Statistical Sampling (AS Unit 2: Applied Mathematics A) — WJEC A-Level Mathematics
Statistical Sampling (AS Unit 2: Applied Mathematics A) explained
Statistical Sampling (AS Unit 2: Applied Mathematics A) covers 4 specification points, including Understand and use the terms 'population' and… and Use samples to make informal inferences about….
Read the full explanation
Open a point below for its explanation and revision guidance.
Understand and use the terms 'population' and 'sample'.
In statistical theory, a population comprises the entire collection of individuals, objects, or measurements about which information is sought, while a sample is a selected subset of the population. A complete enumeration of every population member constitutes a census. Although a census provides completely accurate parameters with zero sampling error, it is frequently unfeasible due to prohibitive costs, excessive time constraints, or the destructive nature of testing (such as testing the breaking strain of cables). Sampling offers a practical, cost-effective alternative, though it introduces sampling variation that must be accounted for during inference.
Use samples to make informal inferences about the population.
Calculated sample statistics serve as point estimates for unknown population parameters. Informal inference involves examining sample characteristics—such as the sample proportion, central tendency, or dispersion—and making reasoned deductions about the wider population while acknowledging the presence of sampling uncertainty. For example, finding that 35% of a random sample of voters support an initiative allows an informal estimate that roughly 35% of the electorate does so. Candidates must recognize how sample size governs precision: larger samples reduce sampling variability, producing more reliable informal estimates.
Understand and use sampling techniques, including simple random sampling, systematic sampling and opportunity sampling.
Different sampling techniques select members from a target group using probabilistic or non-probabilistic designs. Simple random sampling allocates every member an equal probability of selection, typically executed by numbering a sampling frame from 1 to N and picking items via pseudo-random numbers without replacement. Systematic sampling selects every k-th member (where k ≈ N/n) from an ordered list following a random starting point between 1 and k, but risks bias if the list possesses periodic structure. Opportunity (convenience) sampling chooses subjects who are readily available; while swift and inexpensive, it introduces severe selection bias.
Select or critique sampling techniques in the context of solving a statistical problem, including understanding that different samples can lead to different conclusions about the population.
Selecting an effective sampling design requires matching theoretical methods to practical constraints. Candidates must recognise that independent random samples drawn from the identical population will naturally yield varying estimates due to sampling variability. Non-random methods like opportunity sampling introduce systematic bias that cannot be mitigated by merely increasing sample size. When critiquing survey designs, learners assess coverage errors, non-response bias, exclusion of key demographics, and sampling frame availability, proposing refined probability-based alternatives.
Your focus
- Distinguish between an entire population and a representative sample in statistical investigations.
- Evaluate the practical advantages and limitations of a census compared to taking a sample.
- Identify real-world scenarios where destructive testing makes a full census impossible.
Show all 12 objectives
- Estimate unknown population parameters using sample proportions and sample means.
- Explain how sample size influences the variability and reliability of informal inferences.
- Formulate cautious, evidence-based statements about a population from sample data.
- Implement simple random sampling using a numbered sampling frame and random number generation.
- Calculate the sampling interval and random starting index for systematic sampling procedures.
- Compare the operational trade-offs between probability sampling and opportunity sampling.
- Critique proposed sampling designs by identifying coverage errors and selection bias.
- Explain why natural sampling variation leads different random samples to produce different inferences.
- Select and justify appropriate sampling techniques for complex real-world statistical problems.
Statistical Sampling (AS Unit 2: Applied Mathematics A) exam tips
Marking Points
- defining a population as the entire set of items under statistical consideration
- defining a sample as a finite subset chosen from a population
- identifying a practical constraint (such as cost, time, or destructive testing) that necessitates sampling
- calculating the relevant sample statistic such as a sample mean or sample proportion
- stating an informal inference about the corresponding population parameter
- explicitly acknowledging sampling variability or uncertainty in the inferred conclusion
- describing how to number a sampling frame and choose items using random numbers without replacement
- determining the sampling interval k = N/n and selecting a random start in systematic sampling
- identifying opportunity sampling and explaining why it is susceptible to selection bias
- identifying a specific source of bias or structural flaw in a proposed sampling method
- explaining that natural sampling variability causes different samples to yield differing conclusions
- recommending and justifying an improved sampling methodology suited to the population structure
Examiner Tips
- 💡Define populations precisely with explicit geographical and temporal boundaries.
- 💡Cite destructive testing whenever a question asks why a manufacturer cannot inspect every product.
- 💡Use cautious phrasing such as 'the evidence suggests' or 'it is estimated that' rather than claiming proof.
- 💡Check whether sample selection was unbiased before generalising findings to the wider population.
- 💡State all three steps for simple random sampling: assign unique numbers, generate random numbers, ignore duplicates.
- 💡Remember that opportunity sampling lacks a sampling frame and cannot guarantee equal selection chances.
- 💡Contextualise your critique directly (e.g. interviewing shoppers on a weekday morning excludes full-time workers).
- 💡Distinguish between natural random sampling error and systematic bias caused by poor design.
Common Mistakes
- assuming a census is always preferable without recognising practical drawbacks like destructive testing
- confusing sample statistics (such as x̄ and s) with population parameters (such as μ and σ)
- stating sample conclusions as absolute certainties about the population rather than tentative estimates
- ignoring the role of sample size when evaluating the reliability of an informal inference
- forgetting to discard duplicate random numbers when executing simple random sampling without replacement
- failing to recognise that cyclical patterns in a list invalidate systematic sampling
- criticising a sampling method purely on sample size without identifying structural selection bias
- assuming that different sample results imply that an error was made during data collection