| Exam Board | Edexcel |
|---|---|
| Module | S2 (Statistics 2) |
| Year | 2021 |
| Session | January |
| Marks | 14 |
| Paper | Download PDF ↗ |
| Mark scheme | Download PDF ↗ |
| Topic | Hypothesis test of binomial distributions |
| Type | Multiple binomial probability calculations |
| Difficulty | Standard +0.3 This is a standard S2 question covering routine binomial distribution calculations and a straightforward hypothesis test. Parts (a)-(c) require basic binomial probability calculations, part (d) uses normal approximation (standard S2 technique), and part (e) is a textbook one-tailed binomial test. All techniques are directly taught with no novel problem-solving required, making it slightly easier than average. |
| Spec | 2.04b 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-tail |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Mark | Guidance |
| \(B(30, 0.05)\) | B1 | Must include B(inomial), \(n = 25\) and \(p = 0.05\). Do not allow \(p = 0.95\) in part (a) |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Mark | Guidance |
| The probability of an oyster surviving/not surviving is constant | B1 | For either correct assumption in context. Ignore extraneous non-contradicting comments |
| The survival of each oyster is independent of the others | B1 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Mark | Guidance |
| \(^{30}C_{24}(0.05)^6(0.95)^{24}\) oe | M1 | Allow \(^{30}C_6\) oe or \(P(X \leq 6) - P(X \leq 5)\) with one correct probability |
| \(= 0.002708\ldots\) awrt \(0.0027\) | A1 | awrt \(0.0027\) (correct answer scores 2 out of 2) |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Mark | Guidance |
| \(P(Y \geq 3) = 1 - P(Y \leq 2)\) from \(Y \sim B(30, 0.05)\) or \(P(X \leq 27)\) from \(X \sim B(30, 0.95)\) | M1 | Writing/using \(1 - P(Y \leq 2)\) with \(B(30, 0.05)\) or writing/using \(P(X \leq 27)\) with \(B(30, 0.95)\) |
| \(= 1 - 0.8122\) | ||
| \(= 0.1878\) awrt \(0.188\) | A1 | awrt \(0.188\) (correct answer scores 2 out of 2) |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Mark | Guidance |
| \(A \sim Po(10)\) | B1 | Writing or using \(Po(10)\) — sight of \(0.1301\) or \(0.8699\) can imply this mark |
| \(P(A \geq n) > 0.8\) | ||
| \(P(A \leq n-1) < 0.2\) or \(P(A \leq 6) = 0.1301\ldots\) awrt \(0.13\) or \(P(A \geq 7) = 0.8699\ldots\) awrt \(0.87\) | M1 | Allow \(P(A < n) < 0.2\) or \(P(A < 7) =\) awrt \(0.13\) or \(P(A > 6) =\) awrt \(0.87\) |
| \(n = 7\) | A1cao | \(n = 7\) which must come from use of \(Po(10)\) or \(N(10, 9.5)\) |
| Note: Use of normal approx. with \(\mu = 10\) and \(\sigma^2 = 9.5\) leading to \(n < 7.4\ldots\) can score M1; Exact binomial gives \(P(A \leq 6) = 0.14\) / \(P(A \geq 7) = 0.86\) scores B0M0A0 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer | Mark | Guidance |
| \(H_0: p = 0.05\), \(H_1: p > 0.05\) | B1 | Both hypotheses correct (allow use of \(p\) or \(\pi\)). Allow \(H_0: p = 0.95\), \(H_1: p < 0.95\) |
| Using \(C \sim B(25, 0.05)\) and \(P(C \geq 4)\) \ | Using \(D \sim B(25, 0.95)\) and \(P(D \leq 21)\) | M1 |
| \(P(C \geq 4) = 0.0341\) / CR \(C \geq 4\) \ | \(P(D \leq 21) = 0.0341\) / CR \(D \leq 21\) | A1 |
| Evidence to reject \(H_0\), in the CR, significant | dM1 | (dep on 1st M1) A correct non-contextual statement (do not allow contradicting non-contextual comments) which is consistent with their prob and \(0.05\) (If not stated, may be implied by A1) |
| There is evidence that the proportion of oysters not surviving has increased/Jim's belief is supported | A1cso | All previous marks must be awarded. Correct contextual conclusion with bold words (oe) |
| SC: 2-tail — Use of two-tailed test can score max: B1M1A1M1A0, but must not reject \(H_0\) for 2nd M1 |
# Question 1:
## Part (a)
| Answer | Mark | Guidance |
|--------|------|----------|
| $B(30, 0.05)$ | B1 | Must include B(inomial), $n = 25$ and $p = 0.05$. Do not allow $p = 0.95$ in part (a) |
## Part (b)
| Answer | Mark | Guidance |
|--------|------|----------|
| The **probability** of an oyster surviving/not surviving is **constant** | B1 | For either correct assumption in context. Ignore extraneous non-contradicting comments |
| The survival of each oyster is **independent** of the others | B1 | |
## Part (c)(i)
| Answer | Mark | Guidance |
|--------|------|----------|
| $^{30}C_{24}(0.05)^6(0.95)^{24}$ oe | M1 | Allow $^{30}C_6$ oe or $P(X \leq 6) - P(X \leq 5)$ with one correct probability |
| $= 0.002708\ldots$ awrt $0.0027$ | A1 | awrt $0.0027$ (correct answer scores 2 out of 2) |
## Part (c)(ii)
| Answer | Mark | Guidance |
|--------|------|----------|
| $P(Y \geq 3) = 1 - P(Y \leq 2)$ from $Y \sim B(30, 0.05)$ or $P(X \leq 27)$ from $X \sim B(30, 0.95)$ | M1 | Writing/using $1 - P(Y \leq 2)$ with $B(30, 0.05)$ or writing/using $P(X \leq 27)$ with $B(30, 0.95)$ |
| $= 1 - 0.8122$ | | |
| $= 0.1878$ awrt $0.188$ | A1 | awrt $0.188$ (correct answer scores 2 out of 2) |
## Part (d)
| Answer | Mark | Guidance |
|--------|------|----------|
| $A \sim Po(10)$ | B1 | Writing or using $Po(10)$ — sight of $0.1301$ or $0.8699$ can imply this mark |
| $P(A \geq n) > 0.8$ | | |
| $P(A \leq n-1) < 0.2$ or $P(A \leq 6) = 0.1301\ldots$ awrt $0.13$ or $P(A \geq 7) = 0.8699\ldots$ awrt $0.87$ | M1 | Allow $P(A < n) < 0.2$ or $P(A < 7) =$ awrt $0.13$ or $P(A > 6) =$ awrt $0.87$ |
| $n = 7$ | A1cao | $n = 7$ which must come from use of $Po(10)$ or $N(10, 9.5)$ |
| **Note:** Use of normal approx. with $\mu = 10$ and $\sigma^2 = 9.5$ leading to $n < 7.4\ldots$ can score M1; Exact binomial gives $P(A \leq 6) = 0.14$ / $P(A \geq 7) = 0.86$ scores B0M0A0 | | |
## Part (e)
| Answer | Mark | Guidance |
|--------|------|----------|
| $H_0: p = 0.05$, $H_1: p > 0.05$ | B1 | Both hypotheses correct (allow use of $p$ or $\pi$). Allow $H_0: p = 0.95$, $H_1: p < 0.95$ |
| Using $C \sim B(25, 0.05)$ and $P(C \geq 4)$ \| Using $D \sim B(25, 0.95)$ and $P(D \leq 21)$ | M1 | Using $B(25, 0.05)$ and writing/using $P(C \geq 4)$ or if CR given $P(C \geq 3)$ using $B(25, 0.95)$ and writing/using $P(D \leq 21)$ or if CR given $P(D \leq 20)$ |
| $P(C \geq 4) = 0.0341$ / CR $C \geq 4$ \| $P(D \leq 21) = 0.0341$ / CR $D \leq 21$ | A1 | Correct probability to 3sf (must not go on and give incorrect CR) or correct CR (ignore upper tail) |
| Evidence to reject $H_0$, in the CR, significant | dM1 | (dep on 1st M1) A correct non-contextual statement (do not allow contradicting non-contextual comments) which is consistent with their prob and $0.05$ (If not stated, may be implied by A1) |
| There is evidence that the proportion of **oysters** not surviving has **increased**/Jim's **belief** is supported | A1cso | All previous marks must be awarded. Correct contextual conclusion with bold words (oe) |
| **SC: 2-tail** — Use of two-tailed test can score max: B1M1A1M1A0, but must **not reject** $H_0$ for 2nd M1 | | |
---
\begin{enumerate}
\item Jim farms oysters in a particular lake. He knows from past experience that $5 \%$ of young oysters do not survive to be harvested.
\end{enumerate}
In a random sample of 30 young oysters, the random variable $X$ represents the number that do not survive to be harvested.\\
(a) Write down a suitable model for the distribution of $X$.\\
(b) State an assumption that has been made for the model in part (a).\\
(c) Find the probability that\\
(i) exactly 24 young oysters do survive to be harvested,\\
(ii) at least 3 young oysters do not survive to be harvested.
A second random sample, of 200 young oysters, is taken. The probability that at least $n$ of these young oysters do not survive to be harvested is more than 0.8\\
(d) Using a suitable approximation, find the maximum value of $n$.
Jim believes that the level of salt in the lake water has changed and it has altered the survival rate of his oysters. He takes a random sample of 25 young oysters and places them in the lake.\\
When Jim harvests the oysters, he finds that 21 do survive to be harvested.\\
(e) Use a suitable test, at the $5 \%$ level of significance, to assess whether or not there is evidence that the proportion of oysters not surviving to be harvested is more than $5 \%$. State your hypotheses clearly.
\hfill \mbox{\textit{Edexcel S2 2021 Q1 [14]}}