OxfordAQA IGCSE Grading: The 9-1 Scale and What Grade Boundaries Mean

A grading scale built to be read the same way everywhere

Students arriving at an OxfordAQA school from a system that uses percentages, a ten-point scale, or a European Baccalaureate-style average often need a moment to adjust to the 9-1 scale used across the oxfordaqa igcse. It is worth explaining plainly, because a number on its own tells you nothing until you understand what it represents. Grade 9 is the highest attainable grade, reserved for the very strongest performance in a subject; grade 1 is the lowest passing grade; and a result below that threshold is reported as U, for unclassified. Between 9 and 1, each whole grade represents a distinct, examiner-defined standard of achievement, not simply an equal slice of the available marks.

Coming from a European system where a 15/20 or an 8/10 average maps neatly onto a fixed percentage band, it can feel unfamiliar that a 9-1 grade is not a fixed percentage at all. That single fact, that the scale is standards-based rather than percentage-based, is the key to understanding everything else in this guide.

Why OxfordAQA does not use a fixed percentage cut-off

If every exam paper were identical in difficulty every single year, a fixed percentage cut-off for each grade would work perfectly well: score 80% and you always get a 7, score 90% and you always get a 9. In practice, no exam board can guarantee that every paper in every subject in every series is exactly as demanding as the one before it, however carefully it is written and trialled. A paper that turns out, after the fact, to have been slightly harder than intended would unfairly punish an entire cohort of students if grades were awarded on a rigid percentage basis.

To correct for this, OxfordAQA, like every major exam board, sets grade boundaries after each paper has been sat and marked, rather than before. A grade boundary is simply the raw mark a candidate needs to reach a given grade in that specific paper, in that specific series. If a particular Mathematics paper turns out to have been more demanding than the previous year's, the boundary for a grade 7 might be set a few marks lower to compensate, so that a candidate of genuinely equivalent ability earns the same grade regardless of which year's paper they happened to sit.

How the boundary-setting process actually works

After a series of exams has been sat and marked, OxfordAQA's senior examiners review a combination of statistical evidence and script-level judgement before recommending where each grade boundary should sit. Statistical evidence includes how the current cohort's marks compare with previous cohorts on the same specification, adjusted for any known differences in the group of candidates. Judgemental evidence includes senior examiners physically reviewing scripts sitting right at the edge of a proposed boundary, checking that a script scoring one mark above the line genuinely demonstrates a meaningfully stronger standard than one scoring one mark below it.

1. The paper is set, trialled where possible, and sat by the full cohort of candidates that series.
2. Marking is completed and statistical patterns across the cohort are analysed against historical data.
3. Senior examiners review scripts at proposed boundary points to confirm the raw mark reflects the intended standard.
4. Boundaries are finalised and results are issued to schools and candidates on results day.

This process exists precisely so that a grade earned in a slightly harder year and a grade earned in a slightly easier year represent the same underlying standard, even though the raw marks needed to reach them differ.

A worked comparison across two series

Imagine a Physics paper worth 90 marks. In one series, examiners judge the paper to have been fairly standard in difficulty and set the grade 7 boundary at 63 marks. The following year, a slightly more demanding set of questions leads examiners to set the grade 7 boundary at 58 marks for that series. A candidate scoring 59 in the second year and a candidate scoring 64 in the first year have both cleared their respective grade 7 boundary, and both are certified as having reached the same standard, even though their raw scores differ by five marks. This is normal, expected, and the entire point of setting boundaries after the paper rather than before it.

Reading your own results against boundaries

When results are published, OxfordAQA also publishes the grade boundaries for that series alongside them, which lets a student see exactly how close they were to the grade above or below the one they achieved. This is genuinely useful information for planning a resit or for understanding how much room for improvement exists, but it should be read with a note of caution: a boundary that sat at 58 this year gives no guarantee that next year's boundary for the same grade will be anywhere close to that figure, because it depends entirely on how demanding next year's specific paper turns out to be.

Treat last year's grade boundary as background context, not a target. The only reliable way to secure a grade is to master the full specification, not to calculate the smallest number of marks that might scrape you across a line that could move.

How this compares with familiar European grading conventions

Students and parents used to a continental European grading convention, where a score out of 20 or out of 10 maps to a fixed threshold regardless of how hard that year's assessment was, sometimes find the boundary-setting approach counter-intuitive at first. The European convention generally assumes assessments are calibrated to be equally difficult every time, an assumption that works reasonably well for continuous school assessment but is harder to guarantee for a single, high-stakes national exam paper written fresh each year. The 9-1 approach trades a small amount of year-to-year predictability in raw marks for a much stronger guarantee that the underlying standard behind each grade stays genuinely comparable across years, which matters enormously once universities are comparing applicants who sat their exams in different series.

What this means for a study guide and revision plan

Because grade boundaries are unknown until after the exam is sat, a sound study guide never plans revision around hitting a specific raw mark. Instead, it targets full and confident coverage of the syllabus, since a candidate who has genuinely mastered the specification will clear whatever boundary is eventually set for that grade, whether the paper that year turns out easier or harder than usual. Building revision around full syllabus coverage rather than a guessed percentage target is the single most reliable strategy across every OxfordAQA subject.

• Study the full specification rather than trying to predict which topics will appear.
• Use past grade boundaries as general context for how much movement is typical, not as a fixed target.
• Practise full papers under timed conditions so your raw score reflects genuine exam performance, not familiarity with one specific paper.
• Remember that command words and mark allocations, not raw percentages, determine how much detail an answer needs.

The bigger picture

Understanding how OxfordAQA sets grades removes a surprising amount of anxiety from the final weeks before an igcse exam series, because it replaces a vague fear of an unknown percentage cut-off with a clear picture of what is actually being measured: a stable standard, checked and re-checked by senior examiners, expressed through a boundary that moves so the standard itself does not have to. Whatever country a student is sitting their papers in, and whatever grading convention they grew up expecting, the 9-1 scale used by OxfordAQA is designed to answer one question consistently: did this candidate demonstrate the standard of understanding that this specific grade represents.

A brief note on the move from A*-G to 9-1

Older students, parents and some school records may still reference the previous A*-G grading scale used before the 9-1 system became standard across the international gcse landscape. Broadly, the very top of the old scale (A* and the top of A) maps onto grades 8 and 9 under the new system, a solid pass under the old C grade maps roughly onto grade 4, and the two systems align closely enough in the middle of the range that comparisons remain useful for context. What matters for a current OxfordAQA candidate is simply that results are now reported on the 9-1 scale exclusively, and any comparison with the older lettered scale is for historical reference only, not something a current student needs to convert their own results into.

Common misunderstandings worth clearing up

A few misconceptions about grading come up repeatedly among students and parents new to the system. The first is assuming that a fixed percentage, say 90%, always guarantees a grade 9; as explained above, the actual mark needed shifts with each paper's measured difficulty. The second is assuming that a lower grade boundaries figure in one year means that year's exam was "easier to pass"; in fact, a lower boundary usually signals the opposite, that the paper was judged more demanding and the boundary was lowered to compensate. The third is treating a single subject's boundary as representative of every subject; boundaries are set independently for each paper, so a Mathematics boundary and a Geography boundary in the same series can differ substantially and neither tells you anything reliable about the other.

Frequently asked questions

• Are grade boundaries published before the exam? No, they are only finalised and published after marking is complete, since they depend on how the cohort actually performed on that specific paper.
• Can a school appeal a grade boundary? Individual candidates can request a review of marking, but grade boundaries themselves are set for the whole cohort based on statistical and judgemental evidence, not adjusted for an individual case.
• Do harder subjects get more generous boundaries? Boundaries reflect the demand of that specific paper in that specific series, not a subject's general reputation for difficulty, so this should never be assumed without checking the actual published figures.

How grade boundaries interact with tiered papers

In subjects that use Foundation and Higher tiers, each tier has its own independently set boundaries, since the two papers are different documents assessing different ranges of demand. A Foundation tier paper typically caps the highest available grade below the very top of the scale, while opening up the lower grades to a more accessible standard of question, and its grade boundaries are set specifically for that paper rather than borrowed from the Higher tier. This is another reason raw marks across a tiered subject cannot be compared directly between Foundation and Higher entrants: they are, in a very real sense, sitting different exams even though both count towards the same subject.

Keeping perspective across a whole course

It is easy, especially in the final months before results day, to fixate on the mechanics of grade boundaries as though they were the whole story. They are one part of a much larger picture that includes two years of steady content coverage, regular practice, and honest self-assessment against past materials. A student who understands how boundaries work is better equipped to interpret their results sensibly, neither panicking over a near-miss nor becoming complacent after a comfortable pass, but that understanding should sit alongside solid subject knowledge, not replace it.

Why boundaries differ so much between subjects

A grade 7 boundary in Mathematics might sit at a very different percentage of total marks than a grade 7 boundary in English Literature, and this is entirely expected rather than a sign that one paper is easier than the other. Different subjects assess different skills through different question styles, some heavily calculation-based, others reliant on extended written analysis, and the raw mark distributions that result are simply not comparable across subjects. Never use one subject's boundary as a benchmark for judging another.