Edexcel S2 2021 January — Question 3 17 marks

Exam BoardEdexcel
ModuleS2 (Statistics 2)
Year2021
SessionJanuary
Marks17
PaperDownload PDF ↗
Mark schemeDownload PDF ↗
TopicPoisson distribution
TypePoisson hypothesis test
DifficultyStandard +0.3 This is a standard S2 Poisson question with routine applications: (a) basic Poisson probability calculation, (b) binomial probability using result from (a), (c) normal approximation with continuity correction leading to a given quadratic (heavily scaffolded), and (d) standard hypothesis test. All parts follow textbook procedures with no novel insight required, making it slightly easier than average.
Spec2.04b Binomial distribution: as model B(n,p)2.04c Calculate binomial probabilities2.04d Normal approximation to binomial2.05b Hypothesis test for binomial proportion2.05c Significance levels: one-tail and two-tail5.02i 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 = lambda

3. The number of water fleas, in 100 ml of pond water, has a Poisson distribution with mean 7
  1. Find the probability that a sample of 100 ml of the pond water does not contain exactly 4 water fleas. Aja collects 5 separate samples, each of 100 ml , of the pond water.
  2. Find the probability that exactly 1 of these samples contains exactly 4 water fleas. Using a normal approximation, the probability that more than 3 water fleas will be found in a random sample of \(n \mathrm { ml }\) of the pond water is 0.9394 correct to 4 significant figures.
    1. Show that \(n - 1.55 \sqrt { \frac { n } { 0.07 } } - 50 = 0\)
    2. Hence find the value of \(n\) After the pond has been cleaned, the number of water fleas in a 100 ml random sample of the pond water is 15
  3. Using a suitable test, at the \(1 \%\) level of significance, assess whether or not there is evidence that the number of water fleas per 100 ml of the pond water has increased. State your hypotheses clearly. \includegraphics[max width=\textwidth, alt={}, center]{f63c39df-cfc9-4a6b-838d-67613710b0ce-11_2255_50_314_34}
    VIXV SIHIANI III IM IONOOVIAV SIHI NI JYHAM ION OOVI4V SIHI NI JLIYM ION OO

Question 3:
Part (a):
AnswerMarks Guidance
Answer/WorkingMark Guidance
\(P(X \neq 4) = 1 - P(X=4)\), \(= 1 - \frac{e^{-7}7^4}{4!}\) or \(1-(0.1730-0.0818)\)M1 For \(1-P(X=4)\) or \(1-P(X \leq 4)+P(X \leq 3)\) oe
\(= 0.90877\ldots\)A1 awrt 0.909
Part (b):
AnswerMarks Guidance
Answer/WorkingMark Guidance
\(P(Y=1) = (1-\text{"0.90877..."})(\text{"0.90877..."})^4 \times {}^5C_1\)M1M1 \((1-\text{"their 0.909"})^1(\text{"their 0.909"})^4\) or \((1-\text{"their 0.909"})(\text{"their 0.909"})^4\) allow values to 2s.f.; \(P(Y=1)=(1-\text{"their 0.909"})(\text{"their 0.909"})^4 \times {}^5C_1\) allow values to 2s.f.
\(= 0.311\ldots\)A1 awrt 0.312 or awrt 0.311
Part (c)(i):
AnswerMarks Guidance
Answer/WorkingMark Guidance
\(\lambda = 0.07n\)B1 Writing or using mean as \(0.07n\)
\(A \sim N(0.07n, 0.07n)\)M1 Normal with mean = variance which must be in terms of \(n\)
\(\frac{3.5 - \text{"0.07n"}}{\sqrt{\text{"0.07n"}}}\)M1 Standardising with their mean and their \(\sqrt{\text{var}}\); allow 2.5, 3, 3.5, 4, 4.5 (correct standardisation implies B1M1M1)
\(\frac{3.5-0.07n}{\sqrt{0.07n}} = -1.55\) or \(\text{"0.07n"} - (1.55\sqrt{0.07})\sqrt{n} - 3.5 = 0\)B1 Their standardisation \(= \pm 1.55\)
\(n - \left(\frac{1.55}{0.07}\sqrt{0.07}\right)\sqrt{n} - \frac{3.5}{0.07} = 0 \Rightarrow n - 1.55\sqrt{\frac{n}{0.07}} - 50 = 0\)A1cso Must come from compatible signs; need at least one step between standardisation indicating division by 0.07 and correct equation
Part (c)(ii):
AnswerMarks Guidance
Answer/WorkingMark Guidance
\(\sqrt{n} = \frac{\frac{1.55}{\sqrt{0.07}} \pm \sqrt{\left(\frac{1.55}{\sqrt{0.07}}\right)^2 + 4\times 50}}{2} = \text{awrt } {-4.72}\ldots \text{ or awrt } 10.6\ldots\ (4\sqrt{7})\)M1 Correct method to solve given quadratic or sight of awrt \(-4.72\) or awrt \(10.6\)
\(n = 112\)A1cao 112 only (must reject 2nd answer if found); an answer of 112 only scores M1A1
Part (d):
AnswerMarks Guidance
Answer/WorkingMark Guidance
\(H_0: \lambda = 7 \quad H_1: \lambda > 7\)B1 Both hypotheses correct in terms of \(\lambda\) or \(\mu\) [using \(p\) scores B0]
\(P(X \geq 15) = 1 - P(X \leq 14)\), \(P(X \geq 14) = 0.0128\)M1 For \(1-P(X \leq 14)\) or for CR: one of \(P(X \geq 14)=0.0128\) or \(P(X \geq 15)=0.0057\)
\(= 1 - 0.9943\), \(P(X \geq 15) = 0.0057\)
\(= 0.0057\), CR \(X \geq 15\)A1 awrt 0.0057 or correct CR allow \(X > 14\)
Reject \(H_0\), in the CR, SignificantdM1 (dep on 1st M1) A correct non-contextual statement consistent with their prob and 0.01
There is evidence that the number of water fleas per 100 ml of the pond water has increasedA1 All previous marks must be awarded. Correct context, conclusion with increase(oe) and fleas
# Question 3:

## Part (a):
| Answer/Working | Mark | Guidance |
|---|---|---|
| $P(X \neq 4) = 1 - P(X=4)$, $= 1 - \frac{e^{-7}7^4}{4!}$ or $1-(0.1730-0.0818)$ | M1 | For $1-P(X=4)$ or $1-P(X \leq 4)+P(X \leq 3)$ oe |
| $= 0.90877\ldots$ | A1 awrt 0.909 | |

## Part (b):
| Answer/Working | Mark | Guidance |
|---|---|---|
| $P(Y=1) = (1-\text{"0.90877..."})(\text{"0.90877..."})^4 \times {}^5C_1$ | M1M1 | $(1-\text{"their 0.909"})^1(\text{"their 0.909"})^4$ or $(1-\text{"their 0.909"})(\text{"their 0.909"})^4$ allow values to 2s.f.; $P(Y=1)=(1-\text{"their 0.909"})(\text{"their 0.909"})^4 \times {}^5C_1$ allow values to 2s.f. |
| $= 0.311\ldots$ | A1 | awrt 0.312 or awrt 0.311 |

## Part (c)(i):
| Answer/Working | Mark | Guidance |
|---|---|---|
| $\lambda = 0.07n$ | B1 | Writing or using mean as $0.07n$ |
| $A \sim N(0.07n, 0.07n)$ | M1 | Normal with mean = variance which must be in terms of $n$ |
| $\frac{3.5 - \text{"0.07n"}}{\sqrt{\text{"0.07n"}}}$ | M1 | Standardising with their mean and their $\sqrt{\text{var}}$; allow 2.5, 3, 3.5, 4, 4.5 (correct standardisation implies B1M1M1) |
| $\frac{3.5-0.07n}{\sqrt{0.07n}} = -1.55$ or $\text{"0.07n"} - (1.55\sqrt{0.07})\sqrt{n} - 3.5 = 0$ | B1 | Their standardisation $= \pm 1.55$ |
| $n - \left(\frac{1.55}{0.07}\sqrt{0.07}\right)\sqrt{n} - \frac{3.5}{0.07} = 0 \Rightarrow n - 1.55\sqrt{\frac{n}{0.07}} - 50 = 0$ | A1cso | Must come from compatible signs; need at least one step between standardisation indicating division by 0.07 and correct equation |

## Part (c)(ii):
| Answer/Working | Mark | Guidance |
|---|---|---|
| $\sqrt{n} = \frac{\frac{1.55}{\sqrt{0.07}} \pm \sqrt{\left(\frac{1.55}{\sqrt{0.07}}\right)^2 + 4\times 50}}{2} = \text{awrt } {-4.72}\ldots \text{ or awrt } 10.6\ldots\ (4\sqrt{7})$ | M1 | Correct method to solve given quadratic or sight of awrt $-4.72$ or awrt $10.6$ |
| $n = 112$ | A1cao | 112 only (must reject 2nd answer if found); an answer of 112 only scores M1A1 |

## Part (d):
| Answer/Working | Mark | Guidance |
|---|---|---|
| $H_0: \lambda = 7 \quad H_1: \lambda > 7$ | B1 | Both hypotheses correct in terms of $\lambda$ or $\mu$ [using $p$ scores B0] |
| $P(X \geq 15) = 1 - P(X \leq 14)$, $P(X \geq 14) = 0.0128$ | M1 | For $1-P(X \leq 14)$ or for CR: one of $P(X \geq 14)=0.0128$ or $P(X \geq 15)=0.0057$ |
| $= 1 - 0.9943$, $P(X \geq 15) = 0.0057$ | |  |
| $= 0.0057$, CR $X \geq 15$ | A1 | awrt 0.0057 or correct CR allow $X > 14$ |
| Reject $H_0$, in the CR, Significant | dM1 | (dep on 1st M1) A correct non-contextual statement consistent with their prob and 0.01 |
| There is evidence that the number of water fleas per 100 ml of the pond water has **increased** | A1 | All previous marks must be awarded. Correct context, conclusion with increase(oe) and fleas |

---
3. The number of water fleas, in 100 ml of pond water, has a Poisson distribution with mean 7
\begin{enumerate}[label=(\alph*)]
\item Find the probability that a sample of 100 ml of the pond water does not contain exactly 4 water fleas.

Aja collects 5 separate samples, each of 100 ml , of the pond water.
\item Find the probability that exactly 1 of these samples contains exactly 4 water fleas.

Using a normal approximation, the probability that more than 3 water fleas will be found in a random sample of $n \mathrm { ml }$ of the pond water is 0.9394 correct to 4 significant figures.
\item \begin{enumerate}[label=(\roman*)]
\item Show that $n - 1.55 \sqrt { \frac { n } { 0.07 } } - 50 = 0$
\item Hence find the value of $n$

After the pond has been cleaned, the number of water fleas in a 100 ml random sample of the pond water is 15
\end{enumerate}\item Using a suitable test, at the $1 \%$ level of significance, assess whether or not there is evidence that the number of water fleas per 100 ml of the pond water has increased. State your hypotheses clearly.

\includegraphics[max width=\textwidth, alt={}, center]{f63c39df-cfc9-4a6b-838d-67613710b0ce-11_2255_50_314_34}\\

\begin{center}
\begin{tabular}{|l|l|l|}
\hline
VIXV SIHIANI III IM IONOO & VIAV SIHI NI JYHAM ION OO & VI4V SIHI NI JLIYM ION OO \\
\hline
\end{tabular}
\end{center}
\end{enumerate}

\hfill \mbox{\textit{Edexcel S2 2021 Q3 [17]}}