7 ways to get more from a parent survey for schools

Most schools already run some form of parent survey. The usual problems are familiar to anyone who has run one: a response rate that tells you little about who you have heard from, results that arrive after the decision they could have informed, and findings that never reach families or governors in a form that prompts action. A parent survey for schools earns its place when it works as a planning input, linked to the things leaders are accountable for, such as attendance, behaviour and pupil retention. Parents who understand how to reach the school, and who hear about progress in time to help, tend to be the parents whose children are in school and settled.

The seven strategies below are for headteachers, SLT members and trust leaders. Each one is a practical change you can make to the survey you already run.

1. Decide the decision before you write the questions

Surveys that start from a topic list tend to end with a report nobody owns. Surveys that start from a decision produce answers somebody has to act on. The mechanism is simple: a question with a decision attached has a reader waiting for the result, and a question without one has none.

Suppose a trust education director needs to decide whether to standardise Year 7 induction across six secondary academies. Every question is then tied to that choice: how parents heard about induction, whether they had early contact from a named member of staff, and whether their child seemed settled after half a term. A question about the school canteen or the car park has no place in that survey, however reasonable it is as a topic.

To put it into practice:

1. Write down the two decisions this survey must inform, in a single sentence each.
2. Draft the questions you think you need.
3. Next to each question, write the decision it informs and the person who will act on the answer.
4. Delete any question that has neither a decision nor an owner.
5. Agree review dates with the owners before the survey launches, so the results land in a meeting that already exists.

The common mistake is writing a broad survey out of habit, copying last year's version because it is there, and then having no idea what to do with the results. A long list of average scores gives a leadership team plenty to discuss and nothing to decide. If you find yourself keeping a question "because we always ask it", ask what you would do differently if the score dropped by ten points. If the answer is nothing, remove it.

To measure whether this is working, count the share of questions with a named owner, then check how many of those owners had completed an action by the next review date. If fewer than half of your questions led to an action, the survey is still too broad.

2. Ask about experience, not just satisfaction

Satisfaction questions invite a polite, general answer. A parent who is broadly content will tick "agree" to almost anything, and the score will move very little between a school with strong communication and one with weak communication. Experience statements ask whether something specific happened, which is harder to answer politely and easier to act on.

For example, imagine a school that currently asks parents to respond to "I am satisfied with communication". Replace it with "I know who to contact when I have a concern" and "I hear about my child's progress in time to help". The first can be fixed with a named contact on the website and in the planner. The second points to the timing of reports and progress updates. Neither fix could be derived from the satisfaction score.

It is a common assumption that high satisfaction means high engagement. It does not. A parent can be satisfied and still not know who to call, how their child is doing in maths, or whether the school would want to hear from them. Experience statements about welcome, communication and belonging reveal that gap.

Practical steps:

1. Write 10 to 15 statements that parents can agree or disagree with, using one consistent scale throughout.
2. Add one or two open questions, such as "What is one thing the school could do to help you support your child's learning?"
3. Pilot the draft with a few parents, ideally including at least one who does not usually attend school events, and ask them what each question meant to them.

The common mistake is writing leading or double-barrelled questions. "The school communicates clearly and promptly" combines two ideas, and a parent who finds messages clear but late has no honest answer. Split it. Avoid phrasing that signals the right answer, such as "Our excellent staff..."

Measure the spread of scores across questions, because a survey where everything scores within two points of everything else is not discriminating. Also track the number of usable free-text comments per response. A comment that names a specific event, person or process is usable. "Fine" is not.

3. Make it easy to answer on a phone, in the parent's language

Response rate is the share of invited parents who complete the survey. Low response is rarely a sign of apathy. More often the survey is hard to find, hard to open, too long, or written in a language the parent does not read comfortably. Every one of those barriers falls hardest on the families a school most needs to hear from.

Illustratively, a primary school sends the link by text on a Tuesday afternoon, with a single line explaining the purpose: "We are planning next year's parent meetings and want to know what would help you most." The message offers translation. A parent can open it at the school gate, answer in a few minutes and close it. Compare that with a link buried in the Friday newsletter that opens a login page.

To set this up:

1. Open the survey on a phone yourself and complete it, on mobile data rather than school wifi.
2. Cap completion time at five minutes. If it takes longer, cut questions using the test from strategy 1.
3. Send the link through the contact channels already held in your MIS, so every parent with a recorded number or email receives it directly.
4. Switch on translation and read-aloud options so parents with English as an additional language, or with limited literacy, can take part on equal terms.
5. Send one reminder, a few days before the window closes.

The common mistake is relying on a newsletter link or a login-protected form. Both filter out exactly the parents who are least connected to the school. A longer survey is also not a richer one. Beyond a certain length parents stop reading and start ticking, so extra questions cost you quality as well as responses.

Measure response rate overall, then by home language and year group. A strong headline rate can hide a Year 9 rate of 12 per cent or almost no responses from families who speak a language other than English. Those breakdowns tell you where to change the method next time.

4. Time the survey around the school year, and repeat it

One annual snapshot cannot show you whether anything you did made a difference. A survey is most useful as a series, where the same questions asked at the same point each year let you read change rather than guess at it. A pulse survey, meaning a short survey of a handful of questions, can sit between the main surveys to check a specific point without asking much of parents.

For example, imagine a secondary school that runs a short survey six weeks into Year 7 each autumn. By then parents have had a first progress report and a first parents' evening, and they have a view on how well their child has settled. Because the school asks the same questions every year, it can compare this year's Year 7 parents with last year's and see whether a change to induction made any difference.

To put this in place:

1. Fix a two-week window each year, away from report distribution, exam periods and the weeks either side of a holiday.
2. Keep the core wording unchanged from one cycle to the next. Add new questions at the end, never in place of the old ones.
3. Send a reminder at the mid-point of the window.
4. Add a short transition survey for the points where parents are most likely to feel uncertain, such as Year 6 to 7.

The most common mistake is changing questions every year because someone has a better idea. Each change breaks the trend line. If you improve a question, keep the old version alongside it for one cycle so you can see how the two behave.

Measure year-on-year change in your core question scores, and the response rate for the same window. A fall in response between cycles often arrives before a fall in scores, and it is worth investigating in its own right.

5. Break results down by group to find who is not being heard

A whole-school average describes the typical parent and hides the rest. Cohort segmentation means splitting results by groups of pupils whose parents answered, such as year group, SEND, Pupil Premium, EAL or a combination, so that gaps show up. The aim is to find which families the school is serving less well, and then to act on it.

Suppose a school scores well on "I felt welcomed when my child joined". Broken down, the score for Year 7 parents of pupils with SEND is noticeably lower. That finding is specific enough for the SENCO to run a drop-in for those parents in the first half term, which a school-wide score of 84 per cent never would have prompted.

The same breakdown applies to response rates. If very few Pupil Premium families respond, the score you have is a score for the parents who were already easy to reach.

Steps to follow:

1. Pull cohorts from your MIS so that groups are defined by the data you hold, not by what parents choose to say.
2. Set a minimum reporting group size before you see any results. Many schools use ten, though your data protection officer may advise a higher figure. Report nothing below it.
3. Compare scores and response rates for each group with the whole-school figure.
4. Assign one action to each meaningful gap, with a named owner.

The mistakes here are over-interpreting very small groups and exposing individuals. Where a group is small, a few parents can swing a percentage by 20 points, and a result from three families may identify them to staff. Combine groups or wait for another cycle rather than publish a number that cannot be trusted.

Measure the gap between your lowest-scoring group and the whole-school score on each core question across successive surveys. A narrowing gap is the clearest sign that the actions are reaching the families they were meant for.

6. Use the same survey across schools to compare fairly in a trust

Trustees usually hear about parent engagement through headteacher summaries, and those are written by different people, in different formats, about different questions. A common core of questions, asked on the same scale in the same window, lets a board see parent voice across every academy on equal terms. This is where a platform such as Perceptions helps, by running the same survey in each school and bringing the results into one view, but the principle holds with any tool.

For example, imagine a trust finds that one primary academy scores markedly higher than the others on "I know who to contact when I have a concern". Rather than treat that as a ranking, the education director asks the headteacher what they do. The answer is a named key contact for every family, shared on a card in September. Other academies adopt the practice and the next survey shows whether it travelled.

To set it up:

1. Agree a core set of questions centrally, drawing on the decisions from strategy 1 at trust level.
2. Allow each academy two or three local questions for its own priorities.
3. Run every academy in the same window, so seasonal effects affect all schools equally.
4. Report results in one trust-wide view, with the same chart layout for every school.

The common mistake is using comparison to league-table academies. Once heads believe a low score will be held against them, scores become cautious and response is nudged. Present comparisons as a way to find practice worth sharing, and ask the highest-scoring academy to explain what it does before asking anyone to explain a low score. Context matters too, because a small rural primary and a large urban secondary will not share the same baseline.

Measure the spread between academies on each core question, and count the practice-sharing actions taken between schools. A shrinking spread with no sharing activity should prompt a question about how it happened.

7. Close the loop with a "you said, we did" response

Parents who never hear back stop answering. A short public response shows that the survey changed something, and it is the cheapest way to improve next year's response rate. It also gives governors and trustees something concrete to scrutinise, because a list of owned actions is easier to hold to account than a set of percentages.

For example, imagine a one-page summary posted on the school app. It lists three findings, such as "Parents told us progress updates arrive too late to help at home", three actions, each with a named member of staff, and a date for the follow-up survey. Nothing in it is long or elaborate, and a parent can read it on a phone in a minute.

A workable sequence:

1. Share the summary within a few weeks of the survey closing, while parents still remember taking part.
2. Choose two or three actions, each with an owner and a date. More than that, and none of them gets done.
3. Report the same findings and actions to governors or trustees, so the commitments are visible to the people who oversee delivery.
4. Reference the changes in the invitation to the next survey, so parents see that last time's answers led somewhere.

Two mistakes are common. The first is promising too much, announcing a dozen actions when the school can deliver three. The second is sharing only the positive findings. Parents notice when a summary leaves out what they said, and an honest line such as "Many of you told us you were unsure who to contact, and we are changing that" builds more credibility than a page of good news.

Measure the response rate in the next survey, and the score on a statement such as "The school acts on parent feedback". If that score does not rise after you have published a response, the actions are not visible enough or are not the ones parents wanted.

Where to start this term

Begin with strategies 1 and 2. Fix the purpose of the survey, name the decisions and owners, and rewrite the questions as experience statements. That work costs staff time and nothing else, and it determines whether the results are usable.

Plan strategy 7 before you launch. Decide who will write the summary, when it will go out, and which governor or trustee meeting will receive it, so that the response is ready when the results arrive rather than drafted weeks later. Segmentation, trust-wide comparison and repeat timing can follow in the next cycle, once you have a first set of results to build on.

If you want to see how a common survey, cohort breakdowns and a trust-wide view work in practice, Learn more about our services.