GCSE Statistics Higher Paper 2
An 80-mark, 90-minute paper in the structure of Edexcel 1ST0 (two papers per tier), drawn across the whole specification.
Questions
Question 1 [2 marks]
Sampling Methods
A youth club has 40 members. The club leader wants to choose a simple random sample of 8 members.
Describe a method she could use to choose this simple random sample.
Question 2 [3 marks]
Combined Events, Venn Diagrams and Tree Diagrams
In a survey of 40 people, 14 like hiking only, 11 like cycling only, and 6 like both hiking and cycling.
Find the number of people who like neither hiking nor cycling.
Question 3 [3 marks]
Time Series, Index Numbers and Standardised Scores
A cafe's weekly ice cream sales (in units) for 4 consecutive weeks are: Week 1: 60, Week 2: 74, Week 3: 68, Week 4: 82.
Calculate the 3-week moving average covering weeks 2 to 4.
Question 4 [3 marks]
Spearman's Rank and Regression
Two judges give the same ranking to five dancers: 1, 2, 3, 4, 5.
Use rs = 1 - (6 x sum(d^2)) / (n(n^2 - 1)), where d is the difference between the two ranks for each item and n is the number of items.
Work out Spearman's rank correlation coefficient and explain what it shows.
Question 5 [3 marks]
Quality Assurance and Control Charts
A control chart has been used to plot the means of 200 samples from a process that is working correctly.
Warning lines are set so that about 1 in 20 sample means fall outside them. About how many of the sample means would you expect to be outside the warning lines? What would you expect for the action lines?
Question 6 [3 marks]
Population Pyramids and Rates of Change
Over 10 years the crude birth rate of a country fell from 18 per 1000 to 14 per 1000. The population stayed at 6.2 million.
Work out how many fewer births there were in the later year.
Question 7 [3 marks]
Geometric Mean, Standard Deviation and Outliers
The masses of eggs have a mean of 50 g and a standard deviation of 4 g. A value is an outlier if it is more than 3 standard deviations from the mean.
Work out the limits beyond which a value is an outlier. Is an egg of mass 63 g an outlier?
Question 8 [3 marks]
Percentiles and Interpercentile Ranges
In a mock exam the 75th percentile mark is 68 and the 90th percentile mark is 79. Ben scored 71.
Between which two percentiles does Ben's mark lie? Estimate the range of percentages of students who scored higher than Ben.
Question 9 [3 marks]
Averages from Grouped Data
A student estimating the mean from a grouped frequency table uses 20, the lower boundary, instead of the midpoint 25 for the class 20 <= x < 30. This class has a positive frequency.
Explain how the mistake changes the estimated mean compared with using 25. Explain why the midpoint is a better representative of the whole class than the lower boundary.
Question 10 [3 marks]
Tables and Frequency Distributions
A researcher is grouping continuous data on the diameter (in mm) of 80 ball bearings. She proposes the class intervals 5-10, 10-15, 15-20, and so on, each of width 5.
Rewrite the first two class intervals using inequality notation so that a diameter of exactly 10mm is placed unambiguously in only one class, and state which of your two intervals it belongs to.
Question 11 [3 marks]
Index Numbers
Between two years the price of a loaf of bread rose from GBP 1.40 to GBP 1.47 and the price of a pair of shoes rose from GBP 30.00 to GBP 33.60.
Work out the price index for each item (the first year is the base year) and say which item has increased more, in percentage terms.
Question 12 [4 marks]
Histograms and Frequency Density
A histogram for 108 students' reading times has two classes of equal width 6 minutes. The frequency density of the second class is exactly double the frequency density of the first class.
Let the first class have frequency density d. Form an expression for each class's frequency in terms of d, form an equation using the fact the two frequencies sum to 108, and solve to find d. Find both class frequencies and check their total.
Question 13 [4 marks]
Probability
A two-way table for 60 students shows: 25 boys and 35 girls in total; 18 of the boys and 21 of the girls own a games console.
A student is picked at random. Find the probability that the student is a girl who owns a games console, and the probability that the student owns a games console, given that they are a boy.
Question 14 [4 marks]
Averages and Range
Six numbers have a mean of 8. A seventh number is added and the new mean of all seven numbers is 9.
Find the value of the seventh number.
Question 15 [3 marks]
Binomial and Normal Distributions
The times taken to complete a task are normally distributed with mean 40 minutes. About 95% of the times are between 32 minutes and 48 minutes.
Work out the standard deviation.
Question 16 [4 marks]
Risk, Expected Frequency and Bias
In a study of cycling accidents, 18 in every 1000 cyclists who did not wear a helmet were injured, and 6 in every 1000 cyclists who wore a helmet were injured.
Work out the relative risk of injury for cyclists without a helmet compared with cyclists with a helmet. Explain what it means, and work out the difference in the expected number of injuries in a group of 5000 cyclists of each type.
Question 17 [5 marks]
Binomial and Normal Distributions
The marks of 400 students are normally distributed with mean 80 and standard deviation 10. About 68% of marks are within 1 standard deviation of the mean and about 95% are within 2 standard deviations.
Work out the number of students expected to score between 70 and 90, and the number expected to score more than 100.
Question 18 [6 marks]
Statistical Diagrams
A researcher has grouped continuous data on the heights of 200 plants and wants to choose between drawing a histogram and drawing a box plot to present the shape of the distribution.
Explain one advantage and one disadvantage of using a box plot rather than a histogram, and describe what information a histogram shows that a box plot does not. Develop both the advantage and the disadvantage, and give two distinct examples of distribution detail a histogram can reveal.
Question 19 [6 marks]
Collecting Data
A pilot survey of 40 students found that 25% owned a smartwatch. A researcher uses this proportion to predict the number of owners in the whole year group of 220 students.
A full census of all 220 students later finds that 68 of them own a smartwatch.
Calculate the pilot's predicted number of owners, calculate the census's actual percentage of owners (1 decimal place), and explain what this comparison suggests about how reliable the small pilot's estimate was.
Question 20 [6 marks]
Correlation and Regression
A line of best fit passes through the points (5, 42) and (15, 82) on a scatter graph of hours studied (x) against test score (y).
Find the equation of this line of best fit in the form y = mx + c, and use it to estimate the test score for a student who studies 10 hours.
Question 21 [6 marks]
Quartiles, Interquartile Range and Standard Deviation
A farmer records the mass (grams) of 50 apples from Orchard A: mean = 180 g, standard deviation = 18 g. He also records the mass (grams) of 50 pears from Orchard B: mean = 220 g, standard deviation = 20 g.
The coefficient of variation (CV) is defined as CV = (standard deviation / mean) x 100%.
Calculate the CV for each orchard's fruit, and state which orchard's fruit has more consistent mass RELATIVE to its own average size.
Give both coefficients of variation to 2 decimal places. Explain why CV is more useful here than comparing the raw standard deviations.
Model solutions
| Question 1[2 marks] | |
|---|---|
| Answer or working | Marks |
| uniquely labelling all 40 members and uniformly selecting 8 distinct labels without replacement | M1 |
| selecting the matching members so every subset of 8 members has the same probability | A1 |
| Final answer: Label the members 1 to 40. Generate 8 distinct numbers uniformly from 1 to 40, ignoring repeats, and select their members. Every size-8 subset then has the same probability of selection. | |
| Question 2[3 marks] | |
|---|---|
| Answer or working | Marks |
| the number who like at least one activity: 14 + 11 + 6 | M1 |
| 31 people | A1 |
| neither = 40 - 31 = 9 people | B1 |
| Final answer: 9 people like neither. | |
| Question 3[3 marks] | |
|---|---|
| Answer or working | Marks |
| 74 + 68 + 82 | M1 |
| dividing by 3 | M1 |
| 74.7 (1 dp) | A1 |
| Question 4[3 marks] | |
|---|---|
| Answer or working | Marks |
| every difference d = 0, so sum(d^2) = 0 | M1 |
| rs = 1 - 0 = 1 | A1 |
| perfect agreement between the judges | B1 |
| Final answer: rs = 1: perfect agreement | |
| Question 5[3 marks] | |
|---|---|
| Answer or working | Marks |
| 200 / 20 (1 in 20 outside the warning lines) | M1 |
| about 10 sample means | A1 |
| almost none outside the action lines | B1 |
| Final answer: About 10 outside the warning lines; almost none outside the action lines. | |
| Question 6[3 marks] | |
|---|---|
| Answer or working | Marks |
| (18 - 14) / 1000 x 6200000 | M1 |
| 24800 | A1 |
| 111600 births compared with 86800 births, a difference of 24800 | B1 |
| Final answer: 24800 fewer births | |
| Question 7[3 marks] | |
|---|---|
| Answer or working | Marks |
| 50 +/- 3 x 4 | M1 |
| limits 38 g and 62 g | A1 |
| 63 g is an outlier, since it is greater than 62 g | B1 |
| Final answer: Limits 38 g and 62 g; 63 g is an outlier. | |
| Question 8[3 marks] | |
|---|---|
| Answer or working | Marks |
| Ben's mark lies between the 75th and the 90th percentiles | B1 |
| 100 - 90 = 10% scored higher than the 90th percentile mark | B1 |
| between 10% and 25% of students scored higher than Ben | B1 |
| Final answer: Between the 75th and 90th percentiles; between 10% and 25% of students scored higher than Ben. | |
| Question 9[3 marks] | |
|---|---|
| Answer or working | Marks |
| explaining that 20 is 5 below 25, decreasing the weighted total and hence the midpoint mean estimate | B1 |
| explaining that the midpoint is central and balances the possible endpoint errors | B1 |
| explaining that the lower boundary represents an extreme rather than the whole interval | B1 |
| Final answer: The weighted total decreases by 5 times the class frequency, so the estimate is lower than the correct midpoint estimate. The midpoint is central and balances distances to the endpoints, whereas the lower boundary is the smallest possible value. Neither choice guarantees the true class mean. | |
| Question 10[3 marks] | |
|---|---|
| Answer or working | Marks |
| the first interval, e.g. 5 <= d < 10 | B1 |
| the second interval, e.g. 10 <= d < 15 | B1 |
| stating a diameter of exactly 10mm belongs to the second interval (10 <= d < 15) | B1 |
| Final answer: 5 <= d < 10, 10 <= d < 15. A diameter of exactly 10mm belongs to the second interval. | |
| Question 11[3 marks] | |
|---|---|
| Answer or working | Marks |
| bread: 1.47 / 1.40 x 100 = 105 | B1 |
| shoes: 33.60 / 30.00 x 100 = 112 | B1 |
| shoes have increased more (12% compared with 5%) | B1 |
| Final answer: Bread 105, shoes 112; shoes have increased more. | |
| Question 12[4 marks] | |
|---|---|
| Answer or working | Marks |
| the two frequencies: 6d and 6 x 2d = 12d | M1 |
| the equation: 6d + 12d = 108 | M1 |
| d = 6 | A1 |
| the frequencies: 36 and 72 (checking 36 + 72 = 108) | B1 |
| Final answer: d = 6. Frequencies: first class = 36 students, second class = 72 students (36 + 72 = 108). | |
| Question 13[4 marks] | |
|---|---|
| Answer or working | Marks |
| the number of girls who own a console: 21 out of 60 total | M1 |
| P(girl and console) = 21/60 = 7/20 (= 0.35) | A1 |
| recognising the conditional probability restricts the sample space to boys only: 18 out of 25 boys | M1 |
| P(console | boy) = 18/25 | A1 |
| Final answer: P(girl and owns console) = 7/20 (0.35). P(owns console | boy) = 18/25 (0.72). | |
| Question 14[4 marks] | |
|---|---|
| Answer or working | Marks |
| the sum of the original six numbers: 6 x 8 = 48 | M1 |
| the sum of all seven numbers: 7 x 9 = 63 | M1 |
| subtracting: 63 - 48 | M1 |
| the seventh number = 15 | A1 |
| Final answer: 15 | |
| Question 15[3 marks] | |
|---|---|
| Answer or working | Marks |
| 95% lie within 2 standard deviations of the mean | B1 |
| 48 - 40 = 8 = 2 standard deviations | M1 |
| 4 minutes | A1 |
| Question 16[4 marks] | |
|---|---|
| Answer or working | Marks |
| 18 / 6 | M1 |
| relative risk = 3 | A1 |
| cyclists without a helmet are 3 times as likely to be injured | B1 |
| 5000 x (18 - 6) / 1000 = 60 more injuries | B1 |
| Final answer: Relative risk = 3: 3 times as likely; 60 more injuries. | |
| Question 17[5 marks] | |
|---|---|
| Answer or working | Marks |
| 70 to 90 is within 1 standard deviation of the mean | M1 |
| 0.68 x 400 | M1 |
| 272 students | A1 |
| 100 is 2 standard deviations above the mean, so 2.5% score above this | M1 |
| 0.025 x 400 = 10 students | A1 |
| Final answer: 272 students between 70 and 90; 10 students above 100 | |
| Question 18[6 marks] | |
|---|---|
| Answer or working | Marks |
| an advantage of the five-number summary, e.g. explicit median and quartiles giving centre and spread | B1 |
| developing that advantage, e.g. compact plots on a common scale allow direct comparison of centres and IQRs | B1 |
| the limitation that observations within each quartile are not shown | B1 |
| explaining the consequence: different within-quartile arrangements can have identical five-number summaries | B1 |
| histogram detail about class frequencies or densities, e.g. locating modal classes or multiple peaks | B1 |
| a second distinct histogram detail, e.g. empty class intervals reveal gaps hidden by the quartile summary; do not credit the same peaks example twice | B1 |
| Final answer: Box plot advantage: shows median, quartiles, range and outliers clearly, and is easy to compare across groups. Disadvantage: hides the detailed shape within each quarter. A histogram shows frequency density across intervals, revealing peaks, gaps or skew a box plot cannot. | |
| Question 19[6 marks] | |
|---|---|
| Answer or working | Marks |
| the predicted number: 0.25 x 220 | M1 |
| 55 students predicted | A1 |
| the actual percentage: 68/220 x 100 | M1 |
| 30.9% (1 dp) | A1 |
| comparing the two: the pilot's 25% underestimated the true rate of about 30.9% | B1 |
| a valid explanation, e.g. a pilot of only 40 students is a small sample and so is subject to considerable sampling variation, meaning its estimate can differ noticeably from the true population value | B1 |
| Final answer: Pilot predicted 55 owners (25%). The census shows the true rate is 30.9% (1 dp), so the small pilot underestimated the true proportion, illustrating that a small pilot sample can differ substantially from the population value due to sampling variation. | |
| Question 20[6 marks] | |
|---|---|
| Answer or working | Marks |
| the gradient: (82 - 42)/(15 - 5) | M1 |
| gradient m = 4 | A1 |
| substituting a point to find c: 42 = 4 x 5 + c | M1 |
| c = 22 | A1 |
| substituting x = 10 into y = 4x + 22 | M1 |
| y = 62 | A1 |
| Final answer: y = 4x + 22. Estimated test score for 10 hours studied = 62. | |
| Question 21[6 marks] | |
|---|---|
| Answer or working | Marks |
| Orchard A's CV: 18/180 x 100 | M1 |
| CV(A) = 10% | A1 |
| Orchard B's CV: 20/220 x 100 | M1 |
| CV(B) = 9.09% (2 dp) | A1 |
| concluding Orchard B's pears are relatively more consistent, since 9.09% is smaller than 10% | B1 |
| explaining why CV rather than the raw standard deviation is needed, e.g. the two fruits have different mean masses, so comparing standard deviations alone would not account for their different scales | B1 |
| Final answer: CV(Orchard A) = 10%, CV(Orchard B) = 9.09% (2 dp). Orchard B's pears have more consistent mass relative to their own average, since they have the smaller coefficient of variation. | |