AQA Further AS Paper 2 Statistics 2020 June — Question 8 9 marks

Exam BoardAQA
ModuleFurther AS Paper 2 Statistics (Further AS Paper 2 Statistics)
Year2020
SessionJune
Marks9
PaperDownload PDF ↗
Mark schemeDownload PDF ↗
TopicPoisson distribution
TypePoisson hypothesis test
DifficultyChallenging +1.2 This is a hypothesis test for Poisson distributions requiring students to set up a two-tailed test with combined parameter λ=25, calculate P(X≤16), and find the actual significance level. While it involves multiple steps and understanding of Type I error, the mechanics are straightforward applications of standard Further Maths statistics techniques without requiring novel insight or complex problem-solving.
Spec5.02j Poisson formula: P(X=x) = e^(-lambda)*lambda^x/x!5.02k Calculate Poisson probabilities5.05b Unbiased estimates: of population mean and variance

There are two hospitals in a city. Over a period of time, the first hospital recorded an average of 20 births a day. Over the same period of time, the second hospital recorded an average of 5 births a day. Stuart claims that birth rates in the hospitals have changed over time. On a randomly chosen day, he records a total of 16 births from the two hospitals.
  1. Investigate Stuart's claim, using a suitable test at the 5% level of significance. [6 marks]
  2. For a test of the type carried out in part (a), find the probability of making a Type I error, giving your answer to two significant figures. [3 marks]

Question 8:

AnswerMarks Guidance
8(a)States both hypotheses using
correct language.2.5 B1
0
H : λ ≠ 25
1
X ~ Po(25)
P(X ≤ 16)
= 0.038 > 0.025
Accept H
0
There is no significant evidence to
suggest that the total birth rate in
the two hospitals has changed.
Selects and uses a Poisson model
with λ = 25 to find
P(number of births ≤ 16) or
P(number of births < 16) or the
lower or upper part of the critical
AnswerMarks Guidance
region (X ≤ 15 or X ≥ 36)3.3 M1
Obtains 0.038 (AWRT) or both
parts of the critical region (X ≤ 15
AnswerMarks Guidance
and X ≥ 36)3.4 A1
Evaluates the Poisson model by
comparing ‘their’ p-value with 0.025
or the sample value with ‘their’
AnswerMarks Guidance
critical region3.5a R1
Infers H not rejected.
0
FT comparison of ‘their’ p-value
AnswerMarks Guidance
with 0.025 or 0.052.2b E1F
Concludes in context.
(The conclusion must not be
definite.)
FT their incorrect rejection of H if
0
stated or ‘their’ comparison of p-
AnswerMarks Guidance
value and 0.025/0.05 if not3.2a E1
QMarking Instructions AO

AnswerMarks
8(b)Selects a method to determine the
probability of a Type I error by
considering the probability of the
upper or lower part of the critical
region e.g. by finding P(X ≤ 15) or
P(X ≤ 16)
FT one tailed test used in 8(a).
FT their Po(20) or Po(5) if used in
AnswerMarks Guidance
8(a).3.1a M1
P(X ≤ 16) = 0.03775
P(X ≥ 35) = 0.03384
P(X ≥ 36) = 0.02246
P(Type I error) =
0.02229 + 0.02246
= 0.045
Develops their method by
considering the other part of the
critical region e.g. by finding
P(X ≥ 35) or P(X ≥ 36)
Do not FT one tailed test used in
8(a).
FT their Po(20) or Po(5) if used in
AnswerMarks Guidance
8(a).3.1a M1
Obtains P(Type I error)
= AWRT 0.045
FT one tailed test used in 8(a) .
FT their Po(20) or Po(5) if used in
AnswerMarks Guidance
8(a).1.1b A1
Total9
Paper total40
Question 8:
--- 8(a) ---
8(a) | States both hypotheses using
correct language. | 2.5 | B1 | H : λ = 25
0
H : λ ≠ 25
1
X ~ Po(25)
P(X ≤ 16)
= 0.038 > 0.025
Accept H
0
There is no significant evidence to
suggest that the total birth rate in
the two hospitals has changed.
Selects and uses a Poisson model
with λ = 25 to find
P(number of births ≤ 16) or
P(number of births < 16) or the
lower or upper part of the critical
region (X ≤ 15 or X ≥ 36) | 3.3 | M1
Obtains 0.038 (AWRT) or both
parts of the critical region (X ≤ 15
and X ≥ 36) | 3.4 | A1
Evaluates the Poisson model by
comparing ‘their’ p-value with 0.025
or the sample value with ‘their’
critical region | 3.5a | R1
Infers H not rejected.
0
FT comparison of ‘their’ p-value
with 0.025 or 0.05 | 2.2b | E1F
Concludes in context.
(The conclusion must not be
definite.)
FT their incorrect rejection of H if
0
stated or ‘their’ comparison of p-
value and 0.025/0.05 if not | 3.2a | E1
Q | Marking Instructions | AO | Marks | Typical Solution
--- 8(b) ---
8(b) | Selects a method to determine the
probability of a Type I error by
considering the probability of the
upper or lower part of the critical
region e.g. by finding P(X ≤ 15) or
P(X ≤ 16)
FT one tailed test used in 8(a).
FT their Po(20) or Po(5) if used in
8(a). | 3.1a | M1 | P(X ≤ 15) = 0.02229
P(X ≤ 16) = 0.03775
P(X ≥ 35) = 0.03384
P(X ≥ 36) = 0.02246
P(Type I error) =
0.02229 + 0.02246
= 0.045
Develops their method by
considering the other part of the
critical region e.g. by finding
P(X ≥ 35) or P(X ≥ 36)
Do not FT one tailed test used in
8(a).
FT their Po(20) or Po(5) if used in
8(a). | 3.1a | M1
Obtains P(Type I error)
= AWRT 0.045
FT one tailed test used in 8(a) .
FT their Po(20) or Po(5) if used in
8(a). | 1.1b | A1
Total | 9
Paper total | 40
There are two hospitals in a city.

Over a period of time, the first hospital recorded an average of 20 births a day.

Over the same period of time, the second hospital recorded an average of 5 births a day.

Stuart claims that birth rates in the hospitals have changed over time.

On a randomly chosen day, he records a total of 16 births from the two hospitals.

\begin{enumerate}[label=(\alph*)]
\item Investigate Stuart's claim, using a suitable test at the 5% level of significance.
[6 marks]

\item For a test of the type carried out in part (a), find the probability of making a Type I error, giving your answer to two significant figures.
[3 marks]
\end{enumerate}

\hfill \mbox{\textit{AQA Further AS Paper 2 Statistics 2020 Q8 [9]}}