B4 — AQA GCSE Statistics
Test yourself on B4 with AQA GCSE practice questions.
Start free
7 days Premium · Then free forever · No card, no charge
Your focus
- Know the key features to be considered when planning data collection:
B4 exam tips
Quick Revision Summary (Key Takeaway)
B4 in AQA GCSE Statistics covers the analysis of bivariate data, focusing on scatter diagrams, correlation, and the calculation and interpretation of the product moment correlation coefficient (PMCC). Students must be able to describe relationships, identify outliers, and understand the difference between correlation and causation.
Topic Overview
B4 in AQA GCSE Statistics focuses on bivariate data analysis, where you explore the relationship between two variables. You will learn to construct and interpret scatter diagrams, calculate and interpret the product moment correlation coefficient (PMCC), and understand the distinction between correlation and causation. This topic is essential for analysing real-world data and making informed predictions.
This section builds on your knowledge of data representation and summary statistics from earlier units. It is a key component of the GCSE Statistics course because it develops critical thinking about data relationships and prepares you for more advanced statistical techniques. Mastery of B4 is crucial for exam success and for understanding how statistics is applied in fields such as science, economics, and social research.
Key Concepts
- →Scatter diagrams: plotting paired data to visually assess the relationship between two variables, including identifying outliers and describing correlation (positive, negative, none) and strength (strong, moderate, weak).
- →Product moment correlation coefficient (PMCC): a numerical measure of linear correlation between -1 and 1, calculated using the formula r = (nΣxy - ΣxΣy) / sqrt((nΣx² - (Σx)²)(nΣy² - (Σy)²)).
- →Interpretation of PMCC: values close to 1 or -1 indicate strong linear correlation, values close to 0 indicate weak or no linear correlation. The sign indicates the direction of the relationship.
- →Correlation vs causation: a strong correlation does not imply that one variable causes the other; there may be a lurking variable or the relationship may be coincidental.
- →Line of best fit: a straight line drawn through the centre of the data points on a scatter diagram, used to make predictions within the range of the data (interpolation) but not outside it (extrapolation).
Examiner Tips
- 💡Always interpret the PMCC in the context of the variables given. For example, say 'There is a strong positive correlation between hours studied and exam score' rather than just 'There is a strong positive correlation'.
- 💡When asked to comment on the suitability of a model, consider the strength of correlation, the presence of outliers, and whether extrapolation is being used. Mention that predictions outside the data range are unreliable.
- 💡Show all stages of the PMCC calculation, including the table of values and sums, to gain method marks even if the final answer is incorrect.
Common Mistakes
- Students often think that a strong correlation means one variable causes the other. Correction: Correlation does not imply causation; a third variable could be responsible for the relationship.
- Students may believe that a PMCC of 0 means there is no relationship at all. Correction: A PMCC of 0 indicates no linear relationship, but there could be a non-linear relationship.
- Students sometimes confuse the strength of correlation with the steepness of the line of best fit. Correction: Strength is about how closely points cluster around the line, not the gradient.
Revision Plan
- 1Step 1: Review the basics of scatter diagrams and correlation by watching video tutorials and practising plotting points from data sets.
- 2Step 2: Learn the PMCC formula and practise calculating it using summary statistics. Work through at least five examples with varying strengths of correlation.
- 3Step 3: Study the interpretation of PMCC values, including significance testing using critical values. Complete exercises that require you to compare calculated r to critical values.
- 4Step 4: Explore the difference between correlation and causation with real-world examples. Identify possible lurking variables in given scenarios.
- 5Step 5: Attempt past paper questions on B4, focusing on structured questions that require both calculation and interpretation. Review mark schemes to understand examiner expectations.
Exam Question Types
- 📋Calculation of PMCC: You will be given summary statistics or raw data and asked to calculate the product moment correlation coefficient. Show all working and interpret the result in context.
- 📋Interpretation of scatter diagrams: You may be asked to describe the correlation shown, identify outliers, or comment on the appropriateness of a line of best fit. Use precise language and refer to the variables.
- 📋Correlation vs causation: A question may present a strong correlation and ask you to explain why it does not prove causation. Suggest possible confounding variables and explain their effect.
- 📋Significance testing: You might be given a PMCC and a critical value table and asked to determine whether the correlation is statistically significant at a given level. State the null hypothesis and compare values.
Command Word Expectations (AQA)
Calculate
You must use the correct formula and show all steps. A numerical answer is required, often to 3 significant figures. Method marks are awarded for correct substitution and intermediate working.
Interpret
You must explain what the calculated value or diagram means in the context of the problem. For PMCC, state the strength and direction, and relate it to the variables. For scatter diagrams, describe the relationship and mention outliers if present.
Explain
You must provide a reasoned justification, often linking to statistical concepts such as correlation not implying causation. Use clear, logical sentences and refer to the context.
How Students Lose Marks (Examiner Pitfalls)
Pitfall: Students often describe correlation as 'positive' or 'negative' without specifying it is a relationship between two variables, or they confuse correlation with causation. They also frequently misinterpret the strength of correlation from a scatter diagram, calling a weak correlation 'strong' if points are loosely clustered.
❌ Weak Answer (Loses Marks):There is a positive correlation. As one variable increases, the other increases. This means one causes the other.
Example improved answer:There is a moderate positive correlation between the two variables. As the explanatory variable increases, the response variable tends to increase. However, correlation does not imply causation; a third variable may be responsible for the relationship.
Examiner Tip: Always refer to the variables by name, state the type (positive/negative) and strength (strong/moderate/weak), and explicitly mention that correlation does not prove causation unless you have experimental evidence.
Pitfall: When calculating the PMCC, students often forget to square the values or misapply the formula, leading to an incorrect value outside the range -1 to 1. They also fail to interpret the calculated value in context or compare it to a critical value for significance.
❌ Weak Answer (Loses Marks):r = 0.85, so there is a strong correlation. The result is significant.
Example improved answer:The PMCC is calculated as r = 0.85. This indicates a strong positive linear relationship between the variables. Comparing to the critical value at the 5% significance level for n = 20 (critical value = 0.444), since 0.85 > 0.444, the correlation is statistically significant.
Examiner Tip: Show all steps in the PMCC calculation, including the table of values and sums. Always state the critical value and compare your calculated r to it to determine significance. Interpret r in the context of the question.
Step-by-Step Worked Solutions
Question: A researcher collects data on the number of hours studied (x) and the exam score (y) for 10 students. The summary statistics are: Σx = 50, Σy = 600, Σx² = 300, Σy² = 38000, Σxy = 3200. Calculate the product moment correlation coefficient (PMCC) and interpret your result.
- 1.Step 1: Identify given facts: n = 10, Σx = 50, Σy = 600, Σx² = 300, Σy² = 38000, Σxy = 3200.
- 2.Step 2: Apply the PMCC formula: r = (nΣxy - ΣxΣy) / sqrt((nΣx² - (Σx)²)(nΣy² - (Σy)²)).
- 3.Step 3: Substitute values: r = (10*3200 - 50*600) / sqrt((10*300 - 50²)(10*38000 - 600²)) = (32000 - 30000) / sqrt((3000 - 2500)(380000 - 360000)) = 2000 / sqrt(500 * 20000) = 2000 / sqrt(10000000) = 2000 / 3162.27766 = 0.632.
- 4.Step 4: Interpret: r = 0.632 indicates a moderate positive correlation between hours studied and exam score. As hours studied increase, exam scores tend to increase.
Final Answer: PMCC r = 0.632, indicating a moderate positive correlation between hours studied and exam score.
Question: A scatter diagram shows a strong negative correlation between the age of a car (in years) and its value (in £1000s). The PMCC is calculated as r = -0.92. Explain what this means in context and state whether you would expect the value of a 10-year-old car to be accurately predicted from this model.
- 1.Step 1: Identify given facts: r = -0.92, variables are age and value, strong negative correlation.
- 2.Step 2: Interpret r: The value -0.92 indicates a strong negative linear relationship. As the age of the car increases, its value tends to decrease.
- 3.Step 3: Assess prediction accuracy: Since the correlation is strong, the model can predict value reasonably well for the range of ages in the data. However, for a 10-year-old car, if the data only covers ages 1-8, extrapolation may be unreliable.
- 4.Step 4: State conclusion: The model suggests a strong negative relationship, but predictions outside the data range should be made with caution.
Final Answer: r = -0.92 means a strong negative correlation: older cars tend to have lower value. Prediction for a 10-year-old car may be unreliable if data does not include such ages.
Active Recall Memory Test
What does a PMCC of -0.9 indicate about the relationship between two variables?
Key Fact: It indicates a strong negative linear correlation: as one variable increases, the other tends to decrease.
Why does correlation not imply causation?
Key Fact: Because a third (confounding) variable may be influencing both variables, or the relationship could be coincidental.
What is the range of possible values for the product moment correlation coefficient?
Key Fact: The PMCC ranges from -1 to 1 inclusive.
How do you determine if a correlation is statistically significant?
Key Fact: Compare the calculated PMCC to the critical value from a table for the given sample size and significance level. If the absolute value of r exceeds the critical value, the correlation is significant.
Frequently Asked Questions
What is the difference between correlation and causation in GCSE Statistics?
Correlation means there is a statistical relationship between two variables, while causation means one variable directly causes the other to change. In GCSE Statistics, you must always remember that correlation does not imply causation. For example, a correlation between umbrella sales and rainfall does not mean umbrellas cause rain; rather, rain causes both. You should look for possible confounding variables that could explain the relationship.
How do I calculate the product moment correlation coefficient (PMCC) in AQA GCSE Statistics?
To calculate the PMCC, use the formula r = (nΣxy - ΣxΣy) / sqrt((nΣx² - (Σx)²)(nΣy² - (Σy)²)). First, create a table with columns for x, y, x², y², and xy. Sum each column to find Σx, Σy, Σx², Σy², and Σxy. Then substitute these values into the formula. The result will be between -1 and 1. Always show your working to gain method marks.
What does a strong negative correlation mean in a scatter diagram?
A strong negative correlation means that as one variable increases, the other variable tends to decrease, and the points on the scatter diagram lie closely around a downward-sloping line of best fit. The PMCC would be close to -1. For example, there is a strong negative correlation between the age of a car and its value: older cars tend to be worth less.
How do I know if a correlation is statistically significant?
To determine significance, compare your calculated PMCC (ignoring the sign) to a critical value from a table. The critical value depends on the sample size (n) and the significance level (usually 5% or 1%). If the absolute value of your PMCC is greater than the critical value, the correlation is significant, meaning it is unlikely to have occurred by chance. If it is less, the correlation is not significant.
What are common mistakes students make with PMCC in exams?
Common mistakes include: not squaring the sums correctly, misapplying the formula leading to values outside -1 to 1, forgetting to interpret the result in context, and not comparing to a critical value for significance. Also, students often confuse correlation with causation. To avoid these, practise the calculation thoroughly, always interpret in the context of the variables, and remember that correlation does not prove causation.
How can I revise B4 effectively for AQA GCSE Statistics?
Start by reviewing scatter diagrams and correlation concepts. Then, practise calculating PMCC using past paper questions and check your answers against mark schemes. Focus on interpreting results and understanding significance testing. Use flashcards for key definitions and create mind maps linking correlation, causation, and confounding variables. Finally, attempt timed exam questions to build confidence and identify areas for improvement.