written comment, from staff, parents and pupils, is checked against words and phrases that may indicate harm. Comments in scripts the check cannot read, such as Arabic or Urdu, are always passed on.
Within about 5 minutes
Safeguarding leads are emailed
The email does not contain the comment: leads sign in to read it. Nobody else on the school’s team sees it, and it is kept out of the report, the AI commentary and the download. If an email has not gone out after 15 minutes, the whole team sees a warning.
After 2 hours, if not yet read
A reminder, and the owner is told
Leads are reminded, and the account owner is told an alert is waiting, without seeing what it says.
When a lead reviews it
The outcome is recorded
No concern, logged in the school’s safeguarding system, action taken in school, or referred.
30 days after review
The comment text is deleted
The school keeps its own record in its own safeguarding system.
For pupils. Every pupil survey has a “Need to talk to someone?” link with Childline’s number, 0800 1111, and tells pupils the school will not know who wrote a comment.
What it will not do. Alerts will not catch everything: the check is word-based and can miss indirect language, sarcasm and unusual spellings. Alerts can arrive in the evening, at weekends and in holidays, when nobody may be watching. Alerts support a school’s safeguarding procedures; they do not replace them.
What respondents are told
Every survey opens with a short introduction. These are the words people see, quoted in full from two of the surveys. If you have been sent a survey, read what happens to your answers.
This survey is anonymous. Your school cannot tell who answered. Results are only shown for groups of 10 or more people. Written comments are read by the school. If one suggests that a child or an adult may be at risk of harm, including a concern about the conduct of someone who works with children, it goes straight to the school's safeguarding lead and is kept out of the report.
This survey asks what school is like for you, so we can make it better. It takes about 5 minutes. It is not a test and there are no right or wrong answers. You can skip any question you do not want to answer. Your teachers will not see who said what: answers are put together, and results are only shown for groups of 5 or more pupils. If you write something that makes us worried about your safety or someone else's, the school's safeguarding lead will read that comment. They will not know who wrote it, so they cannot come and find you. This survey is not the way to get help quickly, so if you need help now, talk to an adult you trust in school.
For your data protection lead
The data protection pack has the data processing agreement, the sub-processors and where data is held, the retention schedule, the security overview and a pre-filled DPIA. The privacy notice for staff can go out with a survey invitation.
Run your first survey today, free.
Sign in with your school email. No password and no card needed. You choose whether to pay after it closes.