Statistical sampling opens the Statistics section of Paper 3 and is AS (Year 1) content, so by the final exams it's often assumed knowledge rather than something freshly taught. It covers population versus sample versus census, and the selection, description and critique of five named techniques: simple random, stratified, systematic, quota and opportunity (convenience) sampling.
The topic is small on its own — typically around 3 of the 300 marks available, usually as a short part-question — which makes it easy to under-revise. It's also one of the few statistics topics regularly set against the pre-released Large Data Set, so a question may ask you to describe or critique a sampling method drawn from the weather-station data rather than an abstract scenario.
Because marks rarely cluster here, precision matters more than depth: examiners want the correctly named technique, described with enough contextual detail to show real understanding, not an extended calculation.
The specification statements this topic covers. AS = Year-1 content, also assessed in the standalone AS course (8MA0); A2 = full A level only. Typical share of a 300-mark series: ≈3 marks — our estimate from the 2018–2025 papers, not an official weighting.
| Ref | Spec statement | Level |
|---|---|---|
| 1.1 | Sampling techniques | AS |
Examiners repeatedly report a quota or opportunity sample being labelled stratified, or strata named without any random selection inside them. If a question specifically asks for stratified sampling, a method where the researcher chooses who takes part earns no credit, however sensible it sounds — state the random step explicitly, such as numbering each stratum and using a random number generator.
A stratified or quota description that never names the actual strata — year group, weather station, whatever the scenario specifies — is marked as incomplete even when the method itself is right. Write the technique in the language of the question, not in generic terms.
A question can be set against the pre-released weather data rather than a generic population, asking how a sample of days or stations was, or should be, selected. Detailed recall of values isn't required, but not knowing the data set's basic shape — several UK and overseas stations, two named years — costs marks unrelated to sampling theory itself.
When asked to comment on or criticise a method, a generic line such as 'it isn't random so it's biased' is too vague to credit. Examiners want the specific reason it matters in that situation — for example, that an opportunity sample taken only on weekday mornings would miss anyone not around then.
Choosing convenient or targeted individuals within each group and still calling it stratified is a common slip. Stratified sampling needs random selection inside every named stratum — without that, it's quota or opportunity sampling, and won't earn stratified-method marks.
Not saying which groups the population was divided into, or how selection inside each group is actually randomised, reads as incomplete even when the right technique is chosen. Check any answer against both requirements: named groups, genuine randomness within them.
In LDS-linked questions, it's easy to assume the terminology and structure will be obvious rather than revised beforehand — for instance, not knowing there are several UK and overseas stations across two specific years. That unfamiliarity costs marks that have nothing to do with sampling method itself.
We haven’t published checked questions for this topic yet — a worked sample appears here only once a question has passed every check. In the meantime you can practise in the app.
Sampling carries no formulae to memorise — the whole topic is definitions and judgement, so the useful revision is learning to describe each technique precisely: what makes it random or not, how it uses the context's natural groups, and one genuine advantage and disadvantage against a census. A short, consistent template for each technique means you reach for the same wording under pressure, including the random step and the named context.
There's no calculator skill specific to this topic, so time is better spent on the Large Data Set itself where your course uses it: knowing the stations, the years covered and the rough shape of the variables means a sampling question set in that context doesn't cost marks before you've reached the sampling method.
When practising, check your own answers against what examiners actually flag — did you name the strata, state the randomness, give a reason tied to the scenario rather than a generic one? Sampling part-questions are short, so work through several from different series in one sitting and compare your wording against the mark scheme's accepted answers.
All 19 topics: Edexcel A level Maths topic guides. Reference: formula booklet vs memorise and grade boundaries.
Ready to practise this topic properly? Start free — no card needed.
Original questions written for the Pearson Edexcel A Level Mathematics (9MA0) specification. Not affiliated with or endorsed by Pearson.