OCR MEI S1 2011 June — Question 8 18 marks

Exam BoardOCR MEI
ModuleS1 (Statistics 1)
Year2011
SessionJune
Marks18
PaperDownload PDF ↗
Mark schemeDownload PDF ↗
TopicData representation
TypeDraw cumulative frequency graph from frequency table (equal class widths)
DifficultyModerate -0.3 This is a standard S1 statistics question covering routine cumulative frequency diagrams, quartiles, outlier detection using the 1.5×IQR rule, and basic probability calculations. All techniques are textbook exercises requiring straightforward application of learned methods with no novel problem-solving, making it slightly easier than average for A-level.
Spec2.02a Interpret single variable data: tables and diagrams2.02f Measures of average and spread2.02g Calculate mean and standard deviation2.02h Recognize outliers5.01a Permutations and combinations: evaluate probabilities

The heating quality of the coal in a sample of 50 sacks is measured in suitable units. The data are summarised below.
Heating quality (\(x\))9.1 \(\leqslant x <\) 9.39.3 \(< x \leqslant\) 9.59.5 \(< x \leqslant\) 9.79.7 \(< x \leqslant\) 9.99.9 \(< x \leqslant\) 10.1
Frequency5715167
  1. Draw a cumulative frequency diagram to illustrate these data. [5]
  2. Use the diagram to estimate the median and interquartile range of the data. [3]
  3. Show that there are no outliers in the sample. [3]
  4. Three of these 50 sacks are selected at random. Find the probability that
    1. in all three, the heating quality \(x\) is more than 9.5, [3]
    2. in at least two, the heating quality \(x\) is more than 9.5. [4]

(i)
AnswerMarks Guidance
Answer/Working:B1 for cumulative frequencies; G1 for scales; G1 for labels; G1 for points (Provided plotted at correct UCB positions); G1 for joining points 5
Guidance: May be implied from graph. Condone omission of 0 at this stage. Linear horizontal scale. Linear vertical scale: 0 to 50 (no inequality scales - Not even <9.1, <9.3, <9.5 ...). Heating quality or \(x\) and Cumulative frequency or just CF or similar but not just frequency or fd nor cumulative fd. Plotted as (UCB, their cf). Ignore (9.1,0) at this stage. No midpoint or LCB plots. Plotted within ½ small square. For joining all of 'their points' (line or smooth curve) AND now including (9.1,0) dep on previous G1. Mid point or LCB plots may score first three marks. Can get up to 3/5 for cum freq bars. Allow full credit if axes reversed correctly
Lines of best fit could attract max 4 out of 5.
(ii)
Answer/Working: Median = 9.67
AnswerMarks
Marks: B1 FT; Allow answers between 9.66 and 9.68 without checking curve. Otherwise check curve.3
Guidance: Based on 25th to 26th value on a cumulative frequency graph. If their mid-point plot (not LCB's) approx 9.57 for m.p. plot. Allow 9.56 to 9.58 without checking. B0 for interpolation
Question 8 continued
Answer/Working: \(Q1 = 9.51\), \(Q3 = 9.83\). Inter-quartile range = \(9.83 - 9.51 = 0.32\)
AnswerMarks
Marks: B1 FT for Q3 or Q1; B1 FT for IQR providing both Q1 and Q3 are correct. Allow answers between 9.50 and 9.52 and between 9.82 and 9.84 without checking curve. Otherwise check curve.3
Guidance: Based on 12th to 13th and 37th to 38th values on a cumulative frequency graph. If their mid-point plot (not LCB's) approx Q1 = 9.42; Q3 = 9.73. Allow 9.41 to 9.43 and 9.72 to 9.74 without checking. B0 for interpolation. Allow correct IQR from graph if quartiles not stated. Lines of best fit: B0 B0 B0 here.
(iii)
Answer/Working: Lower limit \(9.51 - 1.5 \times 0.32 = 9.03\). Upper limit \(9.83 + 1.5 \times 0.32 = 10.31\). Thus there are no outliers in the sample.
AnswerMarks
Marks: B1 FT their \(Q_1\), IQR; B1 FT their \(Q_3\), IQR; E1; NB E mark dep on both B marks3
Guidance: Any use of median ± 1.5 IQR scores B0 B0 E0. If FT leads to limits above 9.1 or below 10.1 then E0. No marks for ± 2 or 3 IQR. In this part FT their values from (ii) if sensibly obtained (eg from LCB plot) or lines of best fit, but not from location ie 12.5, 37.5 or cumulative fx's or similar. For use of mean ± 2s, Mean = 9.652, s = 0.235, Limits 9.182, 10.122 gets M1 for correct lower limit, M1 for correct upper limit, zero otherwise, but E0 since there could be outliers using this definition
(iv)
Answer/Working: (A) \(P(\text{All 3 more than 9.5}) = \frac{38}{50} \times \frac{37}{49} \times \frac{36}{48} = 0.4304\) (=\(50616/117600 = 2109/4900\))
AnswerMarks
Marks: M1 for 38/50 × (triple product); M1 for product of remaining fractions; A1 CAO3
Guidance: (38/50)³ which gives answer 0.4389 scores M1M0A0 so watch for this. M0M0A0 for binomial probability including 0.76¹⁰⁰ but ³\(C_0×0.24^0×0.76^3\) still scores M1 (\(k/50\))³ for values of \(k\) other than 38 scores M0M0A0. M1M0A0 for binomial probability including 0.76¹⁰⁰ but ³\(C_0×0.24^0×0.76^3\) still scores M1 (\(k/50\))³ for values of \(k\) other than 38 scores M0M0A0. \(\frac{k}{50} \cdot \frac{(k-1)}{49} \cdot \frac{(k-2)}{48}\) for values of \(k\) other than 38 scores M1M0A0
Answer/Working: (B) \(P(\text{At least 2 more than 9.5}) = 3 \times \frac{38}{50} \times \frac{37}{49} \times \frac{12}{48} + 0.4304 = 3 \times 0.1435 + 0.4304 = 0.4304 + 0.4304 = 0.861\) (=\(101232/117600 = 4218/4900 = 2109/2450\))
AnswerMarks
Marks: M1 for product of 3 correct fractions seen; M1 for 3 × a sensible triple or sum of 3 sensible triples; M1 indep for + 0.4304; FT (providing it is a probability); A1 CAO4
Guidance: Allow unsimplified fraction as final answer 50616/117600. Or (38/12)/50=0.4304 gets first two M1M1's. SC1 for \(3 \times \frac{38}{50} \times \frac{38}{50} \times \frac{12}{50}\) or other sensible triple and SC2 if this + their 0.4304 (= 0.8549). Allow 0.86 or 2109/2450 or 4218/4900, but only M3A0 for other unsimplified fractions. Use of 1 – method 'with replacement' SC1 for \(3 \times \frac{12}{50} \times \frac{12}{50} \times \frac{38}{50}\) SC2 for whole of \(1 - 3 \times \frac{12}{50} \times \frac{12}{50} \times \frac{38}{50} - \frac{12}{50} \times \frac{12}{50} \times \frac{12}{50}\) (= 1 – (0.1313 + 0.0138) = 1 – 0.1451 = 0.8549)
NOTE RE OVER-SPECIFICATION OF ANSWERS
If answers are grossly over-specified (see instruction 8), deduct the final answer mark in every case. Probabilities should also be rounded to a sensible degree of accuracy. In general final non probability answers should not be given to more than 4 significant figures. Allow probabilities given to 5 sig fig. In general accept answers which are correct to 3 significant figures when given to 4 or 5 significant figures.
## (i)
**Answer/Working:** | B1 for cumulative frequencies; G1 for scales; G1 for labels; G1 for points (Provided plotted at correct UCB positions); G1 for joining points | 5

**Guidance:** May be implied from graph. Condone omission of 0 at this stage. Linear horizontal scale. Linear vertical scale: 0 to 50 (no inequality scales - Not even <9.1, <9.3, <9.5 ...). Heating quality or $x$ and Cumulative frequency or just CF or similar but not just frequency or fd nor cumulative fd. Plotted as (UCB, their cf). Ignore (9.1,0) at this stage. No midpoint or LCB plots. Plotted within ½ small square. For joining all of 'their points' (line or smooth curve) AND now including (9.1,0) dep on previous G1. Mid point or LCB plots may score first three marks. Can get up to 3/5 for cum freq bars. Allow full credit if axes reversed correctly

Lines of best fit could attract max 4 out of 5.

## (ii)
**Answer/Working:** Median = 9.67

**Marks:** B1 FT; Allow answers between 9.66 and 9.68 without checking curve. Otherwise check curve. | 3

**Guidance:** Based on 25th to 26th value on a cumulative frequency graph. If their mid-point plot (not LCB's) approx 9.57 for m.p. plot. Allow 9.56 to 9.58 without checking. B0 for interpolation

---

# Question 8 continued

**Answer/Working:** $Q1 = 9.51$, $Q3 = 9.83$. Inter-quartile range = $9.83 - 9.51 = 0.32$

**Marks:** B1 FT for Q3 or Q1; B1 FT for IQR providing both Q1 and Q3 are correct. Allow answers between 9.50 and 9.52 and between 9.82 and 9.84 without checking curve. Otherwise check curve. | 3

**Guidance:** Based on 12th to 13th and 37th to 38th values on a cumulative frequency graph. If their mid-point plot (not LCB's) approx Q1 = 9.42; Q3 = 9.73. Allow 9.41 to 9.43 and 9.72 to 9.74 without checking. B0 for interpolation. Allow correct IQR from graph if quartiles not stated. Lines of best fit: B0 B0 B0 here.

## (iii)
**Answer/Working:** Lower limit $9.51 - 1.5 \times 0.32 = 9.03$. Upper limit $9.83 + 1.5 \times 0.32 = 10.31$. Thus there are no outliers in the sample.

**Marks:** B1 FT their $Q_1$, IQR; B1 FT their $Q_3$, IQR; E1; NB E mark dep on both B marks | 3

**Guidance:** Any use of median ± 1.5 IQR scores B0 B0 E0. If FT leads to limits above 9.1 or below 10.1 then E0. No marks for ± 2 or 3 IQR. In this part FT their values from (ii) if sensibly obtained (eg from LCB plot) or lines of best fit, but not from location ie 12.5, 37.5 or cumulative fx's or similar. For use of mean ± 2s, Mean = 9.652, s = 0.235, Limits 9.182, 10.122 gets M1 for correct lower limit, M1 for correct upper limit, zero otherwise, but E0 since there could be outliers using this definition

## (iv)
**Answer/Working:** (A) $P(\text{All 3 more than 9.5}) = \frac{38}{50} \times \frac{37}{49} \times \frac{36}{48} = 0.4304$ (=$50616/117600 = 2109/4900$)

**Marks:** M1 for 38/50 × (triple product); M1 for product of remaining fractions; A1 CAO | 3

**Guidance:** (38/50)³ which gives answer 0.4389 scores M1M0A0 so watch for this. M0M0A0 for binomial probability including 0.76¹⁰⁰ but ³$C_0×0.24^0×0.76^3$ still scores M1 ($k/50$)³ for values of $k$ other than 38 scores M0M0A0. M1M0A0 for binomial probability including 0.76¹⁰⁰ but ³$C_0×0.24^0×0.76^3$ still scores M1 ($k/50$)³ for values of $k$ other than 38 scores M0M0A0. $\frac{k}{50} \cdot \frac{(k-1)}{49} \cdot \frac{(k-2)}{48}$ for values of $k$ other than 38 scores M1M0A0

**Answer/Working:** (B) $P(\text{At least 2 more than 9.5}) = 3 \times \frac{38}{50} \times \frac{37}{49} \times \frac{12}{48} + 0.4304 = 3 \times 0.1435 + 0.4304 = 0.4304 + 0.4304 = 0.861$ (=$101232/117600 = 4218/4900 = 2109/2450$)

**Marks:** M1 for product of 3 correct fractions seen; M1 for 3 × a sensible triple or sum of 3 sensible triples; M1 indep for + 0.4304; FT (providing it is a probability); A1 CAO | 4

**Guidance:** Allow unsimplified fraction as final answer 50616/117600. Or (38/12)/50=0.4304 gets first two M1M1's. SC1 for $3 \times \frac{38}{50} \times \frac{38}{50} \times \frac{12}{50}$ or other sensible triple and SC2 if this + their 0.4304 (= 0.8549). Allow 0.86 or 2109/2450 or 4218/4900, but only M3A0 for other unsimplified fractions. Use of 1 – method 'with replacement' SC1 for $3 \times \frac{12}{50} \times \frac{12}{50} \times \frac{38}{50}$ SC2 for whole of $1 - 3 \times \frac{12}{50} \times \frac{12}{50} \times \frac{38}{50} - \frac{12}{50} \times \frac{12}{50} \times \frac{12}{50}$ (= 1 – (0.1313 + 0.0138) = 1 – 0.1451 = 0.8549)

---

# NOTE RE OVER-SPECIFICATION OF ANSWERS

If answers are grossly over-specified (see instruction 8), deduct the final answer mark in every case. Probabilities should also be rounded to a sensible degree of accuracy. In general final non probability answers should not be given to more than 4 significant figures. Allow probabilities given to 5 sig fig. In general accept answers which are correct to 3 significant figures when given to 4 or 5 significant figures.
The heating quality of the coal in a sample of 50 sacks is measured in suitable units. The data are summarised below.

\begin{tabular}{|c|c|c|c|c|c|}
\hline
Heating quality ($x$) & 9.1 $\leqslant x <$ 9.3 & 9.3 $< x \leqslant$ 9.5 & 9.5 $< x \leqslant$ 9.7 & 9.7 $< x \leqslant$ 9.9 & 9.9 $< x \leqslant$ 10.1 \\
\hline
Frequency & 5 & 7 & 15 & 16 & 7 \\
\hline
\end{tabular}

\begin{enumerate}[label=(\roman*)]
\item Draw a cumulative frequency diagram to illustrate these data. [5]
\item Use the diagram to estimate the median and interquartile range of the data. [3]
\item Show that there are no outliers in the sample. [3]
\item Three of these 50 sacks are selected at random. Find the probability that
\begin{enumerate}[label=(\Alph*)]
\item in all three, the heating quality $x$ is more than 9.5, [3]
\item in at least two, the heating quality $x$ is more than 9.5. [4]
\end{enumerate}
\end{enumerate}

\hfill \mbox{\textit{OCR MEI S1 2011 Q8 [18]}}