7 October 2026 · Hugo Harrabin
Foundation tops out at a grade 5. On AQA Higher in 2026, missing a 4 by more than 17 marks meant a U. What each tier risks, who decides, and what to ask the school.
Sometime in Year 10 or Year 11, a parent hears that their child has been put in for the Foundation paper, or for the Higher one, or that they are "on the borderline". Both tiers carry a risk, and the school makes the decision.
This guide sets out what each tier offers and what it risks, with the marks behind the trade-off. It also covers the regulator's advice and the questions to ask while the entry can still change. If you want the wider picture of the qualification first, start with our parents' guide to GCSE Maths.
Key Takeaways
- Foundation awards grades 1 to 5. Higher awards 4 to 9, plus an "allowed" grade 3 for a narrow miss. Below that, a Higher student gets a U (unclassified) and no maths grade.
- On AQA in 2026, a grade 4 took 154 marks out of 240 on Foundation and 63 on Higher. The Foundation paper is easier, so the same grade takes more marks.
- On Higher, the allowed grade 3 started at 46 marks in 2026. A child who scored under 46 got a U.
- Ofqual, the exams regulator, has told schools that a student expected to get a grade 4 or 5 should be entered for Foundation, and that a 4 or 5 is worth the same on either tier. Certificates don't show the tier.
- The school decides, for most students early in the GCSE course through their maths set. Later moves are usually down, from Higher to Foundation. Entries for summer 2027 close on 21 February. In 2026, AQA let schools change a tier without a fee until 21 April; the 2027 date isn't out yet. Ask early.
Foundation tier awards grades 1 to 5. Higher tier awards grades 4 to 9, with an allowed grade 3 for a student who just misses a 4. Grades 3, 4 and 5 can be reached on either tier, and a student can only be entered for one tier in an exam series (AQA GCSE Mathematics 8300, general administration, retrieved 7 October 2026).
On AQA, both tiers are three papers of 90 minutes, 80 marks each, one without a calculator. The difference is in what the papers ask for and where the hardest questions fall.
| Foundation | Higher | |
|---|---|---|
| Grades available | 1 to 5 | 4 to 9, plus an allowed grade 3 |
| Worst outcome | Grade 1 | U (unclassified), no grade |
| Marks on number, ratio, proportion and rates of change | About 50% | About 35% |
| Marks on algebra | About 20% | About 30% |
| Marks for routine methods (standard techniques) | 50% | 40% |
| Where the grade 4 and 5 questions sit | Towards the end of each paper | At the start of each paper |
Sources: AQA 8300 specification; Pearson, GCSE Mathematics (9–1): Tiering guidance; DfE, GCSE mathematics: subject content and assessment objectives. The number, algebra and question-position rows are from Pearson's Edexcel guidance. Retrieved 7 October 2026.
Common questions are questions that appear on both tiers' papers, aimed at grades 4 and 5. They make up at least 20% of the marks on each tier (Pearson, Tiering guidance). On Foundation they come at the end, after the easier questions. On Higher they come first, and the paper gets harder from there.
So a child working at a grade 4 or 5 meets some of the same questions on either tier. Higher also gives more of its marks to reasoning and problem solving, where examiners report marks being lost when a question doesn't name the method (Eduqas, Examiners' Report, Summer 2025). We explain that gap in why a child can do the drill but not the exam question.
By design, neither. Exam boards are required to make it no easier and no harder to get a grade 4 or 5 on one tier than the other, and the regulator says a grade 4 on Foundation "is the same as a grade 4 gained on the higher tier" (Ofqual, letter to headteachers on tier entries, 23 January 2020, retrieved 7 October 2026).
Foundation asks for more marks on easier questions; Higher asks for far fewer marks on harder ones.
On AQA in June 2026, a grade 4 took 154 marks out of 240 on Foundation, which is 64%, and 63 on Higher, which is 26%. A grade 5 took 187 on Foundation and 97 on Higher (AQA, Grade boundaries, June 2026, retrieved 20 August 2026). Boundaries are set after marking and move each year, and Edexcel and OCR set their own. Our guide to GCSE Maths grade boundaries has the full tables for both tiers from 2023.
That can make Higher look like a free option: the same grade for fewer marks, with a 6 still in reach. For a child working at a 4 or 5, it carries more risk, because of how many marks on each paper are within their reach.
On Foundation, a child working at a 4 can attempt almost every question. Half of Foundation's marks are aimed at grade 3 and below, and at least another fifth at grades 4 and 5 (Pearson, Tiering guidance). In 2026 they could drop 86 of the 240 marks and still come away with a grade 4.
On Higher, most of the paper is beyond them. Half of Higher's marks are aimed at grades 7 to 9, and the grade 4 and 5 questions are concentrated in the common questions at the start of each paper (Pearson, as above). So a grade 4 child's 63 marks have to come from a much smaller pool of questions, covering fewer topics. If one or two of those topics go badly on the day, or a few careless slips cost marks, there is little else on the paper they can make the marks up from, and below 46 marks the result is a U.
The grade is worth the same either way. On Higher it depends on far fewer questions going right, which is why Ofqual advises Foundation for a child expected to get a 4 or 5.
The allowed grade 3, sometimes called the safety net, is the grade a Higher student gets for narrowly missing a 4. One who misses by more gets a U (unclassified) and no maths grade. Ofqual calls the allowed grade a "narrow 'safety net'" (Ofqual, January 2020).
On AQA in 2026 the allowed grade 3 started at 46 marks, 17 below the grade 4 at 63. That's half the 34-mark width of a grade 4, and about six marks a paper. A Higher student who scored 45 out of 240 left with no grade at all. Our GCSE Maths resit plan shows the 17-mark window on a chart.
Reporting on Higher tier in 2025, the Eduqas principal examiner wrote that for candidates "gaining less than 15% of the available marks", entry to Foundation "would have probably led to greater success and more confidence gained" (Eduqas, GCSE Mathematics Examiners' Report, Summer 2025, retrieved 10 August 2026).
The examiner was writing about children who scored very little, but the point about confidence applies just as much to a child working at a 4 or 5.
A teacher's view: "For me, confidence is the outcome that matters most. Picture the same child twice. In one version they sit Foundation, can answer most of the paper, and leave school with a 5 and the belief that maths is something they can do. In the other they spend two years being taught Higher content they can't yet reach, miss most of the marks in the exam, and come away with the same 5 feeling like they're no good at maths and never will be. And that's the good outcome on Higher, because that child is far more exposed to slipping to a 4, a 3 or a U. The certificate says 5 either way. What the child believes about themselves is the part they carry into whatever comes next." Hugo Harrabin, PGCE-qualified teacher and co-founder of Tugo
Foundation carries the opposite risk. It caps the grade at a 5, so a child who could have reached a 6 or 7 on Higher can't get one on Foundation. Sixth forms set their own entry grade for A-Level Maths, so check the ones your child is applying to. Some university courses set a GCSE Maths grade too, as our guide to Russell Group entry requirements sets out. For a child heading that way, Foundation rules out A-Level Maths wherever the entry grade is a 6 or above.
For a child expected to get a 4 or 5, the regulator's advice is clear. Ofqual's letter to headteachers says that because Higher students below the allowed grade get a U, "a student who is expected to achieve a grade 4 or grade 5 should be entered for foundation tier". It adds that some organisations recommend the opposite, and that doing so "puts students at risk of missing out on a grade" (Ofqual, January 2020). The letter is from 2020, but the tier structure it describes hasn't changed.
In practice that gives a rough guide:
Plenty of 4s and 5s are still earned on Higher. In summer 2023, 30% of grade 4s and 67% of grade 5s in GCSE Maths came from Higher, and among 16-year-olds 46% of all maths awards came from Foundation (University of Nottingham Observatory, analysing Ofqual data). The grade 4 figure includes candidates who weren't 16, and 94% of their maths awards came from Foundation, so for first-time candidates the Higher share is probably larger.
Schools also differ in how they enter borderline students. When Ofqual interviewed maths departments in 2017, it found that schools entered students expected to get the same grade into different tiers: "some schools opted to enter the students to the foundation tier, and others chose the higher tier" (Ofqual, GCSE mathematics: understanding schools' approaches to tiering, February 2017). It was a small study of twelve schools, and Ofqual says it may not be representative.
The evidence behind the decision varies as well. The Education Endowment Foundation says a school's predicted grade is usually based on mock results, but can also reflect previous attainment, how the student approaches their learning, target grades, which class they're in "and sometimes parental influence" (EEF, retrieved 7 October 2026). So two children with the same mock grade can end up on different tiers, because one has the stronger record over the previous years.
All of this means the useful question for a parent is what rule this school uses for a child on the borderline, and what evidence put your child on one side of it.
There is little research on which choice serves a borderline child better. The Education Endowment Foundation has commissioned Sheffield Hallam University to study that: the effect on GCSE results of entering pupils predicted a 4 or 5 into Higher or Foundation. It notes that "there is limited available research evidence" for schools to base the decision on. The study's scoping phase is due to finish in autumn 2026, with no results yet (EEF, Evaluating Secondary Maths GCSE Tier of Entry Decisions, retrieved 7 October 2026).
The school decides, and for most students it's decided early. Research summarised by Ofqual found that tier choices were "usually made quite early in a student's GCSE course" through setting, and that moving between tiers was "relatively rare". In the twelve schools Ofqual itself studied in 2017, the decision mostly followed teaching sets: top sets for Higher, the lowest for Foundation, with the discussion about the middle (Ofqual, 2017). Many of those schools taught their middle sets the Higher content first, so those students kept the option. Some were then moved to Foundation if they were struggling as the exam approached. Moves usually went that way, because a late move up leaves little time to learn the extra content.
For that middle group, schools in the study said the Year 11 autumn mock confirmed the tier. After that, changes were mainly for special circumstances, such as an unexpected absence (Ofqual, 2017).
The deadlines sit in the spring. Entries for the June 2027 exams close on 21 February 2027 (JCQ, Key dates June 2027). Changing a student's tier counts as an amendment, and in 2026 AQA let schools make amendments without a fee until 21 April (AQA, Exams officer update, March 2026). Edexcel and OCR set their own amendment dates, so check with the school if your child sits another board.
Dates checked 7 October 2026. The 2027 amendment deadline had not yet been published by AQA; next check January 2027.
So the time to ask is when your child's set is decided. For a child in the middle group, the Year 11 autumn term is the last good chance, because later changes are possible but less common.
Take these to Year 10 or Year 11 parents' evening, or put them in an email to the head of maths:
If you disagree, ask to see the evidence and agree a date to review it. Teachers have seen far more of your child's maths than a single mock score shows, so their reasoning is worth hearing in full.
This is a description of what we do, not a research finding. Our first lesson is free and includes a diagnostic assessment, which shows where a child is losing marks. It doesn't replace the school's mock on the actual paper, but it gives you specific questions to ask. After that, a weekly class of six to eight students, taught by a PGCE-qualified teacher, works through those gaps. You can see how we teach GCSE and A-Level Maths.
No. The highest grade on Foundation is a 5, however many marks a student scores. A child who needs a 6, for example to take A-Level Maths, has to be entered for Higher.
Neither, by design. On AQA in 2026 a grade 4 needed 63 marks out of 240 on Higher and 154 on Foundation, but Higher questions are much harder. Exam boards are required to make a grade 4 equally hard to get on either tier. The difference is the downside: below 46 marks on Higher, the result was a U.
No. Ofqual has confirmed that nothing on the certificate shows the tier of entry (Ofqual, January 2020), and a grade 4 or 5 is worth the same on either tier.
No. A student can be entered for only one tier in an exam series.
Resits work differently, because the student has a real mark from last time to plan from. Our GCSE Maths resit plan covers the tier question for resits.
Sit in on a lesson led by one of our expert PGCE qualified teachers.
Book your free lesson
