This booklet covers every formula the 18 Edexcel International A Level Mathematics Unit S2 papers in our bank needed, from January 2021 to January 2026. S2 tests the binomial and Poisson distributions, continuous random variables, the uniform distribution, normal approximations, sampling and hypothesis tests. Each row is marked Given or Learn it and lists the past questions that used it.
| Board | Official document |
|---|---|
| Pearson Edexcel International A Level Mathematics | Edexcel IAL Maths Unit S2 (WST02) papers January 2021 to January 2026; Mathematical Formulae and Statistical Tables, Issue 2, January 2021 (Pearson Edexcel International Advanced Subsidiary/Advanced Level in Mathematics, Further Mathematics and Pure Mathematics) |
Given means it is in the Mathematical Formulae and Statistical Tables booklet (Issue 2, 2021) handed out in every unit exam. Learn it means a past question needed it and the exam did not print it.
Binomial and Poisson distributions
The booklet prints the probability, mean and variance formulae for both distributions, and the cumulative tables X4 and X5 follow this section. The approximations and the rules for converting probabilities are not printed.
| Formula | Equation | In the exam |
|---|---|---|
| Binomial B(n, p): probabilityn is the number of trials and p the probability of success on each. Use the tables X4 when n and p are in them. Used in: Jan 2026 (A) Q2(a), P(X = 3) for B(10, p) in terms of p; Jun 2023 Q1(a)(i), exactly 26 of 50 adults; Jan 2021 Q1(c)(i), exactly 24 of 30 oysters survive.Given | Given | |
| Binomial B(n, p): meannp is the expected number of successes. Used in: Jun 2025 Q3(a), n and p from a mean of 3 and a variance of 2.55; Jun 2025 Q3(b), the expected number of green sweets in 40; Jan 2026 (A) Q2(c), n = 80 from np = 32.Given | Given | |
| Binomial B(n, p): varianceThe standard deviation is the square root of np(1 - p). It supplies the second parameter of a normal approximation. Used in: Jan 2025 Q6(a), variance of R in terms of p; Oct 2023 Q7, n and p from a normal approximation; Jan 2026 (A) Q2(c), α = 19.2.Given | Given | |
| Poisson Po(λ): probabilityx is a whole number and λ is the mean for the interval in the question. Used in: Jan 2026 (A) Q2(b), λ when P(Y = 3) is 5 times P(Y = 5); Jan 2025 Q1(e), λ from P(Y = 6) and P(Y = 7); Jun 2022 Q1(c), P(W = 2) for Po(4).Given | Given | |
| Poisson Po(λ): meanThe mean is the average rate for the interval asked about, so scale it to that interval first. Used in: Jan 2021 Q3(c), a mean of 7 per 100 ml becomes 0.07n for n ml; Jan 2026 Q6(f), a mean of 5 per 150 m2 becomes 30 per 900 m2; Oct 2025 Q2(a), the sample mean 1.98.Given | Given | |
| Poisson Po(λ): varianceThe variance equals the mean, which is how a question tests whether data fit a Poisson model. Used in: Oct 2025 Q2(a), mean 1.98 and variance 1.95; Jun 2022 Q1(a), the variance of W ~ Po(4); Jan 2021 Q3(c), N(0.07n, 0.07n).Given | Given | |
| Cumulative probabilities from the tablesThe tables X4 and X5 print P(X <= x) only, so convert 'more than', 'at least' and 'between' first and take care with the end values. Used in: Jan 2026 Q1(d), more than 40% of 50 customers; Jan 2026 (A) Q1(b), at least 3 of 8 rods; Jan 2026 Q3(a)(i), at least 4 detections.Learn it | Learn it | |
| At least one success in n trialsSet it above the target probability and solve for n with logarithms. Dividing by log(1 - p), which is negative, reverses the inequality. Used in: Oct 2025 Q5(b), the least number of tulips for P(at least 1) > 0.95; Jun 2025 Q4(b), the least n for P(Y >= 1) > 0.95.Learn it | Learn it | |
| Poisson over a different intervalThe specification gives Po(λ) per minute becoming Po(5 λ) per 5 minutes. Scale the mean to the interval in the question before you use the tables. Used in: Jan 2026 Q3(a), 2.5 per 15 minutes becomes 5 per 30 minutes; Jan 2026 (A) Q7(b), a rate of 1 per 2 months gives Po(0.5m); Oct 2025 (A) Q5(a), 3 per 5 minutes gives 6 per 10 minutes.Learn it | Learn it | |
| Poisson approximation to the binomialState the conditions: n large and p small, and the schemes allow 'p < 0.1' or 'p close to 0'. Used in: Jan 2026 (A) Q1(c), more than 5 of 60 rods with p = 0.05; Jun 2023 Q4(b), no more than 6 of 125 candles; Jan 2025 Q1(b), P(X = 2) for 100 trials.Learn it | Learn it | |
| Percentage error of an approximationUse the exact binomial probability as the denominator. Used in: Jan 2025 Q1(c), the percentage error of the Poisson approximation.Learn it | Learn it |
Continuous random variables
The specification lists P(a < X <= b) and f = dF/dx as formulae that will not appear in the booklet. The booklet prints the mean, variance, E(g(X)) and F(x) in integral form.
| Formula | Equation | In the exam |
|---|---|---|
| Probability from a p.d.f.The specification lists it as a formula you are expected to know. A single value of a continuous variable has probability 0, so < and <= give the same answer. Used in: Jan 2026 (A) Q6(c), show that P(X > 2) = 0.8; Jun 2023 Q3(b), P(2 < X < 3); Jan 2026 Q2(b), P(X = 5) = 0.Learn it | Learn it | |
| P.d.f. from the c.d.f.The specification lists it as a formula you are expected to know. Differentiate each piece of F separately and keep the same intervals, with f(x) = 0 elsewhere. Used in: Jan 2026 Q2(d), specify the p.d.f.; Oct 2025 Q4(d), r(x) and s(x); Oct 2025 (A) Q1(c).Learn it | Learn it | |
| Probability from the c.d.f.Read the correct piece of F for each value. Used in: Jan 2026 Q2(c), P(5 < X < 7); Oct 2025 Q4(c); Oct 2025 (A) Q1(a), P(X > 4).Learn it | Learn it | |
| Total probability and constantsUse it to find constants: put F at the top of the range equal to 1 and make the pieces of a c.d.f. equal where they join. Used in: Jan 2026 Q2(a), show that k = 1; Oct 2025 Q4(a), show that k = 1/48; Jun 2023 Q5(a)(b), a, b and c in a c.d.f.Learn it | Learn it | |
| Expectation (mean)Differentiate F to get f first when the c.d.f. is given, then integrate x f(x) over the range. Used in: Jan 2026 (A) Q3(b)(i), E(X) by calculus; Jan 2026 Q6(c), the mean weight of watermelons; Jan 2021 Q4(b), use algebraic integration to show a = 6.Given | Given | |
| VarianceIntegrate x2 f(x), then subtract μ2. Some questions give E(X), so Var(X) = E(X2) - E(X)2 needs only E(X2). Used in: Jan 2026 (A) Q3(b)(ii); Oct 2025 Q7(b), show Var(Y) = 66/7225; Jan 2021 Q2(c), algebraic integration with E(W) = 1.6.Given | Given | |
| Expectation of a functionRearranged, E(X2) = Var(X) + [E(X)]2 gives E(X2) when the variance is known. Used in: Jan 2026 Q2(g), E(2X2 - 1) using Var(X) = 3 and E(X) = 7; Jun 2023 Q5(e), E(6Y - 5); Jan 2021 Q5(c), E(X2) for a hypotenuse.Given | Given | |
| Cumulative distribution functionIntegrate f(t) piece by piece and fix each constant with F(lowest) = 0 or F(highest) = 1, then write F in every interval. Used in: Jan 2026 (A) Q6(d), define F(x) fully; Jan 2022 Q4(c); Oct 2021 Q6(d).Given | Given | |
| Linear change of a variableFrom Statistics 1, which the specification says is assumed. The constant drops out of the variance and a is squared. Used in: Oct 2025 (A) Q7(a), E(3 - 2X); Oct 2021 Q6(b), Var(2Y - 3); Oct 2025 Q7(c).Learn it | Learn it | |
| Median and quartilesThe specification includes the median and quartiles of a continuous variable. Set the right piece of F equal to the value and solve, checking the answer lies in that interval. Used in: Oct 2025 Q7(a), show that the median is 2.165; Jan 2026 Q6(e), the median lies between 3.7 and 3.8; Oct 2025 (A) Q4(d).Learn it | Learn it | |
| Mode of a continuous variableDifferentiate f and set it to 0, or take an end of the interval if f is greatest there. If the derivative gives a minimum or a value outside the interval, the mode is at an end point. Used in: Jan 2026 Q6(b), mode of a quadratic p.d.f.; Jun 2023 Q3(d)(e), why Kei's method fails; Oct 2025 (A) Q4(b), use calculus.Learn it | Learn it | |
| Derivative of a powerThe specification says differentiation of polynomials is assumed. You need it to turn F into f. Used in: Jan 2026 Q2(d); Oct 2025 Q4(d); Jan 2026 (A) Q3(b).Learn it | Learn it | |
| Integral of a powerThe specification says integration of polynomials is assumed. Questions often say that calculator technology is not acceptable, so show the integral with limits. Used in: Jan 2026 (A) Q3(b), E(X) and E(X2); Jun 2023 Q5(e); Oct 2025 Q7(b).Learn it | Learn it | |
| Quadratic formulaFrom Pure Mathematics 1. Setting F(a) to a probability often gives a quadratic, and you must reject a root outside the range of X. Used in: Jan 2026 (A) Q6(f), a to 3 significant figures; Jan 2026 (A) Q3(a), F(a) = 0.6.Learn it | Learn it |
Continuous uniform distribution
The booklet prints the p.d.f., mean and variance. The specification also asks for the derivation of the mean, variance and c.d.f. Questions often put a uniform length into a shape.
| Formula | Equation | In the exam |
|---|---|---|
| Uniform (rectangular) on [a, b]: p.d.f.The p.d.f. is constant on [a, b] and 0 outside it, and the rectangle under it has area 1. Used in: Jan 2026 (A) Q5(a), L ~ U[20, 40]; Jan 2025 Q3(a), the p.d.f. on [-k, 5k]; Oct 2025 Q6(a), X uniform on [0, 20].Given | Given | |
| Uniform (rectangular) on [a, b]: meanThe midpoint of the interval. Used in: Oct 2025 (A) Q7(a), E(3 - 2X) in terms of a and b; Jan 2025 Q3(c), the mean on [-k, 5k]; Jun 2025 Q6(a), a and b from a mean of 6.Given | Given | |
| Uniform (rectangular) on [a, b]: varianceThe range is b - a, so a range of 18 gives variance 182 / 12. Used in: Jun 2023 Q6(b)(i), a and b from Var(W) = 75; Jun 2025 Q6(c), the exact value of Var(X); Oct 2025 (A) Q7(d).Given | Given | |
| Uniform probabilities and c.d.f.The probability is the length of the interval over the total length. Used in: Jan 2026 (A) Q5(b), a longer piece of 28 cm to the nearest cm; Jan 2025 Q3(d), the c.d.f. on [-k, 5k]; Oct 2025 Q6(b).Learn it | Learn it | |
| Perimeter and area of a rectangle or squareUsed to turn a uniform length into an area, then solve the inequality for X. Used in: Oct 2025 Q6(a), the area X(20 - X) of a rectangle with perimeter 40 cm; Jan 2026 (A) Q5(c), the area of the larger square made from a string.Learn it | Learn it | |
| Pythagoras' theoremc is the hypotenuse. Used in: Jan 2021 Q5(b), show that X2 = 2W2 - 6W + 9 for the longest side of a right-angled triangle.Learn it | Learn it |
Normal approximations
The tables X2 and X3 follow this section. The booklet prints neither the approximations nor the continuity correction, so learn all three.
| Formula | Equation | In the exam |
|---|---|---|
| Normal approximation to the binomialSay which distribution you use. The schemes want n large and p close to 0.5 as the condition, and the standard deviation is the square root of np(1 - p). Used in: Oct 2025 Q5(c), N(25, 18.75); Jun 2023 Q1(b)(i), N(240, 144); Jan 2026 Q7(b), N(81, 44.55).Learn it | Learn it | |
| Normal approximation to the PoissonFor a large mean, after scaling λ to the interval asked about. Used in: Jan 2026 Q6(f), N(30, 30); Jan 2021 Q3(c), N(0.07n, 0.07n); Jan 2022 Q1(b), six weeks gives N(36, 36).Learn it | Learn it | |
| Continuity correctionAdd or subtract 0.5 on the boundary so that the whole integer is included. Used in: Jan 2026 Q7(b), more than k becomes k + 0.5; Oct 2022 Q4(b), fewer than n becomes n - 0.5; Oct 2025 Q5(c), more than 21 but no more than 30 uses 21.5 and 30.5.Learn it | Learn it | |
| Standardising a Normal variableFrom Statistics 1, which the specification says is assumed. Use the mean and standard deviation of the approximating normal, and show the standardisation with numbers. Used in: Jan 2026 Q7(b); Oct 2025 Q5(c); Jun 2023 Q1(b)(i).Learn it | Learn it | |
| Symmetry of the standard NormalThe table X2 starts at z = 0.00, so for a negative z read the positive value and subtract from 1. Used in: Jun 2023 Q1(b)(i), P(Z < -1.46) = 1 - 0.9279; Jan 2022 Q1(b), P(Z < -0.75); Oct 2025 Q5(c).Learn it | Learn it |
Sampling distributions and tests
These rows come from Statistics 1 and are assumed in S2. The method of a test is in the Also learn list below.
| Formula | Equation | In the exam |
|---|---|---|
| Expectation (mean)Multiply each value of the statistic by its probability and add. Used in: Jan 2022 Q6(c), E(T) for a total of breakfast items; Oct 2023 Q4(a), the mean value of the t-shirts.Given | Given | |
| Independent eventsFrom Statistics 1, which the specification says is assumed. Use it for independent periods, observations or samples. Used in: Jan 2026 Q3(a)(ii), 3 independent 30-minute periods; Jun 2022 Q1(c), P(W = 2 and X = 2); Jan 2025 Q2(ii), two independent observations of Y.Learn it | Learn it | |
| Mean and variance of dataFrom Statistics 1. Sxx in the booklet is the top line of the variance. Used in: Oct 2025 Q2(a), show that daily complaints fit a Poisson model.Learn it | Learn it |
Learn these: not in the booklet
- Sampling termsA census asks the whole population, a sample only part of it. A sampling frame is a list of all the population, and a sampling unit is one member of it. Advantages of a sample: cheaper, quicker; a disadvantage: it may not be representative. Used in: Jan 2026 Q1(a)(b); Oct 2025 Q1; Jun 2023 Q2(a)(b).
- Statistic and parameterA statistic is calculated only from sample values and has no unknown parameters; μ is an unknown population parameter and the sample mean is a statistic. Used in: Jun 2023 Q2(c); Jan 2025 Q2(i).
- Sampling distribution of a statisticIt is all the values a statistic can take with their probabilities when every possible sample is considered. Used in: Oct 2025 Q3(a); Jan 2022 Q6, the total of breakfast items for 2 customers; Jan 2026 Q5(c).
- Building a sampling distributionList the possible samples, find the probability of each (multiply, and count the arrangements, for example 3p2 q), group the samples giving the same value of the statistic and check the total is 1. Used in: Jan 2026 Q5(c)(d); Oct 2025 Q3(d)(e); Jun 2025 Q4(a).
- Conditions for a modelBinomial: a fixed number of independent trials with a constant probability of success. Poisson: events occur singly, at random, independently and at a constant rate. Used in: Jan 2021 Q1(a)(b); Oct 2023 Q5(a); Jun 2025 Q2(a)(b).
- HypothesesWrite H0 and H1 in terms of p or λ, never the sample values, and use a one-tailed H1 for 'more than' or 'less than' and a two-tailed H1 for 'changed'. Used in: Jan 2026 Q3(b); Jan 2026 (A) Q4(a)(i); Oct 2025 Q5(d).
- Carrying out a testUnder H0 find the probability of a result at least as extreme as the observed one and compare it with the significance level (halved in each tail for a two-tailed test), or find the critical region and see whether the observation is in it. Give a conclusion about H0, then one in context using the words of the claim. Used in: Oct 2025 Q2(c); Jan 2026 Q4(b); Jan 2026 (A) Q4(b).
- Critical region and actual significance levelThe critical region is the set of values for which H0 is rejected, with each tail as close to the significance level as possible. The actual significance level is the probability of falling in it, so add both tails. Used in: Jan 2026 (A) Q4(a)(ii)(c); Jun 2025 Q2(d)(e); Jan 2026 Q3(c).
- Tests with an approximationWhen the question says to use a suitable approximation, use Po(np) or a normal approximation; an exact binomial answer can lose the marks. Used in: Jan 2026 Q4(b); Oct 2025 Q5(c); Oct 2021 Q1(d).
- Counts from two Poisson intervalsGiven the total from two Poisson intervals, the count in the first is binomial B(total, share of the rate), which the schemes accept as an alternative to conditional probability. Used in: Jan 2026 (A) Q7(e); Jun 2023 Q7(c).
- A valid p.d.f.A p.d.f. is never negative and has total area 1, so a sketch that dips below the axis shows it is not valid. Used in: Oct 2025 (A) Q4(a).
- Skewness of a continuous distributionCompare the mean, median and mode: median close to mode suggests no skew, and a longer upper tail is positive skew. Used in: Oct 2025 (A) Q4(e); Jan 2022 Q4(e).
Download the booklet
This page as a printable PDF, with the full CuratedPrep study booklet for the course. Open the dropdown, then click a title to open it or use Download to save the PDF.
Edexcel International (IAL)1 booklet
Edexcel IAL maths units1 booklet
How to use the formula sheet in the exam
- Write down the three c.d.f. rules before you start a continuous question: P(a < X < b) = F(b) - F(a), f = dF/dx and F(top) = 1. The booklet prints E(X), Var(X) and the uniform formulae, not these.
- The tables X4 and X5 print P(X <= x) only. Convert 'more than', 'at least' and 'between' before you read them, and scale a Poisson mean to the interval in the question first.
- When a question says use an approximation, name it, give its parameters and apply the continuity correction: N(np, np(1 - p)) for a binomial and Po(np) when n is large and p is small. The exact binomial answer loses the marks.
- For a test, state H0 and H1 in terms of p or λ, compare a probability (or critical region) with the significance level, and finish with a conclusion in the context of the claim.
- Where a question says 'algebraic integration' or that calculator technology is not acceptable, show the integral of x f(x) or x² f(x) with limits, and use Var(X) = E(X²) - E(X)².
- For a uniform variable, read the interval carefully: the p.d.f. is 1 / (b - a), the mean is (a + b) / 2 and the variance is (b - a)² / 12, all printed in the booklet.
Common questions
Which formulae does the Edexcel IAL S2 booklet give me?
The Mathematical Formulae and Statistical Tables booklet prints the binomial and Poisson probability, mean and variance formulae, the continuous mean, variance, expectation of a function and cumulative distribution function, the uniform distribution formulae, and the tables X2 to X5. It also notes that S2 candidates may need the S1, P1, P2, P3 and P4 formulae.
Do I need to learn P(a < X <= b) and f = dF/dx?
Yes. The specification lists the integral for P(a < X <= b) and f(x) = dF(x)/dx as formulae you are expected to know, and says they will not appear in the booklet.
Are the normal approximations and the continuity correction in the booklet?
No. The booklet prints the normal distribution tables but not the approximations N(np, np(1 - p)) and N(λ, λ) or the continuity correction. The specification includes the approximations with the continuity correction, so learn them.
Other units
Unit P1 · Unit P2 · Unit P3 · Unit P4 · Unit FP1 · Unit FP2 · Unit FP3 · Unit M1 · Unit M2 · Unit M3 · Unit S1 · Unit S3 · Unit D1. All Edexcel IAL unit formula sheets.
Practise this topic
Real past paper questions, by topic
CuratedPrep worksheets group real past paper questions by syllabus topic, with a full marking scheme and revision notes, for every Edexcel IAL maths topic.
In Dubai or the UAE? Work one to one with a CuratedPrep maths tutor.
Formulas checked against the official documents linked above. Boards update their booklets from time to time: always practise with the version your exam uses.