Statistics: Distributions Depth
In A-level Statistics, Distributions Depth takes the binomial and normal distributions beyond a single direct probability calculation: finding the mean and variance of a linear transformation of a binomial or normal random variable (such as a total cost built from a fixed number of trials), finding an unknown mean or standard deviation of a normal distribution from a given probability, including solving simultaneously when BOTH are unknown, and applying the continuity correction precisely in all four directions when a normal distribution is used to approximate a binomial one. The examiner is testing algebraic control of E(aX+b) and Var(aX+b), fluency with the inverse normal function, and care with strict versus non-strict inequalities under a continuity correction.
Before you start
Make sure you're comfortable with these topics first:
Method
- Decide what the question needs. A new variable defined as Y = aX + b (or built from several independent trials): use E(aX+b) = aE(X) + b and Var(aX+b) = a^2 Var(X). One unknown parameter (mu or sigma) of a normal distribution with one given probability: standardise and use the inverse normal function. BOTH mu and sigma unknown, with two given probabilities: form two equations and solve simultaneously. Approximating a binomial with a normal distribution: check np>5 and n(1-p)>5, then apply a continuity correction.
- For X ~ B(n,p), E(X) = np and Var(X) = np(1-p). For any linear transformation Y = aX + b, E(Y) = aE(X) + b and Var(Y) = a^2 Var(X); adding a constant b never changes the variance, only multiplying by a does.
- To find one unknown parameter from a single probability statement about X ~ N(mu, sigma^2): standardise using z = (x-mu)/sigma, find the exact z-value that gives the stated probability using the inverse normal function, then substitute and rearrange for the unknown.
- When BOTH mu and sigma are unknown, form two equations of the form (x - mu)/sigma = z from two different given probabilities (each with its own x-value and its own z-value), then solve simultaneously, typically by subtracting one equation from the other to eliminate mu first.
- For a normal approximation Y ~ N(np, np(1-p)) to X ~ B(n,p), apply a continuity correction, treating each whole number as the interval from half below it to half above it: P(X > r) becomes P(Y > r+0.5); P(X >= r) becomes P(Y > r-0.5); P(X < r) becomes P(Y < r-0.5); P(X <= r) becomes P(Y < r+0.5).
- Always check np > 5 and n(1-p) > 5 before using a normal approximation to a binomial distribution; if either condition fails, use the exact binomial distribution instead.
Common mistakes
MistakeCalculating Var(aX + b) as a Var(X) + b, or as a^2 Var(X) + b.
InsteadAdding a constant does not change the spread, and multiplying by a multiplies the variance by a^2: Var(aX + b) = a^2 Var(X). Only the mean takes the constant: E(aX + b) = a E(X) + b.
MistakeUsing a positive z-value for a probability below 0.5, so that the unknown ends up on the wrong side of the mean.
InsteadBelow 0.5 the z-value is negative. Rewrite every statement as P(X < x) = p and use the inverse normal function on p, or use symmetry: the z-value for p is minus the z-value for 1 - p.
MistakeApplying a continuity correction to a normal variable that is already continuous, such as a mass or a time.
InsteadThe continuity correction is only for using a normal model to approximate a discrete variable, such as a binomial count. A normal model of a measured quantity has no gaps, so use the values as given.
Worked examples
Example 1
The random variable X ~ N(mu, sigma^2). Given that P(X < 20) = 0.2 and P(X < 35) = 0.9, find the values of mu and sigma, each to 3 significant figures.
- Standardise both probability statements: (20 - mu)/sigma = z1, where P(Z < z1) = 0.2, and (35 - mu)/sigma = z2, where P(Z < z2) = 0.9.
- Find the z-values from the standard normal distribution: z1 = -0.8416 and z2 = 1.2816 (4dp each, using the inverse normal function).
- Write the two equations: 20 - mu = -0.8416 sigma ... (1); 35 - mu = 1.2816 sigma ... (2).
- Subtract equation (1) from equation (2) to eliminate mu: 15 = 2.1232 sigma.
- Solve for sigma: sigma = 15/2.1232 = 7.06 (3 s.f.).
- Substitute back into equation (1): mu = 20 + 0.8416(7.0649) = 25.9 (3 s.f.).
Example 2
The journey time T minutes of a taxi ride is modelled by T ~ N(mu, 6^2). The fare C pounds is given by C = 0.4T + 6. Given that P(C > 22) = 0.05, (a) find the value of mu, (b) find P(C < 15) using this value of mu.
- C is a linear function of a normal variable, so C is normal with E(C) = 0.4mu + 6 and Var(C) = 0.4^2 x 6^2 = 5.76, so the standard deviation of C is 2.4. The constant 6 does not affect the variance.
- P(C > 22) = 0.05 means P(C < 22) = 0.95, so the z-value is 1.6449 (4 d.p.).
- Standardise: (22 - (0.4mu + 6))/2.4 = 1.6449, so 16 - 0.4mu = 3.9478.
- Therefore 0.4mu = 12.0522 and mu = 30.13, which is 30.1 (3 s.f.).
- With mu = 30.13, E(C) = 0.4(30.13) + 6 = 18.05, so for C < 15, z = (15 - 18.05)/2.4 = -1.272.
- Final answer: (a) mu = 30.1, (b) P(C < 15) = P(Z < -1.272) = 0.102 (3 s.f.).
Practice questions
Q1[2 marks]X ~ B(20, 0.3). Find E(4X + 5).Show answer
Answer: E(X) = 20(0.3) = 6, so E(4X+5) = 4(6)+5 = 29.
Q2[2 marks]X ~ B(20, 0.3). Find Var(4X + 5).Show answer
Answer: Var(X) = 20(0.3)(0.7) = 4.2, so Var(4X+5) = 4^2 x 4.2 = 67.2.
Q3[4 marks]X ~ N(mu, 6^2). Given that P(X > 40) = 0.15, find the value of mu, giving your answer to 1 decimal place.Show answer
Answer: P(X<40) = 0.85, so z = 1.0364 (4dp). (40-mu)/6 = 1.0364, giving mu = 40 - 6(1.0364) = 33.8.
Q4[3 marks]X ~ N(50, sigma^2). Given that P(X < 58) = 0.9, find sigma, giving your answer to 2 significant figures.Show answer
Answer: z = 1.2816 (4dp, since P(Z<1.2816)=0.9). (58-50)/sigma = 1.2816, so sigma = 8/1.2816 = 6.2.
Q5[1 mark]X ~ B(80, 0.4) is to be approximated by a normal distribution Y. State the continuity-corrected inequality used to approximate P(X >= 35).Show answer
Answer: P(Y > 34.5).
Q6[2 marks]X ~ B(15, 0.1). Explain, with a calculation, why a normal approximation would NOT be appropriate for X.Show answer
Answer: np = 15(0.1) = 1.5, which is not greater than 5, so the condition np>5 fails. A normal approximation should not be used here; the exact binomial distribution should be used instead.
Q7[4 marks]X ~ B(100, 0.5) is approximated by Y ~ N(50, 25). Use this approximation, with a continuity correction, to estimate P(45 <= X <= 55), giving your answer to 3 significant figures.Show answer
Answer: P(45<=X<=55) becomes P(44.5<Y<55.5). Standardising with mean 50 and standard deviation 5: z = -1.1 and z = 1.1. This gives a probability of 0.729 (3 s.f.).
Exam-style questions
Written in the style of a A Level Maths exam paper, with a full mark scheme.
Exam tips
- When using a normal approximation to a binomial, state Y ~ N(np, np(1 - p)) with the variance in the second place, and take its square root when you standardise.
- Keep z-values to at least four decimal places from the inverse normal function, and round mu and sigma only in the final line.
- Check a value of mu or sigma by using it to recompute one of the given probabilities. The calculator should return the stated value.
Q1[7 marks]A factory finds that 8% of items on a production line are defective. A random sample of 150 items is selected. Let X be the number of defective items in the sample. (a) State two conditions needed for a normal distribution to be a suitable approximation for X, and show that both are satisfied here. (2) (b) Using a suitable normal approximation, find P(X > 16), giving your answer to 3 significant figures. (5)Show mark scheme
Tick each line you got.
Nothing ticked yet - 7 available
Q2[5 marks]A discrete random variable X ~ B(40, 0.25) models the number of faulty bulbs in a box of 40. Each faulty bulb costs the manufacturer 3 pounds to replace, plus a fixed inspection fee of 20 pounds per box regardless of the number of faults. The total cost, in pounds, of a box is modelled by C = 3X + 20. (a) Find E(C) and Var(C). (4) (b) A different box has 60 bulbs with the same fault probability 0.25, using the same cost model C = 3X + 20. Explain, without further calculation, whether E(C) would increase, decrease, or stay the same compared with the 40-bulb box. (1)Show mark scheme
Tick each line you got.
Nothing ticked yet - 5 available
Q3[4 marks]X ~ B(200, 0.45) is approximated by a normal distribution. (a) State the mean and variance of the normal approximation. (2) (b) Write down the continuity-corrected inequality (in terms of the normal variable Y) that should be used to estimate each of: (i) P(X < 85) (ii) P(X >= 95). (2)Show mark scheme
Tick each line you got.
Nothing ticked yet - 4 available
See real A Level Maths past-paper questions, with official mark schemes
This topic in real past papers
We have tagged some boards' past papers question by question (not every board or year yet). This topic comes up in:
- WJEC A Level Maths past papers: 11 questions across 8 papers, latest tagged Summer 2024
- OCR A Level Maths past papers: 10 questions across 6 papers, latest tagged June 2024
- AQA A Level Maths past papers: 2 questions across 2 papers, latest tagged November 2021
Free printable worksheet
Want more practice on paper? Download the Statistics: Distributions Depth worksheet pack - 7 pages of exam-style questions with a full mark scheme. Free to download, no card or account. Print it for personal and classroom use.
This topic is chapter 9 of A Level Maths Statistics Essentials, the whole course as one free printable PDF. Its worksheet is also chapter 9 of A Level Maths Statistics Topic Workbook and chapter 9 of A Level Maths 10-Minute Tests 2, free printable practice books.
Other versions of this worksheet: