University of Pretoria

Department of Industrial and Systems Engineering

Engineering Statistics BES 220

Internal examiner: Mrs. T Malatji

1 Instructions:
This test consists of 8 multiple choice questions (Section A) and 15 questions in the long section (Section B).
In total the question paper contains 23 questions.
Answer all the questions in the provided OCR form.
The assessment ID is: 2 0 2 2 B E S 2 2 0 S 0 2
Ensure that you provide all the necessary details on both sides of the OCR form.
Ensure that you provide numeric answers in the numeric field of the OCR form.
Ensure that you provide text answers in the text field of the OCR form.
The OCR line number must correspond with the question number being answered. (e.g. Q5 on line 05 in OCR form)
You may use the provided answer book for scribbling during the test.
You must submit the OCR form after the test. You may keep the answer book used for scribbling.
The decimal places required for each question is given in boldface.
Please ensure you answer to the correct decimal place.
Marks for each question are indicated in a square frame next to the question in the left margin.
The test is closed book. No material or media may be consulted.
No programmable calculators are allowed. Standard calculators may be used.
You may keep this question paper after completion of the test.

Complete all 23 questions for 50 marks

Total time: 90 minutes

A: Multiple choice questions

Instructions
Select the most appropriate answer for the multiple questions.

Questions
2 1. The standard deviation of the weights of elephants is known to be approximately 15 kg. Fifty newborn
elephants are weighed. The sample mean is 244 kg. The sample standard deviation is 11 kg. A 95%
confidence interval for the mean weight of newborn elephant calves is CI: (239.84, 248.16) is obtained.
What will happen to the confidence interval obtained, if 500 newborn elephants are weighed instead of 50?
A. As the sample size increases, there will be more variability in the mean and so the interval size
increases.
B. As the sample size increases, there will be less variability in the mean and so the
interval size decreases.
C. As the sample size increases, there will be less variability in the mean and so the interval size
increases.
D. As the sample size increases, the standard deviation will increase and so the interval size increases.

2 2. Suppose an airline claims that its flights are consistently on time with an average delay of at most 15
minutes. It claims that the average delay is so consistent that the variance is no more than 150 minutes.
Doubting the consistency part of the claim, a disgruntled traveler calculates the delays for his next 25
flights. The average delay for those 25 flights is 22 minutes with a standard deviation of 15 minutes. If the
traveller tests the airline claim, he will find a chi-square statistic of:
A. 0.24
B. 0.505
C. 36

BES 220 1 of 8 Semester test 2 - Memo – 19 October 2022

 D. 1215
E. 2400

Solution: This is a test of a single variance. The test statistic is given by:

(n − 1)s2
χ2 =
σ2
24(15)2
=
150
= 36

2 3. A study was conducted to test the effectiveness of a juggling class. Before the class started, six subjects
juggled as many balls as they could at once. The class was held the following day and after the class, the
same six subjects juggled as many balls as they could. The differences in the number of balls are calculated.
The differences have a normal distribution. If testing at the 1% significance level what is the appropriate
null and alternative hypotheses?

Subject A B C D E F
Day 1 2 5 3 2 4 3
Day 2 5 6 6 4 5 7
A. H0 : µd ≤ 0
HA : µd > 0
B. H0 : µd ≥ 0
HA : µd < 0
C. H0 : µ1 − µ2 ≤ 0
HA : µ1 − µ2 > 0
D. H0 : µ1 − µ2 ≥ 0
HA : µ1 − µ2 < 0
E. H0 : µ1 − µ2 = 0
HA : µ1 − µ2 > 0

2 4. Which statement is false about the chi-square distribution and the associated tests?
A. The null hypothesis for the test of independence, states that the two factors under test are
independent.
B. The chi-square distribution curve is right skewed and the shape depends on the degrees of freedom.
C. The test of a single variance may be left-, right-, or two-tailed.
D. The distribution of the population under test must be known to use the test of
homogeneity.
E. None of the above.

Solution: A test of homogeneity is used to determine whether two populations come from the same
distribution even if the distribution is unknown.

2 5. A sleeping bag is tested to withstand temperatures of 2 ◦ C. You think the bag cannot stand temperatures
that low. State the Type I error in a complete sentence.
A. The bag cannot withstand 2 ◦ C, and you conclude that it cannot.
B. The bag cannot withstand 2 ◦ C, but you conclude that it can.
C. The bag can withstand 2 ◦ C, and you conclude that it can.
D. None of the above.

BES 220 2 of 8 Semester test 2 - Memo – 19 October 2022

3 6. According to a report on sleep deprivation by the Center for Disease Control and Prevention, the proportion
of Western Cape residents who reported insufficient rest or sleep during each of the preceding 30 days is
8.0%, while this proportion is 8.8% for Gauteng residents. These data are based on simple random samples
of 11,545 Western Cape and 4,691 Gauteng residents. We would like to calculate a 95% confidence interval
for the difference between the proportions of Western Cape and Gauteng residents who are sleep deprived.
Assume the distribution for p′1 − p′2 is approximately normal. What is the margin of error in this case?
(Round your answer to 4 decimal places).
A. 0.0048
B. 0.0080
C. 0.0095
D. 1.9600

Solution: The critical z value or the zα/2 = z0.025 value for a 95% confidence level is 1.96.

M E = z0.025 × SE
s
p′1 (1 − p′1 ) p′2 (1 − p′2 )
= 1.96 × +
n1 n1
s
0.08(1 − 0.08) 0.088(1 − 0.088)
= 1.96 × +
11, 545 4, 691
= 1.96 × 0.0048
= 0.0095

3 7. The length of songs in a collector’s iTunes album collection is uniformly distributed from two to 3.5 minutes.
Suppose we randomly pick five albums from the collection. There are a total of 43 songs on the five albums.
Find the standard deviation for the average song length of the 43 songs. (Round off your answer to 3
decimal places).
A. 0.066
B. 0.194
C. 0.433
D. 2.840
E. None of the above.

Solution:

The distribution for X is given by:

X ∼ U (2, 3.5)
The standard deviation for X is given by:
r
(B − A)2
σ=
12
= 0.433
The standard deviation of the average X̄ therefore is:
σ
SE = √
n
0.433
= √
43
= 0.066

BES 220 3 of 8 Semester test 2 - Memo – 19 October 2022

2 8. The test statistic needed to construct a two-sided confidence interval is given by tα/2 = 2.65. What is the
confidence level and sample size for this statistic?
A. Level 95%, sample size = 14
B. Level 98%, sample size = 14
C. Level 99%, sample size = 14
D. Level 98%, sample size = 13

B: Longer questions
Instructions
Questions in this section are grouped and the text in boldface indicates the question numbers related to the
provided statement.

Use the following information to answer questions 9 to 13.

The article “Analysis of Unwanted Fire Alarm: Case Study” (W. Chow, N. Fong, and C. Ho, Journal of
Architectural Engineering, 1999:62–65) presents a count of the number of a random selection of false alarms at
several sites. The numbers of false alarms each month, divided into those with known causes and those with
unknown causes, are given in the following table. Assume the underlying distribution is normal and that the
alarms for each month are independent.

Months Months
1 2 3 1 2 3
Known 20 13 21 Known 16.26 7.62 18.80
Unknown 12 2 16 Unknown 15.74 7.38 18.20
Observed values for the Expected values for the
number of false alarms number of false alarms.

2 9. Are the conditions met for using the χ2 test of homogeneity? Which statement (or combination of state-
ments) is the most accurate in answering this question?
1. The categories are independent of each other.
2. The population distribution is normal.
3. The observed values are not all ≥ 5.
4. The expected values are ≥ 5.
A. Yes, because of 1, 2 and 4.
B. No, because of 2 and 4.
C. No, because of 3.
D. Yes because of 1 and 4.
2 10. What is the null and alternative hypothesis for this test?
A. H0 : The known and unknown alarm proportions are the same for the different
months.
HA : The known and unknown alarm proportions are not the same for the different
months.
B. H0 : The known and unknown alarm proportions are independent of the month in which an alarm
occurs.
HA : The known and unknown alarm proportions are dependent on the month in which an alarm
occurs.
C. H0 : The known and unknown alarm proportions are not homogeneous across the different months.
HA : The known and unknown alarm proportions are homogeneous across the different months.
D. H0 : p1 ̸= p2 ̸= p3
HA : p1 = p2 = p3

BES 220 4 of 8 Semester test 2 - Memo – 19 October 2022

3 11. What is the test statistic for this test? Round your answer to 4 decimal places.

Solution:
X (observed − expected)2
χ2 =
expected
(20 − 16.26)2 (13 − 7.62)2 (21 − 18.8)2 (12 − 15.74)2 (2 − 7.38)2 (16 − 18.2)2
= + + + + +
16.26 7.62 18.8 15.74 7.38 18.2
= 9.9918

1 12. What is the degrees of freedom for the test? Round off your answer to 0 decimal places.

Solution:

df = k − 1
=3−1
= 2

2 13. What is the p-value for this test? Give the lower bound for the p-value (i.e. p-value < answer), only give
the numeric value. Round off your answer to 4 decimal places.

Solution:
Accept both answers.

0.0050 < p-value < 0.0100

Use the following information to answer questions 14 to 17.

The inside diameter of a randomly selected piston ring is a random variable with mean value 12 cm and
standard deviation 0.04 cm. A sample of 16 rings is inspected to test if they comply with the specification
of 0.1 cm of the diameter 12 cm.
1 14. What is the standard deviation for the sample mean of the sample of 16 rings? (Round off your answer
to 3 decimal places.)

Solution:
σ
SE = √
n
0.04
=√
16
= 0.010

3 15. What is the probability that a random piston ring will have an inside diameter that is within specification?
(Round off your answer to 4 decimal places.)

BES 220 5 of 8 Semester test 2 - Memo – 19 October 2022

 Solution:
We are finding a probability for the random variable X ∼ N (12, 0.04).
The specification under test is ±0.1.

P (11.9 < X < 12.1) = P (X < 12.1) − P (X < 11.9)

   
12.1 − 12 11.9 − 12
=P Z< −P Z <
0.04 0.04
= P (Z < 2.5) − P (Z < −2.5)
= 0.9938 − 0.0062
= 0.9876

2 16. What is the probability that the sample mean of the 16 rings is greater than 12.5 cm? (Round off your
answer to 4 decimal places.)

Solution:
We are finding a probability for the random variable X̄ ∼ N (12, 0.01).

P (X̄ > 12.5) = 1 − P (X̄ < 12.5)

 
12.5 − 12
=1−P Z <
0.01
= 1 − P (Z < 50)
= 1 − 0.9998
= 0.0002

3 17. What is the 99th percentile for the distribution of the sample mean of the 16 rings? (Round off your
answer to 4 decimal places.)

Solution:
We are finding the percentile for the random variable X̄ ∼ N (12, 0.01).

P (X̄ < x̄) = 0.99

This gives us z = 2.33.
x̄ − 12
2.33 =
0.01
x̄ = (2.33)(0.01) + 12
x̄ = 12.0233

Use the following information to answer questions 18 to 23.

A professor at the University of Pretoria recently claimed that the proportion of second year males with
at least one pierced ear is as high as the proportion of second year females. She conducted a survey in her
classes. Out of 107 males, 46 had at least one pierced ear. Out of 92 females, 47 had at least one pierced
ear. She would like to determine if the proportion of males who have piercings is the same as the proportion
of females.

2 18. What is the null and alternative hypothesis for this test? Assume p1 is the proportion of males with at
least one piercing and p2 is the proportion of females with at least one piercing.
A. H0 : p′1 = p′2
HA : p′1 ̸= p′2

BES 220 6 of 8 Semester test 2 - Memo – 19 October 2022

 B. H0 : p1 = p2
HA : p1 < p2
C. H0 : p′1 = p′2
HA : p′1 > p′2
D. H0 : p1 = p2
HA : p1 ̸= p2

3 19. What is the standard error for this test? Round your answer to 4 decimal places.

Solution:

We first have to find the pooled proportion.

x1 + x2
p′ =
n1 + n2
46 + 47
=
107 + 92
= 0.4673
Now we can find the standard error.
s  
′ ′
1 1
SE = p (1 − p ) +
n1 n2
s  
1 1
= 0.4673(1 − 0.4673) +
107 92

= 0.0709

2 20. What is the test statistic for this test? Round your answer to 2 decimal places.

Solution:
(p′1 − p′2 ) − 0
z=
SE
0.43 − 0.5109
=
0.0709
= −1.14

2 21. What is the p-value for this test? Round your answer to 4 decimal places.

Solution:
Remember that this is a two sided test.

p-value = 2 × 0.1271
= 0.2542

2 22. The professor would like to construct a 98% confidence interval for the true difference in proportion of
males and females who have at least one piercing at the University of Pretoria. What is the lower bound
of this confidence interval? (i.e. CI : (Lower bound = answer , Upper bound) Only provide the lower
bound here.) Round your answer to 4 decimal places.

BES 220 7 of 8 Semester test 2 - Memo – 19 October 2022

 Solution:
We must first find the critical z value for a confidence level of 98%.
α 0.02
= = 0.01
2 2
zα/2 = z0.01 = 2.33
CI : (p′1 − p′2 ) ± zα/2 × SE
CI : (0.43 − 0.51) ± (2.33 × 0.0709)
CI : −0.08 ± 0.1652
CI : ( −0.2452 , 0.0852)

2 23. What is the upper bound for the 98% confidence interval that was mentioned in the previous question? (i.e.
CI : (Lower bound, Upper bound = answer ) Only provide the upper bound here.) Round your
answer to 4 decimal places.

Solution:

CI : (−0.2452, 0.0852 )

⋆⋆⋆ end of paper ⋆⋆⋆

BES 220 8 of 8 Semester test 2 - Memo – 19 October 2022