D5 — AQA GCSE Statistics
Test yourself on D5 with AQA GCSE practice questions.
Start free
7 days Premium · Then free forever · No card, no charge
Your focus
- Determine line of best fit by eye.
D5 exam tips
Quick Revision Summary (Key Takeaway)
D5 in AQA GCSE Statistics covers the analysis and interpretation of bivariate data using scatter diagrams, correlation, and linear regression. Students must calculate and interpret the product moment correlation coefficient (PMCC), find the equation of the regression line, and use it to make predictions while understanding the limitations of extrapolation.
Topic Overview
D5 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 determine the equation of the regression line. This topic is essential for understanding how to quantify and model relationships in real-world data, from scientific studies to business trends.
This topic builds on your knowledge of descriptive statistics and probability, extending into inferential techniques. It is a cornerstone of statistical analysis and frequently appears in exams, often combined with other topics like data collection and representation. Mastering D5 will enable you to make informed predictions and critically evaluate claims about correlations in the media and research.
Key Concepts
- →Scatter diagrams visually display the relationship between two quantitative variables, showing the direction, form, and strength of any association.
- →The product moment correlation coefficient (PMCC), denoted r, measures the strength and direction of a linear relationship, ranging from -1 to +1.
- →The regression line of y on x is the line of best fit that minimises the sum of squared residuals, used to predict y from x.
- →Correlation does not imply causation; a strong correlation may be due to a lurking variable or coincidence.
- →Extrapolation, predicting outside the range of observed data, is unreliable and should be avoided or treated with caution.
Examiner Tips
- 💡Always interpret the PMCC in context, stating both the strength (weak/moderate/strong) and direction (positive/negative) of the correlation.
- 💡When making predictions, check if the value is within the data range. If not, comment on the reliability of the extrapolation.
- 💡Show all steps in calculations, especially for the PMCC and regression line, as method marks are often awarded for correct substitution into formulas.
Common Mistakes
- Students often think a high correlation means one variable causes the other. Correction: Correlation only indicates a relationship, not causation; controlled experiments are needed to establish cause.
- Students may believe that a PMCC of 0 means no relationship at all. Correction: r = 0 means no linear relationship, but there could be a non-linear relationship.
- Students sometimes use the regression line to predict x from y without calculating the appropriate regression line of x on y. Correction: The regression line of y on x is not symmetric; to predict x, you need the regression line of x on y.
Revision Plan
- 1Week 1: Review the concepts of scatter diagrams and correlation. Practice calculating PMCC using the formula and interpreting values. Complete past paper questions on PMCC.
- 2Week 1: Learn how to calculate the regression line equation and use it for predictions. Focus on understanding the meaning of the gradient and intercept in context.
- 3Week 2: Work through mixed exam-style questions that combine PMCC and regression, including those requiring interpretation and critique of statistical claims.
- 4Week 2: Create a summary sheet of key formulas and common pitfalls. Test yourself with active recall prompts and practice explaining concepts aloud.
- 5Ongoing: Use online resources or textbooks to attempt challenging questions, and review mistakes to avoid repeating them.
Exam Question Types
- 📋Calculation of PMCC from summary statistics: You will be given Σx, Σy, Σx², Σy², Σxy, and n, and asked to calculate r. Advice: Memorise the formula and show all steps clearly.
- 📋Interpretation of PMCC and correlation: You may be asked to describe the relationship between two variables based on a scatter diagram or a given r value. Advice: Use precise language: 'moderate positive correlation' etc., and avoid causal language.
- 📋Finding and using the regression line: You will calculate the equation y = a + bx and use it to predict a value. Advice: Check the range of data and comment on extrapolation if necessary.
- 📋Critiquing a statistical claim: You may be presented with a statement about correlation and asked to evaluate it. Advice: Consider sample size, outliers, causation, and the context of the data.
Command Word Expectations (AQA)
Calculate
In AQA GCSE Statistics, 'Calculate' requires you to work out a numerical value, showing all necessary steps. For PMCC, you must substitute into the formula and round appropriately (usually to 3 significant figures).
Interpret
You must explain what a statistical measure means in the context of the problem. For PMCC, state the strength and direction of the correlation and relate it to the variables. For regression, explain the meaning of the gradient and intercept.
Evaluate
You must make a judgement about the validity or reliability of a statistical analysis, considering factors like sample size, outliers, extrapolation, and causation. Provide a balanced argument and a clear conclusion.
How Students Lose Marks (Examiner Pitfalls)
Pitfall: Students often confuse correlation with causation, stating that one variable causes the other to change. They also frequently misinterpret the strength of correlation, calling a PMCC of 0.6 'strong' when it is only moderate.
❌ Weak Answer (Loses Marks):There is a strong positive correlation between hours of revision and test score, so revising more causes higher scores.
Example improved answer:There is a moderate positive correlation (r = 0.62) between hours of revision and test score. This means that as revision time increases, test scores tend to increase, but causation cannot be assumed as other factors may be involved.
Examiner Tip: Always use the correct descriptor for the PMCC value: 0.0-0.3 weak, 0.3-0.7 moderate, 0.7-1.0 strong (positive or negative). Never say 'causes' unless the question explicitly states an experiment was controlled.
Pitfall: When using the regression line to predict, students often extrapolate beyond the range of the data without commenting on its unreliability. They also forget to use the correct variable for prediction or misinterpret the gradient and intercept.
❌ Weak Answer (Loses Marks):Using the line y = 2.5x + 10, if x = 100, then y = 260. So the score will be 260.
Example improved answer:The regression line is y = 2.5x + 10, where x is hours of revision (range 0-20 hours). For x = 100 hours, the predicted score is y = 2.5(100) + 10 = 260. However, this is an extrapolation far outside the observed range, so the prediction is unreliable and may not be valid.
Examiner Tip: Always check if the prediction is within the data range. If not, state that it is an extrapolation and therefore unreliable. Also, ensure you substitute into the correct variable and interpret the result in context.
Step-by-Step Worked Solutions
Question: A scatter diagram shows the relationship between the number of hours spent studying (x) and the marks achieved (y) for 10 students. The summary statistics are: Σx = 50, Σy = 600, Σx² = 300, Σy² = 40000, Σxy = 3500, n = 10. Calculate the product moment correlation coefficient (PMCC) and interpret your result.
- 1.Step 1: Calculate the means: x̄ = Σx / n = 50 / 10 = 5, ȳ = Σy / n = 600 / 10 = 60.
- 2.Step 2: Calculate Sxx = Σx² - n(x̄)² = 300 - 10(5)² = 300 - 250 = 50.
- 3.Step 3: Calculate Syy = Σy² - n(ȳ)² = 40000 - 10(60)² = 40000 - 36000 = 4000.
- 4.Step 4: Calculate Sxy = Σxy - n(x̄)(ȳ) = 3500 - 10(5)(60) = 3500 - 3000 = 500.
- 5.Step 5: Use the formula r = Sxy / √(Sxx * Syy) = 500 / √(50 * 4000) = 500 / √200000 = 500 / 447.21 = 1.118. Wait, this is incorrect because r cannot exceed 1. Recalculate: √(50*4000) = √200000 = 447.21, 500/447.21 = 1.118, which is impossible. Check calculations: Sxx = 300 - 10*25 = 50, Syy = 40000 - 10*3600 = 40000 - 36000 = 4000, Sxy = 3500 - 10*5*60 = 3500 - 3000 = 500. r = 500 / sqrt(50*4000) = 500 / sqrt(200000) = 500 / 447.2136 = 1.118. This indicates an error in the given summary statistics because r must be between -1 and 1. However, for the purpose of this example, assume the correct value is r = 0.95 (if data were adjusted). But since the question is hypothetical, we will proceed with the calculated value and note the impossibility. Actually, let's correct the summary statistics to make r valid: Suppose Σxy = 3200, then Sxy = 3200 - 3000 = 200, r = 200 / 447.21 = 0.447. But to keep it simple, we will use the original numbers and state that the calculated r is invalid, so we will use a corrected example. Let's use a different set: Σx = 50, Σy = 600, Σx² = 300, Σy² = 40000, Σxy = 3200, n = 10. Then Sxy = 3200 - 3000 = 200, r = 200 / 447.21 = 0.447. So r = 0.447, indicating moderate positive correlation.
- 6.Step 6: Interpret: r = 0.447, which is a moderate positive correlation. This means that as hours of studying increase, marks tend to increase moderately.
Final Answer: PMCC r = 0.447 (moderate positive correlation). There is a moderate positive linear relationship between hours studied and marks achieved.
Question: For the data in the previous question (with corrected Σxy = 3200), find the equation of the regression line of y on x in the form y = a + bx. Use it to predict the mark for a student who studies for 8 hours.
- 1.Step 1: Calculate the gradient b = Sxy / Sxx = 200 / 50 = 4.
- 2.Step 2: Calculate the intercept a = ȳ - b * x̄ = 60 - 4 * 5 = 60 - 20 = 40.
- 3.Step 3: Write the regression line: y = 40 + 4x.
- 4.Step 4: Substitute x = 8: y = 40 + 4(8) = 40 + 32 = 72.
- 5.Step 5: Interpret: The predicted mark for 8 hours of study is 72.
Final Answer: Regression line: y = 40 + 4x. Predicted mark for 8 hours: 72.
Active Recall Memory Test
What does a PMCC of -0.9 indicate about the relationship between two variables?
Key Fact: A PMCC of -0.9 indicates a strong negative linear correlation, meaning that as one variable increases, the other tends to decrease.
State the formula for the product moment correlation coefficient.
Key Fact: r = Sxy / √(Sxx * Syy), where Sxx = Σx² - n(x̄)², Syy = Σy² - n(ȳ)², and Sxy = Σxy - n(x̄)(ȳ).
Why is extrapolation unreliable when using a regression line?
Key Fact: Extrapolation is unreliable because the relationship between variables may change outside the observed range, and the regression line is only validated for the data range.
What is the difference between correlation and causation?
Key Fact: Correlation means two variables are related, but causation means one variable directly affects the other. Correlation does not imply causation.
Frequently Asked Questions
What is the difference between PMCC and Spearman's rank correlation coefficient?
PMCC measures the strength of a linear relationship between two continuous variables, while Spearman's rank correlation coefficient measures the strength of a monotonic relationship between ranked variables. PMCC is sensitive to outliers and requires interval or ratio data, whereas Spearman's is used for ordinal data or when the relationship is non-linear but monotonic.
How do I know if a correlation is strong or weak?
The strength of correlation is determined by the absolute value of the PMCC. Generally, values between 0.0 and 0.3 are weak, 0.3 to 0.7 are moderate, and 0.7 to 1.0 are strong. However, always consider the context and sample size, as these can affect the interpretation.
Can I use the regression line to predict x from y?
No, the regression line of y on x is used to predict y from x. To predict x from y, you need the regression line of x on y, which is calculated differently. Using the wrong line will give inaccurate predictions.
What does it mean if the PMCC is 0?
A PMCC of 0 means there is no linear relationship between the variables. However, there could still be a non-linear relationship, so it is important to examine the scatter diagram.
How do I calculate the regression line equation?
The regression line of y on x is given by y = a + bx, where b = Sxy / Sxx and a = ȳ - b x̄. You need to calculate the means and the summary statistics Sxx and Sxy from the data.
Why is it important to comment on the range of data when making predictions?
The regression line is only valid within the range of the observed data. Predicting outside this range (extrapolation) is unreliable because the relationship may not hold. Always check if the prediction is within the data range and comment on its reliability.