How to Revise Cambridge A Level Probability & Statistics 2: A Complete Guide

Understanding the Probability & Statistics 2 Syllabus (9709 / 9231)

Probability & Statistics 2 appears as Paper 6 within the Cambridge 9709 Mathematics syllabus and is also included in the 9231 Further Mathematics route. The paper is 50 marks over 1 hour 15 minutes, consisting of structured questions. The syllabus covers six core areas: the Poisson distribution, linear combinations of random variables, continuous random variables, sampling and estimation, hypothesis testing (t-tests and Chi-squared tests), and probability generating functions (in some syllabus years).

The Poisson distribution section requires you to model events occurring randomly in time or space, calculate probabilities, and use the Poisson as an approximation to the binomial. Linear combinations covers the distribution of aX + bY where X and Y are independent, including combinations of normal and Poisson variables. Continuous random variables involves probability density functions, cumulative distribution functions, expectation, variance, and the use of the normal distribution as an approximation with continuity corrections.

Sampling and estimation covers unbiased estimates of population mean and variance, confidence intervals, and the Central Limit Theorem. Hypothesis testing extends S1 by introducing t-tests (when the population variance is unknown), paired-sample tests, and Chi-squared goodness-of-fit and independence tests. Examiners expect correct hypotheses, test statistics, critical values, and a conclusion in context.

Step 1 — Build Your Foundation First

S2 depends heavily on S1 knowledge. Before starting your S2 revision, confirm that you are fluent with: the normal distribution (standardising, finding probabilities, inverse normal), the binomial distribution (including cumulative tables), basic hypothesis testing (null and alternative hypotheses, significance levels, critical regions), and the concepts of expectation and variance. Weakness in any of these will bottleneck your progress in S2.

For each S2 topic, start with the theory before practising calculations. Understand why the Poisson distribution models rare independent events, why the mean equals the variance for a Poisson, and why the Poisson approximation to the binomial works when n is large and p is small. Understanding the "why" behind each distribution means you can reason about which model to use when the exam question describes a real-world scenario — a skill that is tested frequently.

Work through textbook examples for each distribution until you can set up the calculation without hesitation. The goal at this stage is not speed but accuracy — make sure you understand every step before moving to timed practice.

Step 2 — Practice Technique with MCQs

Multiple choice practice is especially valuable for S2 because many questions test your ability to choose the correct distribution, identify the right test, or apply the correct approximation. A well-designed MCQ forces you to decide: is this Poisson or binomial? Do I use a z-test or a t-test? Do I need a continuity correction? These decisions are exactly what the structured exam questions test in their opening parts.

Key concepts to drill with MCQs: the conditions under which the Poisson approximation to the binomial is valid (n > 50, np < 5 as a common rule of thumb), when to use the normal approximation to the Poisson (λ > 15), the difference between a one-tailed and two-tailed test, and how degrees of freedom are calculated for Chi-squared tests (for goodness-of-fit: categories minus 1 minus number of estimated parameters; for independence: (rows - 1)(columns - 1)).

Aim for 25-30 MCQs per topic. After each session, note every question where you chose the wrong distribution or test. These notes form a decision checklist you can review before the exam.

Step 3 — Exam Question Practice (The Most Important Step)

S2 exam questions follow predictable structures. A Poisson question will give a context (defects per metre of fabric, calls per hour), ask you to calculate specific probabilities, then test whether you can combine Poisson variables or use the Poisson approximation. A hypothesis testing question will describe an experiment and ask you to carry out a formal test at a given significance level. Knowing these structures means you can read the question and immediately identify what is being asked.

When using mark schemes, pay close attention to the required level of statistical reasoning. Cambridge examiners expect you to: state H₀ and H₁ using correct notation (e.g. H₀: λ = 3, H₁: λ > 3), calculate the test statistic, compare with the critical value (or calculate the p-value), and state a conclusion in the context of the original problem ("There is sufficient evidence at the 5% significance level to conclude that the mean number of defects has increased"). Omitting any of these steps loses marks.

For continuous random variable questions, practise integrating PDFs to find probabilities, deriving the CDF, and calculating E(X) and Var(X). These are computation-heavy and the marks are in the working — show every line of integration clearly.

Target at least 10 past paper questions per topic. For hypothesis testing and the Poisson distribution — which dominate most papers — aim for 15+. Revisit any question you scored below 75% on after one week.

Step 4 — Use Spaced Repetition for Formulas and Conditions

S2 has a significant number of formulas, distribution properties, and conditions that must be at your fingertips. Create flashcards for: the Poisson PMF (P(X = r) = e^(-λ)λ^r / r!), mean and variance of the Poisson (both equal λ), conditions for Poisson approximation to binomial, conditions for normal approximation to Poisson, the continuity correction rules, the unbiased estimate of population variance (s² = Σ(x - x̄)² / (n-1)), the t-test statistic formula, and the Chi-squared test statistic (Σ(O-E)²/E).

Also create flashcards for the decision rules: when to use a z-test vs. a t-test (known vs. unknown population variance, with a normal population or large sample), one-tailed vs. two-tailed critical values, and how to determine degrees of freedom for different Chi-squared tests. These decision rules are where students lose the most marks — not because the calculation is hard, but because they choose the wrong test.

Five to ten minutes of daily flashcard review, starting at least four weeks before the exam, will make these formulas and conditions automatic. This frees your cognitive load during the exam to focus on problem-solving rather than trying to remember whether the continuity correction adds or subtracts 0.5.

Step 5 — Exam Technique for Probability & Statistics 2

Always state your distribution. Before any calculation, write the distribution you are using: "Let X ~ Po(4.2)" or "X ~ N(50, 16/25)". This earns a method mark and ensures you do not accidentally use the wrong parameters in subsequent steps.

Continuity corrections. When approximating a discrete distribution (Poisson or binomial) with a normal distribution, you must apply a continuity correction. P(X ≥ 10) becomes P(Y > 9.5), P(X ≤ 10) becomes P(Y < 10.5), and P(X = 10) becomes P(9.5 < Y < 10.5). Forgetting the correction or applying it in the wrong direction is one of the most common errors.

Hypothesis test structure. Follow this sequence every time: define the parameter, state H₀ and H₁, state the significance level, calculate the test statistic, find the critical value (or p-value), compare, and conclude in context. Do not skip steps even if the conclusion seems obvious — Cambridge awards marks for the process, not just the verdict.

Chi-squared tests: combine small expected frequencies. If any expected frequency is below 5, you must combine adjacent categories before calculating the test statistic. Failing to do this invalidates the test and will cost you all subsequent marks in the question. State that you are combining and show the new expected frequencies.

Recommended Resources for Probability & Statistics 2

Cambridge past papers (official). Download Paper 6 (Probability & Statistics 2) from the Cambridge International website. Work through papers from the last five to six years. The earlier papers are useful for topic practice; the recent ones are best for timed full-paper simulation.

Nexelia. Provides 809 Cambridge-aligned MCQs and 529 exam questions with full worked solutions for Probability & Statistics 2, organised by chapter. The worked solutions show every step of hypothesis tests — from stating hypotheses through to the contextual conclusion — and include commentary on common errors. The AI study coach can explain any solution and help you understand why a particular test or approximation was chosen.

Your textbook. The endorsed Cambridge coursebook for Probability & Statistics (Chalmers or Dean & Dean) covers the theory with worked examples. Use it for first-pass learning of new distributions and test procedures, then transition to active practice as quickly as possible.

Statistical tables. Make sure you are practising with the same tables you will have in the exam. Cambridge provides cumulative Poisson and normal distribution tables. Know how to read them efficiently — fumbling with tables under time pressure wastes marks.

Common Mistakes Cambridge S2 Students Make

• Using the wrong distribution. The most costly error in S2. Poisson is for independent events occurring at a constant average rate. Binomial is for a fixed number of independent trials with constant probability. If the question describes a rate (per hour, per metre), it is almost certainly Poisson. If it describes trials (10 items tested), it is binomial.
• Forgetting continuity corrections. When approximating a discrete distribution with a continuous one, the continuity correction is not optional. Omitting it typically costs one mark and can change the conclusion of a hypothesis test.
• Using n instead of n-1 for the unbiased variance estimate. The sample variance s² uses (n-1) in the denominator, not n. This is because dividing by n gives a biased underestimate. The formula s² = (Σx² - n·x̄²) / (n-1) must be exact.
• Stating conclusions without context. Writing "reject H₀" is not sufficient. The conclusion must reference the original scenario: "There is significant evidence at the 5% level that the new treatment reduces recovery time." Examiners deduct marks for context-free conclusions.
• Wrong degrees of freedom in Chi-squared tests. For goodness-of-fit: ν = (number of categories after combining) - 1 - (number of parameters estimated from the data). For independence: ν = (rows - 1)(columns - 1). Mixing these up gives the wrong critical value and potentially the wrong conclusion.
• Not checking conditions before applying approximations. Before using a Poisson approximation to the binomial, state that n is large and p is small (or np < 5). Before using a normal approximation to the Poisson, state that λ is sufficiently large. Cambridge examiners award a mark for stating the condition — and withhold it if you jump straight to the approximation.