Skip to topic
    ← Back to course topics

    Section B — AQA GCSE Statistics

    Test yourself on Section B with AQA GCSE practice questions.

    Start free

    7 days Premium · Then free forever · No card, no charge

    Your focus

    1. Recognise the opportunities, constraints and implications for subsequent mathematical analysis involved in obtaining appropriate data through careful design of primary data collection techniques or through the use of reference sources for secondary data to ensure unbiased research.

    Section B exam tips

    Quick Revision Summary (Key Takeaway)

    Section B of the AQA GCSE Statistics examination features in-depth, scenario-based tasks assessing the complete statistical enquiry cycle across a single contextual problem. Students must demonstrate mastery in planning, data representation, mathematical calculation, and critical evaluation within real-world scenarios.

    Topic Overview

    Section B in AQA GCSE Statistics (examined in both Paper 1 and Paper 2) focuses on extended statistical enquiries set within realistic contexts. It assesses your ability to navigate the complete statistical enquiry cycle: from initial hypothesis formulation and sampling design, through data representation and calculation, to critical evaluation.

    Unlike the standalone questions in Section A, Section B questions are interconnected and carry substantial marks. Achieving a high grade requires fluent numerical calculation alongside clear, context-specific statistical commentary that directly references the data and problem provided.

    Key Concepts
    • →The Statistical Enquiry Cycle: Understanding the five core stages (planning, collecting data, processing/analysing, interpreting, and evaluating).
    • →Contextual Justification: Choosing appropriate averages (median vs mean) and measures of spread (IQR vs standard deviation) based on skewness and outliers.
    • →Sampling Evaluation: Identifying biases, selecting between probability and non-probability sampling methods, and assessing the validity of sampling frames.
    • →Critical Synthesis: Comparing multiple representations (e.g. box plots alongside scatter diagrams) to evaluate overarching claims or hypotheses.
    Examiner Tips
    • 💡When asked to compare two distributions, always write two separate sentences: one using medians/means with contextual units, and one using IQRs/standard deviations with contextual units.
    • 💡Check whether later parts of Section B depend on answers to earlier parts; if your calculated value seems unusual, use it consistently to earn follow-through marks.
    • 💡Never write that an experiment or sample is 'completely wrong' or 'useless'; instead, discuss specific limitations and suggest realistic methodological improvements.
    Common Mistakes
    • Treating correlation as proof of causation when evaluating conclusions in extended scenarios.
    • Giving one-sided comparisons between distributions (e.g. stating only that the median is higher without comparing the spread or consistency).
    • Using non-statistical, everyday language in place of standard terms like 'selection bias', 'stratified sample', 'bimodal', or 'outlier'.
    Revision Plan
    1. 1Stage 1: Review the five stages of the AQA statistical enquiry cycle and create a checklist of key evaluation criteria for each stage.
    2. 2Stage 2: Practise linked 4-mark and 6-mark Section B scenario questions from past papers, concentrating on box plot and cumulative frequency comparisons.
    3. 3Stage 3: Master standard scenario calculations, including capture-recapture, standardised rates, index numbers, and weighted means.
    4. 4Stage 4: Complete at least two full Section B sections under timed conditions (approx. 40-45 minutes each) and mark them strictly using AQA mark schemes.
    Exam Question Types
    • 📋Enquiry Evaluation Scenarios: Multi-part questions tracing an investigation from a student's initial hypothesis to a final conclusion requiring critique of their method and findings.
    • 📋Comparative Representation Tasks: Questions presenting dual datasets (such as back-to-back stem-and-leaf diagrams or parallel box plots) requiring structured contextual comparisons.
    • 📋Methodology and Sampling Tasks: Scenarios assessing how data was gathered, demanding identification of bias and specific suggestions for improvement.
    Command Word Expectations (AQA)
    Evaluate

    Review information, identify strengths and limitations of data or methods, and form a balanced conclusion supported by statistical evidence from the text or calculations.

    Compare

    Explicitly discuss similarities and differences between two distributions by quoting and interpreting both an average and a measure of spread with units in context.

    Justify

    Provide clear numerical proof, statistical definitions, or methodological rationale to explain why a particular technique, choice, or conclusion is valid.

    How Students Lose Marks (Examiner Pitfalls)
    Pitfall: Providing generic evaluation statements such as 'the sample is too small' or 'it is biased' without contextual evidence.
    ❌ Weak Answer (Loses Marks):The results are not reliable because her sample size is too small and people might lie on the survey.
    Example improved answer:The sample of 20 students from Year 11 is unrepresentative of the entire school population of 900 pupils across Years 7 to 11. Additionally, sampling only students in the library at lunchtime introduces selection bias towards higher-achieving pupils, meaning the conclusion cannot be generalised to all pupils.
    Examiner Tip: Always reference the specific context, numbers given in the scenario, and the target population when evaluating methodology.
    Pitfall: Comparing distributions by discussing only measures of average while ignoring measures of spread, or omitting units.
    ❌ Weak Answer (Loses Marks):Group A has a higher median than Group B, so Group A did better.
    Example improved answer:Group A had a higher median score (64 marks) compared to Group B (52 marks), indicating that on average Group A performed better. However, Group A also had a larger interquartile range (18 marks compared to 10 marks for Group B), showing that Group A's scores were less consistent.
    Examiner Tip: Structure comparison responses with two distinct statements: one comparing an average (median/mean) in context, and one comparing spread (IQR/standard deviation) in context.
    Step-by-Step Worked Solutions

    Question: An ecologist uses the capture-recapture method to estimate a trout population in a reservoir. She catches, tags, and releases 120 trout. Two weeks later, she catches 80 trout and finds that 15 are tagged. (a) Calculate an estimate for the total trout population. (b) State two assumptions necessary for this estimate to be valid.

    1. 1.Step 1: Set up the capture-recapture proportion using N = (M * n) / m, where M is the initial tagged sample (120), n is the second sample size (80), and m is the number of marked trout in the second sample (15).
    2. 2.Step 2: Substitute the values into the formula: N = (120 * 80) / 15 = 9600 / 15 = 640.
    3. 3.Step 3: State the mathematical estimate clearly as 640 trout.
    4. 4.Step 4: Identify two specific assumptions required: The population remains closed (no significant births, deaths, immigration, or emigration during the two-week period) and tags do not fall off or affect survival/catchability.
    Final Answer: (a) Estimated population is 640 trout. (b) Assumptions: 1. No changes in population (no births/deaths/migration); 2. Tags do not fall off or affect recapture probability.

    Question: A town council records road accidents across two areas with different populations. Area X has 42 accidents among 35,000 residents. Area Y has 27 accidents among 18,000 residents. (a) Calculate the crude accident rate per 1,000 residents for both areas. (b) Explain why crude rates are more appropriate than raw accident counts for comparing road safety between these areas.

    1. 1.Step 1: Calculate the crude rate for Area X: (Accidents / Population) * 1,000 = (42 / 35,000) * 1,000 = 0.0012 * 1,000 = 1.2 per 1,000 residents.
    2. 2.Step 2: Calculate the crude rate for Area Y: (Accidents / Population) * 1,000 = (27 / 18,000) * 1,000 = 0.0015 * 1,000 = 1.5 per 1,000 residents.
    3. 3.Step 3: Compare raw counts versus rates: Raw counts suggest Area X is more dangerous (42 vs 27), but crude rates show Area Y has a higher risk per person (1.5 vs 1.2 per 1,000).
    4. 4.Step 4: Justify why rates are superior: Crude rates take differing population sizes into account, providing a fair basis for comparison.
    Final Answer: (a) Area X: 1.2 per 1,000; Area Y: 1.5 per 1,000. (b) Crude rates account for differing population sizes, allowing for an equitable comparison of risk rather than simply reflecting total population size.
    Active Recall Memory Test
    What two statistical measures must always be included when comparing two datasets in Section B?
    Key Fact: A measure of average (e.g. median or mean) and a measure of spread/dispersion (e.g. IQR or standard deviation), both with units and context.
    Why is stratified sampling often preferred over simple random sampling in Section B enquiry scenarios?
    Key Fact: Because it ensures all key subgroups (strata) of the population are represented in the sample in exact proportion to their presence in the target population.
    When should you use the median and IQR instead of the mean and standard deviation to summarise data?
    Key Fact: When the data is skewed or contains extreme values (outliers), as the median and IQR are resistant to extreme values.
    Frequently Asked Questions
    How is Section B different from Section A in AQA GCSE Statistics?
    Section A contains shorter, disconnected questions that test discrete statistical techniques across various topics. In contrast, Section B consists of larger, linked multi-part questions structured around comprehensive real-world scenarios, testing the full statistical enquiry cycle from planning to final evaluation.
    How many marks is Section B worth on the AQA papers?
    On each of the two 80-mark papers (Paper 1 and Paper 2), Section B accounts for approximately half of the total marks (usually around 35 to 40 marks). This makes strong performance in scenario-based analysis essential for achieving higher grades.
    What is the best way to structure an evaluation question in Section B?
    Use a point-evidence-explanation structure tied directly to the scenario. State a specific strength or weakness of the method (e.g., sample selection, data reliability), cite numerical evidence or contextual facts from the paper, and explain how this impacts the validity of the conclusion.
    Can I get full marks if I use mean and range instead of median and IQR?
    You can get marks for mean and range if the question does not specify, but if the distribution is skewed or has outliers, examiners specifically look for median and interquartile range. Using range instead of IQR is also penalised if extreme values distort the overall measure of dispersion.
    What should I do if my calculation in part (a) of Section B seems incorrect?
    Continue working through the subsequent parts using your calculated answer. AQA examiners award full follow-through (FT) marks for correct method, reasoning, and interpretation based on an incorrect earlier numerical value, provided you do not make the problem trivial.