Level 2 Probability practice paper 2026
Question One
The weights of kiwifruit from a Bay of Plenty orchard are normally distributed with mean 105 g and standard deviation 12 g. Find the z-value for a kiwifruit weighing 125 g, and the probability that a randomly chosen kiwifruit weighs more than 125 g.
Hint
z = (x − μ) ÷ σ. Sketch the curve and shade to the right of 125, then use normal cdf from 125 to a very large number.
Answer
z = 1.667, P(X > 125) = 0.0478
Step by step
- Sketch a bell curve centred on 105, with 125 marked to the right of the middle. Shade the right-hand tail.
- The shaded tail is small, so expect a small answer.
- z = (x − μ) ÷ σ = (125 − 105) ÷ 12 = 20 ÷ 12 = 1.667. It is positive because 125 is above the mean.
- On the calculator use normal cdf with lower 125, upper 1E99, μ = 105 and σ = 12.
- The calculator gives 0.04779, which is 0.0478 to four decimal places.
- Check against the sketch: a thin right tail, and the answer is under 0.5. It matches.
Full working
- z = (125 − 105) ÷ 12 = 1.667125 is above the mean
- lower 125, upper 1E99everything to the right
- P(X > 125) = 0.0478
The weights of kiwifruit from a Bay of Plenty orchard are normally distributed with mean 105 g and standard deviation 12 g.
(i) Kiwifruit that weigh less than 90 g are sent for juice. How many of the fruit in a bin of 2400 would you expect to be sent for juice?
(ii) The heaviest 15% of kiwifruit are packed as "jumbo". Find the minimum weight of a jumbo kiwifruit.
Hint
(i) Find P(X < 90) and multiply by 2400. (ii) The area to the LEFT of the cut-off is 0.85. Use inverse normal, or find z and use x = μ + zσ.
Answer
(i) P(X < 90) = 0.1056, so about 254 fruit. (ii) z = 1.036, minimum weight 117.44 g.
Step by step
- (i) z = (90 − 105) ÷ 12 = −1.25. Shade the left tail below 90.
- Normal cdf with lower −1E99, upper 90, μ = 105, σ = 12 gives 0.1056.
- Expected number = probability × number of fruit = 0.1056 × 2400 = 253.6.
- You cannot have part of a kiwifruit, so about 254 fruit.
- (ii) Jumbo fruit are the top 15%, so 85% of fruit are lighter than the cut-off.
- Inverse normal with area 0.85 (to the left) gives z = 1.036.
- x = μ + zσ = 105 + 1.036 × 12 = 117.44 g.
- Classic mistake: putting 0.15 into inverse normal. Most calculators want the area to the left, which is 0.85.
- Check: 117.44 is above the mean, as the heaviest fruit must be.
Full working
- z = (90 − 105) ÷ 12 = −1.25
- P(X < 90) = 0.1056normal cdf, left tail
- 0.1056 × 2400 = 253.6, so about 254 fruitexpected number
- area to the left = 0.85, z = 1.036inverse normal
- x = 105 + 1.036 × 12x = μ + zσ
- 254 fruit; jumbo from 117.44 g
The weights of kiwifruit from a Bay of Plenty orchard are normally distributed with mean 105 g and standard deviation 12 g. The grower wants no more than 3% of fruit to weigh less than 90 g. Changing how the vines are thinned can raise the mean weight without changing the standard deviation. Find the smallest mean weight that would meet the grower's target.
Hint
Find the z-value with 3% of the area to its left. It will be negative. Then put it into z = (x − μ) ÷ σ with x = 90 and σ = 12, and solve for μ.
Answer
z = −1.881, so μ = 90 + 1.881 × 12 = 112.57 g
Step by step
- We want P(X < 90) = 0.03, so 90 g must cut off the bottom 3% of the curve.
- Inverse normal with area 0.03 (on the standard normal) gives z = −1.881. It is negative because 90 g is below the mean.
- Use z = (x − μ) ÷ σ: −1.881 = (90 − μ) ÷ 12.
- Multiply both sides by 12: −22.57 = 90 − μ.
- So μ = 90 + 22.57 = 112.57 g.
- Any mean of at least 112.57 g keeps the proportion under 90 g at 3% or less.
- Check: normal cdf from −1E99 to 90 with μ = 112.57 and σ = 12 gives 0.0300.
Full working
- P(X < 90) = 0.03target
- z = −1.881inverse normal, area 0.03
- −1.881 = (90 − μ) ÷ 12z = (x − μ) ÷ σ
- μ = 90 + 1.881 × 12
- μ = 112.57 g
Samples of 300 kiwifruit were weighed from two orchards, A and B. The table gives the summary statistics. A histogram of Orchard A's weights is roughly bell-shaped with one peak at 100 to 110 g. A histogram of Orchard B's weights has one peak at 90 to 100 g, with fruit bunched to the left and a long tail stretching out to the right.
(i) Compare the distributions of kiwifruit weights from the two orchards. Discuss shape, centre and spread, with numbers, in context.
(ii) A packer wants to model Orchard B's weights with a normal distribution with mean 102.6 g and standard deviation 15.8 g. Use a calculation to explain whether this model is suitable.
| Statistic (g) | Orchard A | Orchard B |
|---|---|---|
| Minimum | 72 | 80 |
| Lower quartile | 97 | 92 |
| Median | 105 | 98 |
| Upper quartile | 113 | 110 |
| Maximum | 139 | 162 |
| Mean | 105.2 | 102.6 |
| Standard deviation | 12.1 | 15.8 |
Hint
For each of shape, centre and spread, say what each orchard does AND compare them, quoting numbers. For (ii), work out how many of the 300 fruit the normal model says should weigh under 80 g, and compare with the minimum.
Answer
(i) A is symmetric and bell-shaped (mean 105.2 close to median 105); B is skewed to the right (mean 102.6 above median 98). B's fruit tend to be lighter (median 98 g against 105 g). B is more spread out (IQR 18 g against 16 g, standard deviation 15.8 g against 12.1 g, range 82 g against 67 g). (ii) The model gives P(X < 80) = 0.0763, about 23 of 300 fruit, but none weighed under 80 g; the data are skewed, so the model is not suitable.
Step by step
- (i) Shape. Orchard A is roughly symmetric and bell-shaped: the mean (105.2 g) and the median (105 g) are almost equal.
- Orchard B is skewed to the right: its peak is at 90 to 100 g, and a few very heavy fruit (up to 162 g) pull the mean (102.6 g) above the median (98 g).
- Both are unimodal (one peak), but B's peak (90 to 100 g) is lower than A's (100 to 110 g).
- Centre. Orchard B's kiwifruit tend to be lighter than Orchard A's: B's median is 98 g, 7 g below A's 105 g.
- Use the median for B, not the mean, because the long right tail inflates B's mean.
- Spread. B's weights vary more. B's IQR is 110 − 92 = 18 g against A's 113 − 97 = 16 g, so the middle 50% of B's fruit is a bit more spread out.
- B's standard deviation (15.8 g) and range (162 − 80 = 82 g) are also bigger than A's (12.1 g and 139 − 72 = 67 g). Much of B's extra spread comes from the heavy fruit in its tail.
- (ii) Under the model, z = (80 − 102.6) ÷ 15.8 = −1.430 and P(X < 80) = 0.0763.
- So the model expects 0.0763 × 300 = 22.9, about 23 fruit lighter than 80 g. But the minimum is 80 g, so no fruit in the sample weighed less than 80 g.
- A normal curve is symmetric, but B's data are skewed to the right (mean above median, long right tail, short left tail). So a normal model is not suitable for Orchard B.
Full working
- A: mean 105.2 ≈ median 105, so symmetricshape
- B: mean 102.6 > median 98, long right tail, so right-skewedshape
- median B 98 g < median A 105 gcentre
- IQR B 18 g > IQR A 16 g; SD 15.8 g > 12.1 gspread
- model: P(X < 80) = 0.0763, 0.0763 × 300 ≈ 23 fruitz = −1.430
- Sample has 0 fruit under 80 g, so the normal model does not fit B
At a Te Puke packhouse, 10% of a grower's kiwifruit weigh less than 90 g and 5% weigh more than 130 g. Assume the weights are normally distributed.
(i) Find the mean and standard deviation of the weights.
(ii) The packhouse expects 20 000 of this grower's fruit this season. How many would you expect to weigh between 100 g and 120 g?
(iii) In a random sample of 400 of the grower's fruit, 34 weighed less than 90 g. Compare this with the model and comment.
Hint
Find the z-value for each percentage. Write x = μ + zσ twice, once for 90 g and once for 130 g, and solve the two equations together.
Answer
(i) μ = 107.52 g, σ = 13.67 g. (ii) P(100 < X < 120) = 0.5283, so about 10 565 fruit. (iii) 34400 = 0.085 in the sample against 0.10 in the model (40 fruit expected); this is close and fits natural sampling variation, so it does not cast doubt on the model.
Step by step
- (i) 10% are below 90 g: inverse normal with area 0.10 gives z = −1.282.
- 5% are above 130 g, so 95% are below it: inverse normal with area 0.95 gives z = 1.645.
- Using x = μ + zσ: 90 = μ − 1.282σ and 130 = μ + 1.645σ.
- Take the first equation from the second: 40 = 2.926σ, so σ = 40 ÷ 2.926 = 13.67 g.
- Then μ = 90 + 1.282 × 13.67 = 107.52 g.
- Check: normal cdf below 90 with μ = 107.52, σ = 13.67 gives 0.100, and above 130 gives 0.050.
- (ii) Normal cdf from 100 to 120 with μ = 107.52 and σ = 13.67 gives 0.5283.
- Expected number = 0.5283 × 20 000 = 10 565 fruit (answers from about 10 540 to 10 570 are fine, depending on rounding).
- (iii) In the sample, 34 ÷ 400 = 0.085, or 8.5%, weighed under 90 g. The model says 10%, which is 0.10 × 400 = 40 fruit.
- The experimental proportion (0.085) is a little below the model's (0.10), but a random sample will not match the model exactly. A difference of 6 fruit in 400 is small.
- So the sample is consistent with the model. A bigger sample would give a more reliable check.
Full working
- z = −1.282 (area 0.10) and z = 1.645 (area 0.95)inverse normal
- 90 = μ − 1.282σ, 130 = μ + 1.645σx = μ + zσ
- 2.926σ = 40, so σ = 13.67 gsubtract
- μ = 90 + 1.282 × 13.67 = 107.52 g
- P(100 < X < 120) = 0.5283, × 20 000 = 10 565normal cdf
- sample 34400 = 0.085 against model 0.10 (40 fruit)experimental vs model
- μ = 107.52 g, σ = 13.67 g; about 10 565 fruit; the sample fits the model
Question Two
A surf school at Mount Maunganui runs one beginners' lesson each day. On 70% of days the swell is suitable. When the swell is suitable, the lesson goes ahead 90% of the time (otherwise strong wind cancels it). When the swell is not suitable, the lesson is moved to a calmer beach and goes ahead 30% of the time. Draw a probability tree and find the probability that the lesson goes ahead on a randomly chosen day.
Hint
First branches: swell suitable (0.7) or not (0.3). Second branches: lesson goes ahead or is cancelled. Multiply along each 'goes ahead' path, then add.
Answer
P(goes ahead) = 0.7 × 0.9 + 0.3 × 0.3 = 0.72
Step by step
- The tree has two first branches: swell suitable 0.7 and not suitable 1 − 0.7 = 0.3.
- From 'suitable', two branches: goes ahead 0.9 and cancelled 0.1.
- From 'not suitable', two branches: goes ahead 0.3 and cancelled 0.7.
- Each set of branches from one point must add to 1. Check: 0.9 + 0.1 = 1 and 0.3 + 0.7 = 1.
- Multiply along each path that ends in 'goes ahead': 0.7 × 0.9 = 0.63 and 0.3 × 0.3 = 0.09.
- Add the paths: 0.63 + 0.09 = 0.72.
Full working
- suitable 0.7, not 0.3first branches
- 0.7 × 0.9 = 0.63suitable and goes ahead
- 0.3 × 0.3 = 0.09not suitable and goes ahead
- P(goes ahead) = 0.72
A surf school at Mount Maunganui runs one beginners' lesson each day. On 70% of days the swell is suitable. When the swell is suitable, the lesson goes ahead 90% of the time. When the swell is not suitable, the lesson goes ahead 30% of the time.
(i) Given that the lesson went ahead on a particular day, what is the probability that the swell was suitable?
(ii) The summer season has 40 days. How many lessons would you expect to go ahead?
Hint
(i) This is conditional: P(suitable | goes ahead) = P(suitable and goes ahead) ÷ P(goes ahead). (ii) Expected number = probability × number of days.
Answer
(i) 0.63 ÷ 0.72 = 0.875. (ii) 0.72 × 40 = 28.8, so about 29 lessons.
Step by step
- From the tree: P(suitable and goes ahead) = 0.7 × 0.9 = 0.63, and P(not suitable and goes ahead) = 0.3 × 0.3 = 0.09.
- So P(goes ahead) = 0.63 + 0.09 = 0.72.
- (i) We are told the lesson went ahead, so only the 'goes ahead' paths count. That is the new total, 0.72.
- P(suitable | goes ahead) = 0.63 ÷ 0.72 = 0.875.
- Classic mistake: answering 0.9. That is P(goes ahead | suitable), which is the other way round.
- (ii) Expected number = 0.72 × 40 = 28.8.
- Lessons come in whole numbers, so about 29 lessons.
Full working
- P(goes ahead) = 0.63 + 0.09 = 0.72add the paths
- P(suitablegoes ahead) = 0.63 ÷ 0.72 | conditional probability
- = 0.875
- 0.72 × 40 = 28.8expected number
- 0.875; about 29 lessons
The number of students, X, who turn up to a surf lesson has the probability distribution in the table, where k is a constant. Each student pays $65.
(i) Find k.
(ii) On how many of 200 lessons would you expect the school to take at least $260?
(iii) Find the probability that at least 5 students turn up.
| Number of students, x | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|
| P(X = x) | 0.10 | 0.25 | k | 0.20 | 0.15 |
Hint
All the probabilities must add to 1. For (ii), work out how many students make at least $260 at $65 each, add the probabilities for those numbers of students, then multiply by 200.
Answer
(i) k = 0.3. (ii) At least 4 students: P(X ≥ 4) = 0.65, so 0.65 × 200 = 130 lessons. (iii) 0.35
Step by step
- (i) The probabilities add to 1: 0.10 + 0.25 + k + 0.20 + 0.15 = 1.
- 0.70 + k = 1, so k = 0.3.
- (ii) $260 ÷ $65 = 4, so the school takes at least $260 when at least 4 students turn up.
- Check: 3 students pay 3 × $65 = $195, too little; 4 students pay 4 × $65 = $260, enough.
- At least 4 means 4, 5 or 6: P(X ≥ 4) = 0.3 + 0.20 + 0.15 = 0.65.
- Expected number of lessons = 0.65 × 200 = 130 lessons.
- (iii) At least 5 means 5 or 6: 0.20 + 0.15 = 0.35.
Full working
- 0.10 + 0.25 + k + 0.20 + 0.15 = 1, so k = 0.3probabilities add to 1
- 260 ÷ 65 = 4, so at least 4 studentsincome ≥ $260
- P(X ≥ 4) = 0.3 + 0.20 + 0.15 = 0.65add 4, 5 and 6
- 0.65 × 200 = 130expected number of lessons
- P(X ≥ 5) = 0.20 + 0.15 = 0.35
- k = 0.3, 130 lessons, 0.35
Of the students at a surf school, 40% are complete beginners and the rest have surfed before. Of those who have surfed before, 85% stand up on the board during their lesson. Overall, 67% of all students stand up during their lesson.
(i) Draw a probability tree and find the probability that a complete beginner stands up during the lesson.
(ii) A student stood up during the lesson. What is the probability that this student is a complete beginner?
(iii) In a group of 25 students, how many complete beginners would you expect NOT to stand up?
Hint
Let p be the probability a beginner stands up. The 'stands up' paths add to 0.67: 0.4p + 0.6 × 0.85 = 0.67. Solve for p.
Answer
(i) p = 0.4. (ii) 0.16 ÷ 0.67 = 0.2388. (iii) 25 × 0.4 × 0.6 = 6 beginners.
Step by step
- (i) First branches: beginner 0.4, surfed before 0.6.
- From 'surfed before': stands up 0.85, does not 0.15.
- From 'beginner': stands up p (unknown), does not 1 − p.
- The two 'stands up' paths add to the overall 0.67: 0.4 × p + 0.6 × 0.85 = 0.67.
- 0.6 × 0.85 = 0.51, so 0.4p = 0.67 − 0.51 = 0.16, and p = 0.16 ÷ 0.4 = 0.4.
- (ii) P(beginner and stands up) = 0.4 × 0.4 = 0.16.
- P(beginner | stands up) = 0.16 ÷ 0.67 = 0.2388.
- So only about 24% of students who stand up are beginners, even though beginners are 40% of all students, because beginners are much less likely to stand up.
- (iii) P(beginner and does not stand up) = 0.4 × (1 − 0.4) = 0.4 × 0.6 = 0.24.
- Expected number = 0.24 × 25 = 6 beginners.
- Check the tree: 0.16 + 0.24 + 0.51 + 0.09 = 1.
Full working
- 0.4p + 0.6 × 0.85 = 0.67'stands up' paths add to 0.67
- 0.4p = 0.16, so p = 0.4
- P(beginnerstands up) = (0.4 × 0.4) ÷ 0.67 = 0.2388 | conditional
- P(beginner and not) = 0.4 × 0.6 = 0.24multiply along the path
- 0.24 × 25 = 6
- p = 0.4; 0.2388; 6 beginners
After each lesson, the surf school gives the student a wristband in one of four colours. The colour is chosen at random, and each colour is equally likely. A student wants all four colours. A simulation using random whole numbers from 1 to 4 was run 50 times. Each trial counted how many lessons were needed to collect all four colours. The table shows the results.
(i) Use the table to estimate the probability that exactly 4 lessons are needed, the probability that more than 10 lessons are needed, and the mean number of lessons needed.
(ii) Find the theoretical probability that exactly 4 lessons are needed. Compare it with your estimate from (i), and explain the difference. (The theoretical mean is 8.33 lessons.)
| Lessons needed | Number of trials |
|---|---|
| 4 | 2 |
| 5 | 7 |
| 6 | 8 |
| 7 | 7 |
| 8 | 5 |
| 9 | 4 |
| 10 | 3 |
| 11 | 4 |
| 12 | 4 |
| 13 | 2 |
| 14 | 1 |
| 16 | 2 |
| 17 | 1 |
Hint
(ii) The first wristband can be any colour. The second must be one of the 3 colours not yet held, the third one of the 2 left, and the fourth the last 1. Multiply the probabilities.
Answer
(i) P(exactly 4) ≈ 250 = 0.04; P(more than 10) ≈ 1450 = 0.28; mean ≈ 42750 = 8.54 lessons. (ii) 1 × (34) × (24) × (14) = 332 = 0.0938. The estimate 0.04 is well below 0.0938: with only 50 trials we would expect about 4.7 trials needing exactly 4, and 2 is within chance variation. More trials would bring the estimate closer.
Step by step
- (i) Exactly 4 lessons happened in 2 of the 50 trials: 2 ÷ 50 = 0.04.
- More than 10 lessons: 4 + 4 + 2 + 1 + 2 + 1 = 14 trials, so 14 ÷ 50 = 0.28.
- Mean: multiply each number of lessons by its count and add: 8 + 35 + 48 + 49 + 40 + 36 + 30 + 44 + 48 + 26 + 14 + 32 + 17 = 427. Then 427 ÷ 50 = 8.54 lessons.
- (ii) To finish in exactly 4 lessons, all four wristbands must be different colours.
- The first can be any colour: probability 1. The second must be new: 34. The third: 24. The fourth: 14.
- 1 × 34 × 24 × 14 = 664 = 332 = 0.0938.
- The experimental estimate (0.04) is less than half of the theoretical value (0.0938). In 50 trials the theory expects 50 × 0.0938 = 4.7 trials of exactly 4, and the simulation got 2.
- That is because 50 trials is a small number: a rare event can easily come up a few times more or less than expected. The simulated mean (8.54) is close to the theoretical 8.33, so the simulation is working.
- Running many more trials (for example 1000) would give an experimental probability much closer to 0.0938.
Full working
- P(exactly 4) ≈ 250 = 0.04experimental
- P(more than 10) ≈ 1450 = 0.28
- mean ≈ 427 ÷ 50 = 8.54Σ lessons × count ÷ 50
- 1 × 34 × 24 × 14 = 332 = 0.0938theoretical
- 0.04 against 0.0938: few trials, so chance variation; more trials would get closer
Question Three
A study followed 800 secondary school rugby players for one season. The table shows whether each player wore a mouthguard and whether they had a dental injury.
(i) What percentage of the players had a dental injury?
(ii) A player who had a dental injury is chosen at random. What is the probability that this player did not wear a mouthguard?
| Dental injury | No dental injury | Total | |
|---|---|---|---|
| Wore a mouthguard | 18 | 582 | 600 |
| No mouthguard | 27 | 173 | 200 |
| Total | 45 | 755 | 800 |
Hint
(i) Use the Total row. (ii) You are told the player was injured, so look only at the 'Dental injury' column: its total, 45, is the denominator.
Answer
(i) 45800 = 5.625%. (ii) 2745 = 0.6
Step by step
- (i) 45 of the 800 players had a dental injury. 45 ÷ 800 = 0.05625, which is 5.625%.
- (ii) We know the player was injured, so only the 45 injured players count.
- Of those, 27 did not wear a mouthguard. 27 ÷ 45 = 0.6.
- Classic mistake: dividing by 200 or 800. 'Given injured' means divide by the injured total, 45.
Full working
- 45 ÷ 800 = 0.05625 = 5.625%Total row
- 27 ÷ 45given injured, so out of 45
- 5.625%; 0.6
A study followed 800 secondary school rugby players for one season. Of the 600 players who wore a mouthguard, 18 had a dental injury. Of the 200 players who did not wear a mouthguard, 27 had a dental injury. Calculate the relative risk of a dental injury for players who wore a mouthguard compared with players who did not, and interpret it in context.
Hint
Risk = injured ÷ that group's total. Relative risk = risk for mouthguard wearers ÷ risk for non-wearers. A relative risk below 1 means a lower risk.
Answer
Risks 18600 = 0.03 and 27200 = 0.135. RR = 0.03 ÷ 0.135 = 0.222. Mouthguard wearers were 0.222 times as likely to have a dental injury, which is 77.8% less likely (equivalently, non-wearers were 4.5 times as likely).
Step by step
- Risk for mouthguard wearers = 18 ÷ 600 = 0.030.
- Risk for non-wearers = 27 ÷ 200 = 0.135.
- Relative risk = 0.030 ÷ 0.135 = 0.222. The group named first (wearers) goes on top.
- A relative risk below 1 means the risk is lower in the first group.
- In context: players who wore a mouthguard were 0.222 times as likely to have a dental injury as players who did not.
- Another way to say it: 1 − 0.222 = 0.778, so wearers were 77.8% less likely to be injured.
- Or flip it: 0.135 ÷ 0.030 = 4.5, so non-wearers were 4.5 times as likely to have a dental injury.
- Classic mistake: comparing counts. 18 and 27 are not fair to compare because there were three times as many wearers.
Full working
- risk wearers = 18 ÷ 600 = 0.030cases ÷ group total
- risk non-wearers = 27 ÷ 200 = 0.135
- RR = 0.030 ÷ 0.135 = 0.222below 1, so lower risk
- 1 − 0.222 = 0.778, so 77.8% less likely
- RR = 0.222: wearers are 0.222 times as likely to be injured
800 secondary school rugby players were recorded by position and whether they were concussed during the season. Some of the table is missing.
(i) Complete the table.
(ii) What is the probability that a randomly chosen player is a back or was concussed?
(iii) A player who was not concussed is chosen at random. What is the probability that this player is a forward?
| Concussed | Not concussed | Total | |
|---|---|---|---|
| Forwards | 36 | 480 | |
| Backs | 304 | ||
| Total | 52 | 800 |
Hint
Each row and each column must add to its total. For 'back or concussed', add the backs and the concussed players, but do not count the concussed backs twice.
Answer
(i) Forwards not concussed 444; backs concussed 16, backs total 320; total not concussed 748. (ii) 356800 = 0.445. (iii) 444748 = 0.594
Step by step
- (i) Forwards not concussed: 480 − 36 = 444.
- Backs concussed: 52 − 36 = 16. Backs total: 800 − 480 = 320.
- Check the backs row: 16 + 304 = 320. Total not concussed: 444 + 304 = 748, and 52 + 748 = 800.
- (ii) 'Back or concussed' counts every back (320) plus the concussed players who are not backs, the concussed forwards (36).
- 320 + 36 = 356, so the probability is 356 ÷ 800 = 0.445.
- Another way: 320 + 52 − 16 = 356, taking off the 16 concussed backs who were counted twice.
- (iii) Only the 748 players who were not concussed count. Of those, 444 are forwards.
- 444 ÷ 748 = 0.5936, which is 0.594 to three decimal places.
Full working
- 444, 16, 320, 748rows and columns add to totals
- back or concussed = 320 + 36 = 356do not count concussed backs twice
- 356 ÷ 800 = 0.445
- 444 ÷ 748 = 0.5936given not concussed, so out of 748
- 0.445; 0.594
In a season, 36 of 480 school rugby forwards and 16 of 320 backs were concussed. A coach says: "36 forwards were concussed but only 16 backs, so forwards are more than twice as likely to be concussed as backs."
(i) Explain the error in the coach's reasoning.
(ii) Use relative risk to evaluate the coach's claim, and interpret your relative risk in context.
(iii) If backs had the same risk of concussion as forwards, how many backs would you expect to have been concussed?
Hint
The coach compared counts, but the groups are different sizes. Find each group's risk (concussed ÷ group total) and divide the forwards' risk by the backs' risk.
Answer
(i) The coach compared counts (36 ÷ 16 = 2.25) but there are more forwards (480) than backs (320), so the counts are not a fair comparison. (ii) Risks 0.075 and 0.050; RR = 1.5. Forwards were 1.5 times as likely (50% more likely) to be concussed, not more than twice as likely, so the claim is wrong. (iii) 0.075 × 320 = 24 backs.
Step by step
- (i) The coach divided the counts: 36 ÷ 16 = 2.25, which is where 'more than twice' comes from.
- But there are 480 forwards and only 320 backs. A bigger group will have more cases even at the same risk, so counts are not a fair comparison.
- (ii) Risk for forwards = 36 ÷ 480 = 0.075. Risk for backs = 16 ÷ 320 = 0.050.
- Relative risk = 0.075 ÷ 0.050 = 1.5.
- Interpret: forwards were 1.5 times as likely to be concussed as backs, which means 50% more likely.
- 1.5 is well under 2, so the claim that forwards are more than twice as likely is not supported by the data.
- Forwards do have the higher risk, so the coach is right that forwards are more at risk, but wrong about how much.
- (iii) At the forwards' risk, the expected number of concussed backs = 0.075 × 320 = 24.
- Only 16 backs were concussed, which is 1624, or two-thirds, of that. That matches the relative risk the other way round: 0.050 ÷ 0.075 = 0.667.
Full working
- 36 ÷ 16 = 2.25 compares counts, but 480 ≠ 320groups are different sizes
- risk F = 36 ÷ 480 = 0.075, risk B = 16 ÷ 320 = 0.050cases ÷ group total
- RR = 0.075 ÷ 0.050 = 1.5forwards 50% more likely
- 1.5 < 2, so the claim is wrong
- 0.075 × 320 = 24backs at the forwards' risk
- RR = 1.5, claim rejected; 24 backs
A dentist says: "Players with a custom-fitted mouthguard are 0.4 times as likely to have a dental injury as players with a shop-bought mouthguard." In a large rugby competition where every player wears one of these two types, 30% of players wear a custom-fitted mouthguard. The probability that a randomly chosen player wears a shop-bought mouthguard AND has a dental injury is 0.042.
Assume the dentist's claim is correct.
(i) Find the probability that a randomly chosen player has a dental injury.
(ii) A player has a dental injury. What is the probability that this player wears a custom-fitted mouthguard?
(iii) The competition has 1500 players. How many dental injuries would you expect in a season?
Hint
Draw a tree: custom 0.3, shop-bought 0.7. The risk for shop-bought players is P(injured | shop-bought) = 0.042 ÷ 0.7. The custom risk is 0.4 times that.
Answer
(i) Shop-bought risk = 0.042 ÷ 0.7 = 0.06; custom risk = 0.4 × 0.06 = 0.024; P(injury) = 0.3 × 0.024 + 0.042 = 0.0492. (ii) 0.0072 ÷ 0.0492 = 0.146. (iii) 0.0492 × 1500 = 73.8, about 74 injuries.
Step by step
- (i) Tree: first branches custom 0.3 and shop-bought 0.7. Second branches: injured or not.
- The 0.042 is the whole path 'shop-bought AND injured', so 0.7 × P(injured | shop-bought) = 0.042.
- P(injured | shop-bought) = 0.042 ÷ 0.7 = 0.060. This is the risk for shop-bought players.
- The relative risk is 0.4, so the custom risk = 0.4 × 0.060 = 0.024.
- Path custom AND injured = 0.3 × 0.024 = 0.0072.
- P(injured) = 0.0072 + 0.042 = 0.0492.
- Classic mistake: using 0.4 × 0.042. The relative risk compares the risks within each group (0.024 against 0.060), not the path probabilities.
- (ii) Given the player is injured, divide by P(injured): 0.0072 ÷ 0.0492 = 0.1463, which is 0.146.
- So custom players are 30% of all players but only about 15% of injured players, because their risk is lower.
- (iii) Expected injuries = 0.0492 × 1500 = 73.8, so about 74.
- Check: 0.024 ÷ 0.060 = 0.4, which matches the dentist's claim.
Full working
- 0.7 × P(Ishop) = 0.042, so P(I | shop) = 0.060 | path ÷ branch
- P(Icustom) = 0.4 × 0.060 = 0.024 | relative risk 0.4
- P(I) = 0.3 × 0.024 + 0.042 = 0.0072 + 0.042 = 0.0492
- P(customI) = 0.0072 ÷ 0.0492 = 0.146 | conditional
- 0.0492 × 1500 = 73.8expected number
- 0.0492; 0.146; about 74 injuries
AS 91267 past papers · All NCEA Level 2 Maths
A Vertex Academy practice paper, written from NZQA's 2026 assessment specification for AS 91267. It is not an NZQA paper. Grades use the 2025 exam's cut scores: 8, 14 and 20 out of 24 for Achieved, Merit and Excellence.