A-Level Maths Statistics: a complete revision guide

Many A-Level Maths students arrive in Year 12 thinking statistics is the easier half of the course. It is true that the calculations are often more routine than pure mathematics, but the mark schemes are unforgiving on interpretation, assumptions and context. A student who can integrate by parts but cannot explain what a confidence interval means will lose marks they could easily have kept.

This guide covers the statistics content common to the major UK exam boards — Edexcel, AQA, OCR and OCR MEI. There are differences in emphasis and notation, so always check your own specification, but the underlying ideas are the same everywhere.

Start with the data, not the formula

Before you calculate anything, you need to know what kind of data you are dealing with. This determines which summary measures, diagrams and tests are appropriate. The classification is simple, but examiners test it constantly because students rush past it.

A common error is to calculate a mean for a qualitative variable. You cannot average 'North' and 'South'. Another common error is to use a histogram for discrete data with very few distinct values, where a bar chart is clearer.

Measures of location and spread

You need to know when to use the mean, the median and the mode, and when each is misleading. You also need to understand standard deviation, variance and interquartile range as measures of spread.

• Mean: uses every data point, but is pulled by outliers.
• Median: robust to outliers, but ignores most of the data.
• Mode: useful for categorical data or bimodal distributions.
• Standard deviation: the standard measure of spread, but sensitive to outliers.
• Interquartile range: a robust measure of spread for skewed data.

In exam questions, the skill is not just calculating these values. It is choosing which one to report and explaining why. If a distribution is skewed, the median and interquartile range are usually more representative than the mean and standard deviation. If the question asks you to compare two groups, quote both a measure of location and a measure of spread.

Probability: the rules that matter

A-Level probability builds directly on GCSE, but with more formal notation and more complex problems. The rules you need are not long, but you must apply them carefully.

1. Mutually exclusive events: P(A or B) = P(A) + P(B).
2. Independent events: P(A and B) = P(A) × P(B).
3. Conditional probability: P(A given B) = P(A and B) ÷ P(B).
4. Complementary probability: P(not A) = 1 − P(A).

Tree diagrams and Venn diagrams are not just ways to present a problem — they are ways to think about it. If you are unsure whether events are independent, drawing a Venn diagram often makes the structure clear. Probability questions frequently combine with combinatorics, so make sure your factorial and combination notation is secure.

The binomial distribution

The binomial distribution models the number of successes in a fixed number of independent trials, each with the same probability of success. The conditions are important: fixed number of trials, two outcomes, constant probability, independence.

You need to be comfortable with the formula, but in practice most questions are answered faster with tables or the calculator. The real exam skill is recognising when a situation is binomial and stating the assumptions clearly. If the question gives you a context, check whether the probability really is constant and whether the trials really are independent.

The normal distribution

The normal distribution is used for continuous data that is symmetric and bell-shaped. You need to be able to standardise a value using z = (x − μ) / σ, find probabilities from a standard normal table or calculator, and find values given a probability.

Many normal distribution questions are calculator-heavy, so practise with the exact model you will use in the exam. Know how to enter the mean and standard deviation, how to find P(X < a), and how to find the value of a given P(X < a). A sketch of the bell curve, with the area you want shaded, will prevent sign errors.

The normal approximation to the binomial distribution is also worth revising. It is appropriate when n is large and p is close to 0.5, and you must apply a continuity correction because you are approximating a discrete distribution with a continuous one.

Hypothesis testing

Hypothesis testing is the topic that most clearly separates students who understand statistics from students who have only memorised steps. You must know the logical structure, not just the recipe.

1. State the null and alternative hypotheses in words and symbols.
2. Choose the test and the significance level, and justify why it is appropriate.
3. Calculate the test statistic or the probability under the null hypothesis.
4. Compare your result to the critical value or critical probability.
5. Write a conclusion in context, using the wording 'there is sufficient evidence' or 'there is insufficient evidence'.

The most common mistake is to write the conclusion as a proof. You never 'prove' the alternative hypothesis. You only say whether the data provides enough evidence to reject the null hypothesis at the chosen significance level. Another common mistake is to confuse the p-value with the probability that the null hypothesis is true.

Correlation and regression

You need to be able to calculate the product moment correlation coefficient and the equation of a regression line, but the interpretation is where the marks are won and lost. A strong correlation does not mean causation, and a regression line should only be used to predict within the range of the data you have.

• Positive correlation: as one variable increases, the other tends to increase.
• Negative correlation: as one variable increases, the other tends to decrease.
• Correlation near zero: little or no linear relationship.
• Extrapolation: predicting outside the data range is unreliable.

When asked to comment on a correlation or regression result, always refer to the context. 'There is a strong positive correlation between revision hours and mock marks' is better than 'r is close to 1'.

Exam technique for statistics

Statistics questions reward clear communication. A correct number with no explanation will often score fewer marks than a slightly wrong number with the right reasoning clearly set out.

• Define every random variable you introduce.
• State assumptions such as independence, random sampling, or normality.
• Give answers to the required degree of accuracy.
• Interpret every result in the context of the question.
• Use the correct notation for distributions: B(n, p) and N(μ, σ²).

Time management matters too. Statistics questions can be wordy, and it is easy to spend too long on the early parts. Read the whole question before you start, identify where the marks are, and do not get bogged down in a calculation that is only worth one mark.

Resources for further practice

The best way to improve at statistics is to do real exam questions and study the mark scheme carefully. Textbook exercises are useful for learning the method, but only past papers teach you the style of interpretation the examiners expect.

When statistics does not click

Statistics often feels harder than it should because the notation and the wording hide simple ideas. If you can do the pure maths but keep losing marks on the interpretation, a few one-to-one sessions can make a big difference. The tutor can show you exactly where your written explanation is missing the mark scheme point and give you a repeatable way to phrase your answers.

Final revision plan

1. Review variable types and which summaries are appropriate for each.
2. Practise mean, median, standard deviation and interquartile range from grouped and ungrouped data.
3. Revise probability rules and practise conditional probability problems.
4. Do binomial and normal distribution questions until the calculator work is automatic.
5. Write out full hypothesis tests, including the conclusion in context.
6. Practise correlation and regression interpretations, especially causation and extrapolation.
7. Finish with at least one full statistics past paper section under timed conditions.

Statistics rewards methodical revision more than raw talent. If you learn the definitions, practise the interpretations, and study the mark schemes, the marks will follow.