| Exam Board | Edexcel |
|---|---|
| Module | FS1 AS (Further Statistics 1 AS) |
| Year | 2023 |
| Session | June |
| Marks | 16 |
| Paper | Download PDF ↗ |
| Mark scheme | Download PDF ↗ |
| Topic | Poisson distribution |
| Type | Finding maximum n for P(X=0) threshold |
| Difficulty | Standard +0.3 This is a straightforward Further Statistics 1 question testing standard Poisson distribution techniques: finding parameter from probability (inverse calculation with logarithms), binomial probability, expectation/variance, Poisson approximation to binomial, and a one-tailed hypothesis test. All parts follow routine procedures with no novel problem-solving required, though the multi-part structure and hypothesis testing elevate it slightly above average A-level difficulty. |
| Spec | 5.01a Permutations and combinations: evaluate probabilities5.02i Poisson distribution: random events model5.02j Poisson formula: P(X=x) = e^(-lambda)*lambda^x/x!5.02k Calculate Poisson probabilities5.02l Poisson conditions: for modelling5.02m Poisson: mean = variance = lambda5.05c Hypothesis test: normal distribution for population mean |
| Answer | Marks | Guidance |
|---|---|---|
| Working | Mark | Guidance |
| \([F = \text{no. of faults in } A \text{ m}^2]\) \(F \sim Po(A \times 0.4)\) | M1 | For selecting the correct model \(Po(0.4A)\) |
| \([P(F=0) \Rightarrow]\) \(0.0907 = e^{-0.4A}\) | M1 | For a correct equation; may be implied by correct answer with no incorrect working |
| \(A = 6\) | A1 | For \(A = 6\) (or awrt 6.0) |
| Answer | Marks | Guidance |
|---|---|---|
| Working | Mark | Guidance |
| \([T = \text{no. of tablecloths with no faults}]\) \(T \sim B(20, 0.0907)\) | M1 | For selecting a correct model \(B(20, 0.0907)\) |
| \(P(T > 1) = 1 - P(T \leq 1)\) | M1 | For correctly interpreting "more than 1" to reach \(1 - P(T \leq 1)\); may be implied by A1 |
| \(= 0.55276\ldots =\) awrt \(0.553\) | A1 | For awrt 0.553 |
| Answer | Marks | Guidance |
|---|---|---|
| Working | Mark | Guidance |
| \([X \sim B(100, 0.0907)]\) | M1 | For \(X \sim B(100, 0.0907)\) used; may be implied by correct \(E(X)\) or \(Var(X)\) |
| \(E(X) = 100 \times 0.0907 = 9.07\) | A1 | For 9.07 |
| Answer | Marks | Guidance |
|---|---|---|
| Working | Mark | Guidance |
| \(Var(X) = 100 \times 0.0907 \times (1 - 0.0907) = 8.247351\ldots\) awrt \(8.25\) | A1 | For awrt 8.25. SC - award M0A1A0 for: using \(X \sim B(100, 0.4)\) leading to \(E(X)=40\), \(Var(X)=24\); or using \(X \sim B(100,p)\), \(0
|
| Answer | Marks | Guidance |
|---|---|---|
| Working | Mark | Guidance |
| \(X \approx Po(9.07)\) | M1 | For selecting the correct Poisson model — ft their answer to (c)(i) |
| \(P(X = 10) \approx 0.11947\ldots\) \(\mathbf{0.1195}\) or awrt \(0.119\) | A1 | For 0.1195 or awrt 0.119 |
| Answer | Marks | Guidance |
|---|---|---|
| Working | Mark | Guidance |
| \(H_0: \lambda = 0.4\) (or \(\lambda = 12\)) \(\quad H_1: \lambda < 0.4\) (or \(\lambda < 12\)) | B1 | For both hypotheses correct in terms of \(\lambda\) or \(\mu\) |
| \([Y = \text{no. of faults from new machine}]\) \(Y \sim Po(12)\) | M1 | For selecting a suitable model; sight or use of \(Po(12)\); may be implied by 1st A1 |
| \(P(Y \leq 6) = 0.04582\ldots\) | A1 | For a correct probability; must be 0.046 or better |
| [Significant] there is evidence to support the claim | A1 | For a correct conclusion in context using "claim" or "rate of faults" |
| Answer | Marks | Guidance |
|---|---|---|
| Working | Mark | Guidance |
| \(p\text{-value} = 0.04582\ldots\) awrt \(\mathbf{0.0458}\) | B1ft | B1ft for awrt 0.0458 o.e. e.g. 4.58%, or ft their answer to 1st A1 in (e) |
# Question 3:
## Part (a):
| Working | Mark | Guidance |
|---------|------|----------|
| $[F = \text{no. of faults in } A \text{ m}^2]$ $F \sim Po(A \times 0.4)$ | M1 | For selecting the correct model $Po(0.4A)$ |
| $[P(F=0) \Rightarrow]$ $0.0907 = e^{-0.4A}$ | M1 | For a correct equation; may be implied by correct answer with no incorrect working |
| $A = 6$ | A1 | For $A = 6$ (or awrt 6.0) |
## Part (b):
| Working | Mark | Guidance |
|---------|------|----------|
| $[T = \text{no. of tablecloths with no faults}]$ $T \sim B(20, 0.0907)$ | M1 | For selecting a correct model $B(20, 0.0907)$ |
| $P(T > 1) = 1 - P(T \leq 1)$ | M1 | For correctly interpreting "more than 1" to reach $1 - P(T \leq 1)$; may be implied by A1 |
| $= 0.55276\ldots =$ awrt $0.553$ | A1 | For awrt 0.553 |
## Part (c)(i):
| Working | Mark | Guidance |
|---------|------|----------|
| $[X \sim B(100, 0.0907)]$ | M1 | For $X \sim B(100, 0.0907)$ used; may be implied by correct $E(X)$ or $Var(X)$ |
| $E(X) = 100 \times 0.0907 = 9.07$ | A1 | For 9.07 |
## Part (c)(ii):
| Working | Mark | Guidance |
|---------|------|----------|
| $Var(X) = 100 \times 0.0907 \times (1 - 0.0907) = 8.247351\ldots$ awrt $8.25$ | A1 | For awrt 8.25. SC - award M0A1A0 for: using $X \sim B(100, 0.4)$ leading to $E(X)=40$, $Var(X)=24$; or using $X \sim B(100,p)$, $0<p<1$ and $E(X)=100p$, $Var(X)=100p(1-p)$ |
## Part (d):
| Working | Mark | Guidance |
|---------|------|----------|
| $X \approx Po(9.07)$ | M1 | For selecting the correct Poisson model — ft their answer to (c)(i) |
| $P(X = 10) \approx 0.11947\ldots$ $\mathbf{0.1195}$ or awrt $0.119$ | A1 | For 0.1195 or awrt 0.119 |
## Part (e):
| Working | Mark | Guidance |
|---------|------|----------|
| $H_0: \lambda = 0.4$ (or $\lambda = 12$) $\quad H_1: \lambda < 0.4$ (or $\lambda < 12$) | B1 | For both hypotheses correct in terms of $\lambda$ or $\mu$ |
| $[Y = \text{no. of faults from new machine}]$ $Y \sim Po(12)$ | M1 | For selecting a suitable model; sight or use of $Po(12)$; may be implied by 1st A1 |
| $P(Y \leq 6) = 0.04582\ldots$ | A1 | For a correct probability; must be 0.046 or better |
| [Significant] there is evidence to support the claim | A1 | For a correct conclusion in context using "claim" or "rate of faults" |
## Part (f):
| Working | Mark | Guidance |
|---------|------|----------|
| $p\text{-value} = 0.04582\ldots$ awrt $\mathbf{0.0458}$ | B1ft | B1ft for awrt 0.0458 o.e. e.g. 4.58%, or ft their answer to 1st A1 in (e) |
---
\begin{enumerate}
\item A machine produces cloth. Faults occur randomly in the cloth at a rate of 0.4 per square metre.
\end{enumerate}
The machine is used to produce tablecloths, each of area $A$ square metres. One of these tablecloths is taken at random.
The probability that this tablecloth has no faults is 0.0907\\
(a) Find the value of $A$
The tablecloths are sold in packets of 20\\
A randomly selected packet is taken.\\
(b) Find the probability that more than 1 of the tablecloths in this packet has no faults.
A hotel places an order for 100 tablecloths each of area $A$ square metres.\\
The random variable $X$ represents the number of these tablecloths that have no faults.\\
(c) Find\\
(i) $\mathrm { E } ( X )$\\
(ii) $\operatorname { Var } ( X )$\\
(d) Use a Poisson approximation to estimate $\mathrm { P } ( X = 10 )$
It is claimed that a new machine produces cloth with a rate of faults that is less than 0.4 per square metre.
A piece of cloth produced by this new machine is taken at random.\\
The piece of cloth has area 30 square metres and is found to have 6 faults.\\
(e) Stating your hypotheses clearly, use a suitable test to assess the claim made for the new machine. Use a $5 \%$ level of significance.\\
(f) Write down the $p$-value for the test used in part (e).
\hfill \mbox{\textit{Edexcel FS1 AS 2023 Q3 [16]}}