Edexcel A-Level Mathematics (9MA0) is assessed by three two-hour papers of 100 marks each: Paper 1 and Paper 2 on Pure Mathematics and Paper 3 on Statistics and Mechanics. Each paper is worth one third of the A Level. Marks are split across three assessment objectives — AO1 use and apply standard techniques (48–52%), AO2 reason, interpret and communicate (23–27%) and AO3 solve problems and model (23–27%). Examiner reports show most dropped marks are accuracy and communication marks: sign slips, missing units, unsupported calculator answers and wrong rounding.
Who this guide is for
This is for students sitting the full Edexcel A-Level in Mathematics, specification code 9MA0, and for parents trying to understand where the marks go. It assumes you are working through the course content already; the focus here is the shape of the exam and the exam-room habits that turn knowledge into marks. For one-to-one teaching of the content itself, see the A-Level Maths tuition page.
1. How the 9MA0 exam is structured
The qualification is examined entirely at the end of the course — there is no coursework and no module that can be banked early. You sit three papers, each two hours long and each marked out of 100, so each paper carries exactly one third of the final grade.
Two hours, 100 marks, 33.33% of the A Level. Draws on the whole Pure content: proof, algebra and functions, coordinate geometry, sequences and series, trigonometry, exponentials and logarithms, differentiation, integration, numerical methods and vectors.
Two hours, 100 marks, 33.33%. Examines the same Pure content as Paper 1 — the two Pure papers are not split by topic, so any Pure topic can appear on either, and you should revise all of it for both.
Two hours, 100 marks, 33.33%, but split into two clearly separated sections of 50 marks each: Section A Statistics (using the prescribed large data set) and Section B Mechanics.
All questions on every paper are compulsory — there is no choice — and a calculator is permitted throughout. That last point matters more than students expect: because everyone has a powerful calculator, the examiner cannot give marks for the answer alone, so marks migrate onto method and presentation, which is exactly where they get lost.
2. The three assessment objectives and their weightings
Every mark on the paper is tagged to one of three assessment objectives. Knowing roughly how they are distributed tells you why some questions are short and mechanical and others are long, wordy and worth chasing.
| Objective | What it rewards | Share of A Level |
|---|---|---|
| AO1 | Use and apply standard techniques; select and carry out routine procedures; recall facts, terminology and definitions | 48–52% |
| AO2 | Reason, interpret and communicate mathematically; construct rigorous arguments and proofs; make deductions; use notation correctly | 23–27% |
| AO3 | Solve problems in mathematical and real-world contexts; translate situations into models; interpret and evaluate the outcomes | 23–27% |
Roughly half the A Level is AO1 — the bread-and-butter technique you can drill. The other half is split between reasoning (AO2) and problem-solving and modelling (AO3), and this is the half that separates grades. The two Pure papers lean slightly more on AO2 (proof, "show that", "hence" chains), while Paper 3 carries a noticeably larger AO3 share because of its modelling, interpretation and large-data-set questions.
A practical reading: if you only ever practise short technique questions you are training for about half the paper. The grade boundary at A and A* is decided on the AO2/AO3 questions — proofs, "explain", "interpret", "criticise the model" — which reward writing as much as calculating.
3. Where Edexcel students actually lose marks
Edexcel publishes a principal examiner's report after each series, and the same themes recur year after year. Almost none of them are about not knowing the topic.
- Sign errors. The single most-cited slip, clustering in differentiation, integration by parts and vectors — and concentrated in the back half of the paper where students are rushing.
- Missing or wrong units. A correct number with no units, or the wrong units, drops the final accuracy mark. Velocity needs m s-1, force needs newtons, a rate of change needs its context units.
- Distance versus displacement. In mechanics and in kinematics-by-calculus, integrating straight through a change of direction gives displacement, not the total distance the question asked for.
- Unsupported calculator answers. Writing down a root, an integral or a summary statistic with no working forfeits the method marks even when the final number is right. The calculator can check your answer; it cannot earn the method marks.
- Wrong accuracy. Premature rounding mid-calculation, or a final answer given to too few or too many significant figures, costs accuracy marks the examiner is specifically watching for.
- Weak reasoning on "show that" and "prove". Jumping to the printed result without justified steps, or asserting rather than deducing, loses AO2 marks even when the algebra is broadly right.
The encouraging part is that these are recoverable. They are accuracy and communication marks, not knowledge marks, so a disciplined method — show every line, carry exact values, state units, and answer the question that was asked — claws most of them back.
4. Four worked exam-style questions
These four questions are written in the Edexcel style and span all three papers. They are original, not lifted from past papers, but they are pitched at genuine 9MA0 difficulty. For each one there is a full solution and a note on exactly what the examiner is rewarding and where candidates throw marks away.
Question 1 — Pure: differentiation and a stationary point
The curve \(C\) has equation \(y = 2x^3 - 9x^2 + 12x + 5\).
(a) Find \(\dfrac{dy}{dx}\) and hence determine the coordinates of the two stationary points of \(C\). (5)
(b) Use the second derivative to determine the nature of each stationary point. (2)
Solution (a). Differentiate term by term:
Stationary points occur where \(\frac{dy}{dx}=0\):
so \(x=1\) or \(x=2\). Substituting back into the original equation for the \(y\)-coordinates:
The stationary points are \((1,10)\) and \((2,9)\).
Solution (b). Differentiate again:
At \(x=1\): \(12(1)-18 = -6 < 0\), so \((1,10)\) is a local maximum. At \(x=2\): \(12(2)-18 = 6 > 0\), so \((2,9)\) is a local minimum.
Part (a) is mostly AO1: the differentiation and solving the quadratic earn method marks (M), and the correct coordinate pairs earn accuracy marks (A). A classic dropped mark is finding the two \(x\)-values and stopping — you must give full coordinates, so substitute back. Part (b) is AO2 reasoning: the second-derivative values alone are not enough, you must state the conclusion ("maximum", "minimum") and the sign that justifies it. Watch the sign of \(\frac{d^2y}{dx^2}\) — a sign slip here flips the nature of both points and loses the accuracy marks.
Question 2 — Pure: trigonometric identity and equation
(a) Show that the equation \(3\sin^2\theta + 8\cos\theta = 6\) can be written as \(3\cos^2\theta - 8\cos\theta + 3 = 0\). (2)
(b) Hence solve \(3\sin^2\theta + 8\cos\theta = 6\) for \(0^\circ \le \theta \le 360^\circ\), giving answers to one decimal place. (5)
Solution (a). Use the identity \(\sin^2\theta = 1 - \cos^2\theta\):
Expanding and collecting all terms on one side:
and multiplying through by \(-1\) gives the required form \(3\cos^2\theta - 8\cos\theta + 3 = 0\).
Solution (b). Treat this as a quadratic in \(c = \cos\theta\): \(3c^2 - 8c + 3 = 0\). By the quadratic formula,
Numerically, \(c = 2.215\ldots\) or \(c = 0.4514\ldots\). The first root is rejected because \(\cos\theta\) cannot exceed 1. So \(\cos\theta = 0.4514\ldots\), giving
Part (a) is AO2 communication: the mark is for using the Pythagorean identity and showing every line to reach the printed answer — do not skip from the first line to the last, because in a "show that" the steps are the marks. Part (b) blends AO1 (solving the quadratic) and AO3 (handling the domain). Two marks routinely vanish here: candidates forget to reject the root greater than 1, and they give only the principal value \(63.2^\circ\), missing the second solution in the range. Carry the unrounded value of \(\cos\theta\) into the inverse cosine — rounding to \(0.45\) early can shift the final answer at one decimal place.
Question 3 — Statistics: the normal distribution
The masses, in grams, of apples from an orchard are modelled by a normal distribution with mean 168 g and standard deviation 22 g. Let \(M\) denote the mass of a randomly chosen apple.
(a) Find \(P(M > 200)\). (2)
(b) Apples are graded "large" if they are among the heaviest 15% of the crop. Find, to the nearest gram, the minimum mass of a "large" apple. (3)
(c) State one reason why the normal distribution may not be a fully appropriate model here. (1)
Solution (a). With \(M \sim N(168, 22^2)\), standardise:
Then \(P(M > 200) = P(Z > 1.4545\ldots) = 1 - \Phi(1.4545\ldots) = 0.0729\) (to 3 s.f.), reading from the calculator's normal function.
Solution (b). The heaviest 15% lie above the 85th percentile, so we need \(m\) with \(P(M > m) = 0.15\), i.e. \(P(M \le m) = 0.85\). The corresponding \(z\)-value is \(z = 1.0364\ldots\) (inverse normal of 0.85). Then
so the minimum mass of a "large" apple is \(191\) g to the nearest gram.
Solution (c). A normal distribution is symmetric and assigns non-zero probability to negative masses, whereas real apple masses cannot be negative and may be skewed; so the model is only an approximation in the tails.
Parts (a) and (b) are AO1, but the method mark is for the standardisation or the explicit inverse-normal step — write \(z = (200-168)/22\) down, because a bare probability from the calculator with no working can lose the method mark. In (b) the commonest error is using \(z\) for 0.15 (a negative value) instead of 0.85, which puts the apple at the wrong tail. Part (c) is an AO3 modelling mark: a vague "it's not accurate" scores nothing — you must give a concrete reason (skew, or non-negativity of mass). Keep full calculator accuracy until the final rounding.
Question 4 — Mechanics: kinematics with variable acceleration
A particle \(P\) moves in a straight line. At time \(t\) seconds (\(t \ge 0\)) its velocity is \(v = 3t^2 - 12t + 9\) m s-1.
(a) Find the times at which \(P\) is instantaneously at rest. (2)
(b) Find the acceleration of \(P\) when \(t = 3\). (2)
(c) Find the total distance travelled by \(P\) in the interval \(0 \le t \le 3\). (4)
Solution (a). The particle is at rest when \(v = 0\):
so \(P\) is at rest at \(t = 1\) s and \(t = 3\) s.
Solution (b). Acceleration is \(a = \dfrac{dv}{dt} = 6t - 12\). At \(t = 3\), \(a = 6(3) - 12 = 6\) m s-2.
Solution (c). Because the particle changes direction at \(t=1\) (where \(v\) changes sign), the total distance must be computed in two stages, not as a single integral. Displacement is \(\int v \, dt\), and \(\int (3t^2 - 12t + 9)\,dt = t^3 - 6t^2 + 9t\). Evaluating each leg:
The second leg gives a displacement of \(-4\) m (the particle moves back). The total distance travelled is the sum of the magnitudes:
Parts (a) and (b) are AO1: differentiate for acceleration, solve for rest, and state units (m s-2) on the acceleration or lose the final mark. Part (c) is the AO3 discriminator and where most marks are lost across the cohort. The method mark requires recognising that the particle reverses at \(t=1\) and splitting the integral at that point; a single integral \(\int_0^3 v\,dt = 0\) gives displacement and scores almost nothing for "total distance". The accuracy mark is for combining the magnitudes to get 8 m. State units throughout. This is the textbook distance-versus-displacement trap the examiner reports flag every year.
5. Edexcel exam technique that protects marks
Pulling the examiner-report themes together, a handful of habits protect more marks than any extra topic revision at this stage.
- Show the method even when the calculator gives the answer. Write the integral, the standardisation, the quadratic formula. Method marks are awarded for the line, not the number.
- Keep exact values until the very end. Carry surds and full decimals through the working; round only the final answer, and to the accuracy the question demands.
- Read the command word. "Show that" wants every step; "hence" means use the previous part; "state" wants a short answer; "total distance" is not displacement. The command word tells you which assessment objective is being tested.
- Always attach units and a domain check. Units on every contextual answer; reject out-of-range roots; give all solutions inside a stated trigonometric interval.
- Budget time by marks. Roughly one mark per minute over two hours, leaving a little slack. The long AO2/AO3 questions at the end are where grades are made — do not let early arithmetic eat the time they need.
If you would like structured practice on the question types that decide the grade — proof, modelling, "show that" chains and the Statistics and Mechanics paper — that is exactly what targeted A-Level Maths tuition is for.
A-Level Maths tuition
Edexcel A-Level Maths rewards disciplined method as much as raw knowledge. For one-to-one help with Pure, Statistics, Mechanics or exam technique, see A-Level Maths tuition, A-Level & admissions tutoring or book a free consultation.