How to use the figures

Make sure the numbers answer the same question.

Before comparing schools, check what each figure counts and which years it covers. The Schools Record keeps these differences visible so you can see what a comparison supports and what still needs an answer.

Three checks before drawing a conclusion

The same result

A-level A* grades, Pre-U distinctions and IB points describe different qualifications. Similar labels do not make their results interchangeable.

The same group

A percentage of subject entries is different from a percentage of pupils. A university offer is different from a place taken up.

The same years

Compare years in which both schools have a usable figure. A missing result stays missing; it is never treated as zero or estimated to fill a gap.

What is being counted?

One pupil normally takes several subjects. If a school reports that 60% of its A-level entries received A*, that does not mean 60% of its pupils achieved A* in every subject. The percentage’s denominator is the total it is calculated from; where that total is unknown, the record says so.

Terms used in school results and university records
Term	What it counts
Candidates or pupils	People in the group described by the publication.
Exam entries	Results for individual subjects. One pupil usually contributes several entries.
Applications	Applications attributed to a particular admissions cycle and application centre.
Offers or admissions	Positive admissions decisions. They do not, by themselves, establish that a pupil took up a place.
Final destinations	Places reported as taken up by the leavers or entry-year group described.

Why qualifications are kept separate

A-level, Pre-U, IB and the old and new GCSE grading systems have different scales. The comparison tool includes a result only when its qualification, grade and reported group match the selected measure. A change of qualification can therefore create a break in a school’s chart.

Qualification definitions used in the record

A level

A level · A*–E

Denominator: % of published A-level entries

Cumulative bands are shown separately: A*, A*–A, A*–B and any wider band actually reported.

A level

A level · A–E before A*

Denominator: % of published A-level entries

The A* grade did not yet exist. The exact published bands are A, A–B and, where supplied, the source-labelled pass rate; the pass threshold is not inferred.

A level / IB Higher Level

A level / IB HL · school crosswalk

Denominator: % of the source-defined combined grade entries

This pooled school measures is not an A-level-only result and is never mixed with one.

GCSE / IGCSE

GCSE / IGCSE · 9–1

Denominator: % of published numbered-grade entries

Cumulative bands are explicit: grade 9, grades 9–8, grades 9–7 and any wider band reported.

GCSE

GCSE · numbered transition

Denominator: % of the source-defined numbered-grade entries

Only the exact numbered bands printed by the source are shown; no grade-9 value is inferred.

GCSE / IGCSE

GCSE / IGCSE · A*–G

Denominator: % of published subject-grade entries

Cumulative bands are shown separately: A*, A*–A and each wider band actually reported.

GCSE

GCSE · pre-A* scale

Denominator: % of published GCSE entries

The A* grade did not yet exist. Westminster's exact published bands are A and A–C; neither is relabelled as an A* measure.

IB Higher Level

Denominator: % of published IB Higher Level grade entries

Grade-entry percentages are kept separate from the candidate-level mean Diploma points measure.

Cambridge Pre-U

Denominator: % of published principal-grade entries

D1–D3 and wider D1–M bands are distinct cumulative thresholds; neither is relabelled as A level.

Source-defined qualification mix

Published equivalence measure

Denominator: Source-defined denominator

The source wording is retained because the underlying conversion basis is not fully encoded.

A level / Cambridge Pre-U

A level / Pre-U · published crosswalk

Denominator: % of the source-defined combined entries

This mixed-qualification measures is kept separate from pure A-level and pure Pre-U series.

GCSE / IGCSE

GCSE / IGCSE · published crosswalk

Denominator: % of the source-defined combined entries

These are published cross-scale bands, not universal grade equivalences. Numbered-only metrics remain separate.

Qualification not resolved

Unresolved grade measures

Denominator: Not safely encoded

Ambiguous examination percentages are withheld instead of receiving a generic top-grade label.

Why 2020 and 2021 are shown separately

Grades in 2020 were centre assessed and those in 2021 were teacher assessed, rather than awarded through the usual external examinations. Where the records identify these years, they are excluded from like-for-like exam comparisons by default. The comparison tool can show them separately for context.

Offers, acceptances and destinations answer different questions

A university application table may describe an admissions cycle. A school’s destination list may describe a leaving year. The groups and dates can differ, so a count of offers cannot automatically be compared with a count of destinations.

University pages show the school publications that name each institution. A mention alone does not establish how many pupils went there, or the chance that a future applicant will secure a place.

How the comparison is calculated

An annual difference compares the same measure, grading system, reported group and year at two schools. Where a figure is disputed, it is excluded until the sources support a documented choice under the rules for settling a difference. Charts and calculated rates use recorded figures; they do not replace the figures printed in the source.

A median is never taken across a change of qualification, population or outcome definition. If the matched years span such a change, the comparison reports each constant-definition segment on its own and states no combined figure; source observations are never relabelled to make them agree.

The comparison page’s period check tests a fixed set of windows: the full series, the latest three and five available years, and each year left out in turn. The claim checker tests every eligible window with at least three or five matched years and a constant definition. Both show how dependent a conclusion is on the period selected. Neither establishes statistical significance, a probability, or that a school’s teaching caused the results.

How verdicts are graded

A verdict needs at least 3 matched years, and with fewer only the values are shown. A typical gap under 1 percentage point, or a direction shared by fewer than 60% of the tested periods, is reported as marginal.

The verdicts a comparison can give
Verdict	When it is given
Values only	Fewer than 3 matched years. The figures are shown and no direction is stated.
Level	At least 3 matched years, and the typical annual difference is zero.
Marginal	At least 3 matched years, but the typical gap is under 1 percentage point, or fewer than 60% of the tested periods point the same way.
Clear	At least 3 matched years, a typical gap of 1 percentage point or more, and at least 60% of the tested periods pointing the same way.

The size rule applies to percentages; a gap in destination counts is graded on the matched years and the tested periods alone.

Reading percentages and summary differences

A change from 50% to 55% is five percentage points. When the comparison reports a median annual difference, it summarises the differences in the matched years. It is not a combined rate calculated from all pupils or all subject entries across the period.

A published range keeps both bounds. Wording such as “over two-thirds” keeps that limitation, and an estimate remains labelled as an estimate. These figures are not converted into exact observations for the comparison.

What happens when the evidence is incomplete?

The collection includes results tables, dated announcements, annual reports, school magazines and official government or university tables. Each figure keeps the definition and source information available for it. A link to a publication does not, on its own, mean that every year or figure in it has been checked.

Missing years

A missing year may reflect unfinished research, an inaccessible source or a figure published in a form that cannot be compared. Coverage records distinguish recorded figures from checks against original publications. An unsuccessful search is not proof that the result was never published.

Conflicting publications

Initial results, revised results and figures for different groups remain identifiable. Figures counted from a published list are distinguished from totals printed by the publisher.

Since 23 September 2026, a difference between two publications is settled only when one of the rules below is met. Each settlement is dated and published in the revision log of the figures it affects, and earlier versions stay visible. A newer page or a more precise number still does not settle a difference on its own.

How a difference between publications is settled
Outcome	When it applies	What the record does
Revised by the school	The same publisher issued both versions, for example both signed by the school’s Examinations Officer. Both versions count the same population: the same entry total. The changed cells are identified by comparing the versions row by row. The later version is the one the school now links or labels as final.	The record shows the current version. The earlier version stays in the revision log with its date and link, and the corrections ledger records the change.
Consistent at the published precision	The less precise figure equals the more precise figure rounded half up to the same number of decimal places. Where a table prints its grade counts, the exact share from those counts is used, so a rounded figure is never rounded twice. Truncation, or any other convention, does not qualify.	Not a disagreement. Both figures stay visible and the more precise one is used.
Different groups counted	The figures count different populations, for example A-level entries alone and A-level plus Pre-U entries. The difference is established by the publication’s own label, or by reproducing each figure exactly, at its printed precision, from the school’s separate tables.	Both figures are shown, each labelled with what it counts. Only a figure that matches a measure’s definition enters that measure’s comparison.
Sources differ	None of the three rules above is met.	Every version stays visible. The year is left out of comparisons for that measure, and the reason is stated.

What can change a figure or its label

A headline can be labelled as an A-level result while counting Pre-U or IB grades as well. A figure’s value or label changes only on positive evidence: the school’s own re-issued table, the publication’s own wording, or an exact reproduction from the school’s own tables. Not finding a source online is never evidence that a figure is wrong or mislabelled. Many figures come from archived prospectuses, magazines and documents that are no longer online, and they stay as recorded. A figure shown to count a different group keeps its place in the record with its own label.

Corrections and earlier versions

Automated checks flag possible errors; they do not rewrite published results. A correction needs a supporting source and a recorded explanation, and earlier values remain preserved. New observations and source checks are added without rewriting the original collection.

Explore the evidence · See recorded corrections and disagreements

What the figures cannot tell you

Published outcomes can help you question a claim and understand a school’s academic record. They cannot determine whether a school is right for a particular child or predict that child’s results or admission.

The comparisons do not control for prior attainment, admissions selectivity, subject choices, the make-up of a year group or school aims. They cannot isolate teaching quality or establish an overall school ranking.

Try a comparison