CONFIDENTIAL 1 CS/FEB 2022/STA470

UNIVERSITI TEKNOLOGI MARA

FINAL ASSESSMENT

COURSE : STATISTICAL METHODS

COURSE CODE : STA470
EXAMINATION : FEBRUARY 2022
DATE : 22nd FEBRUARY 2022
TIME : 3 HOURS (9.00 AM – 12.00 PM)

INSTRUCTIONS TO CANDIDATES

1. This question paper consists of SIX (6) questions.

2. Write down your FULL NAME, STUDENT ID and GROUP on the first page of your answer
sheets.

3. Additional 30 minutes is allocated for submission. Any submission after 12.30 PM WILL NOT
be accepted.

4. Save your answers as ONE (1) PDF document as GROUP_STUDENT ID.

5. Answer ALL questions in English.

DO NOT TURN THIS PAGE UNTIL YOU ARE TOLD TO DO SO

This examination paper consists of 6 printed pages
© Hak Cipta Universiti Teknologi MARA CONFIDENTIAL
CONFIDENTIAL 2 CS/FEB 2022/STA470

QUESTION 1

a) State whether each of the following statements is True (T) or False (F).

i) Ratio is the highest level of measurement.

ii) The larger the sample size, the larger the standard error of the sample
proportion.

iii) An efficient estimator is the estimator with the highest variance.

iv) A type I error is the error of rejecting a true null hypothesis.

v) The Sign Test can be used to conduct tests about the median for one
population or to compare medians of two dependent populations.

(5 marks)

b) Fill the following blanks with an appropriate term.

i) ___________ statistics involve procedures of making conclusions

on population based on sample data.
ii) _______________ are measurements on the population of interest.

iii) Increasing the sample size will ________________ the margin of error in your
confidence interval.
iv) ________________ is used to make comparisons of more than two population
means.

v) The ____________ test is the non-parametric alternative to the independent

samples t test.
(5 marks)

QUESTION 2

a) Two independent random samples are chosen from two normal populations with the
following information.

Population 1 2
Sample size, n 45 40
p 0.7 0.68

i) Find the mean and standard deviation for the sampling distribution of the
difference between two sample proportions.
(4 marks)

ii) Calculate the probability that the proportion from sample 1 is less than the
proportion from sample 2.
(4 marks)
© Hak Cipta Universiti Teknologi MARA CONFIDENTIAL
CONFIDENTIAL 3 CS/FEB 2022/STA470

b) A taxi company purchase Brand A and Brand B tires for its taxis fleet. To estimate the
difference between the two brands of tire’s life, an experiment was conducted using
12 taxis for each brand. The tires were run until they wear out and the distances (in
kilometers) were measured. The descriptive statistics for the two brands of tires are
shown below.

Brand Population mean (in km) Population standard deviation (in km)
A 36300 5000
B 38100 6100

i) Construct the sampling distribution for the difference between the two brands of
tire’s life.
(3 marks)

ii) Find the probability that the mean distance of tires to wear out for Brand B
exceeds Brand A by more than 1000 kilometers.
(4 marks)

QUESTION 3

a) A random sample of 500 adult residents of Town A found that 385 were in favor of
increasing the highway speed limit to 120 km/h, while another sample of 400 adult
residents of Town B found that 250 were in favor of the increased speed limit.

i) Construct a 95 % confidence interval on the difference in the proportion of the

adult residents who in favor of the increased limit between the two towns. Interpret
the interval obtained.
(6 marks)

ii) Estimate the sample size required for a 99% confidence interval to specify the
percentage of adult residents who were in favor of increasing the speed limit from
Town A within 5% margin of error.
(4 marks)

b) Let X 1 , X 2 , X 3 , X 4 , X 5 be a random sample from a population with mean,  and

variance,  . Suppose M and N are two estimators of  where;
2

i) Are both estimators unbiased estimators of  ?

(5 marks)
ii) Which estimator is more efficient?
(5 marks)

© Hak Cipta Universiti Teknologi MARA CONFIDENTIAL

CONFIDENTIAL 4 CS/FEB 2022/STA470

QUESTION 4

a) An instructor prepares two sets of exam questions from the same topics for two different
classes. She found out that the performance of one class is higher than the other
though she used the same method to teach both classes. In order to understand the
reason why the performance was different, she administered both exam questions to 12
students from other programs. The score of both exams for each of the students are
given below. Assuming the data comes from a normally distributed population.

Student Exam Score 1 Exam Score 2

Nina 63 58
Khatijah 65 57
Julia 98 86
Michael 88 67
Trisya 77 80
Umar 92 88
Suzy 68 69
Selene 87 81
Kelvin 64 70
Ali 90 85
Mages 88 82
Indra 71 72

At 5% level of significance, justify the fairness of the exam questions.

(8 marks)

b) In a final event of a bowling competition, two bowlers are competing to become the
champion. The summary of scores obtained by the bowlers and the analysis output are
given below:

Group Statistics
Bowler N Mean Std. Deviation Std. Error Mean
Kenny 13 228.00 28.337 7.859
Scores
Sufi 11 233.36 21.951 6.619

Independent Samples Test

Scores
Equal variances Equal variances
assumed not assumed
t -.511 -.522
df 22 21.862
t-test for Sig. (2-tailed) .615 .607
Equality of Mean Difference -5.364 -5.364
Means Std. Error Difference 10.501 10.275
95% Confidence Interval Lower -27.141 -26.680
of the Difference Upper 16.414 15.593

© Hak Cipta Universiti Teknologi MARA CONFIDENTIAL

CONFIDENTIAL 5 CS/FEB 2022/STA470

i) Calculate the unbiased estimate of pooled variance for the scores obtained by both
bowlers.
(2 marks)

ii) Construct a 95% confidence interval for the ratio of the variances of the scores
obtained by the two bowlers. What can you conclude on the equality of the
variances?
(5 marks)

iii) Sufi was awarded the champion since his average score was higher than Kenny’s
average score. By considering the results in part (ii) and the given SPSS output
above, is there evidence to indicate that the mean score obtained by Sufi was
significantly higher than Kenny? Use  = 0.05.
(5 marks)

QUESTION 5

Encik Rizal manages a car rental company. He plans to use new engine oil for his fleet of
cars for rental. Many oil producers claim that their product could reduce fuel consumption.
He decides to try four products from four renowned oil producer companies. He plans to use
the engine oil which gives the highest mileage if there is significant difference in the
performance. The mileage per litre was recorded. The summary statistics were tabulated
below:

Mobidick Power Lasso Formula 1 Siput Power Casatel Extra

Engine Oil

Mean 32.5 34.2 35.4 30.7

Standard
Deviation 1.643 2.490 2.073 1.862

Sample Size 6 5 5 6

2
Given SSTOTAL = 142.9 and where s j is the sample variance and
n j is the sample size for each type of engine oil.

a) At 5% level of significance, can Encik Rizal conclude that the average fuel consumption
provided by all four engine oil are significantly different?
(7 marks)

b) Construct the 95% confidence interval mean difference to compare,

i. Siput Power with Mobidick Power.

(2 marks)
ii. Siput Power with Lasso Formula 1.
(2 marks)

© Hak Cipta Universiti Teknologi MARA CONFIDENTIAL

CONFIDENTIAL 6 CS/FEB 2022/STA470

c) Given the price of Siput Power, Mobidick Power and Lasso Formula 1 per litre are as
follow:

Engine Oil Mobidick Power Lasso Formula 1 Siput Power

Price per litre RM2.00 RM2.20 RM2.30

Based on your answer in (b) and the price for each type of the engine oil, which brand
would likely be chosen by Encik Rizal? Explain your choice.
(2 marks)

d) “Tests of Homegeneity of Variances revealed that the Levene Statistic = 0.315 with p-
value = 0.814”. Would this statement make a difference in the findings above? Explain.

(2 marks)

QUESTION 6

a) State TWO (2) advantages and THREE (3) allowable conditions for using non-
parametric methods.
(5 marks)

b) i) The manufacturer of a hot tub is interested in testing two different heating elements
for his product. The element that produces the maximum heat gain after 15 minutes
would be preferable. He obtains 10 samples of each heating unit and tests each one.
The heat gain after 15 minutes is given as follows:

Unit 1 25 27 29 31 30 26 24 32 33 38
Unit 2 31 32 32 35 34 29 39 35 37 40

Assume the heat gain is not normally distributed. At 5% significance level is there
evidence that Unit 1 is more superior to Unit 2?
(10 marks)

ii) Ten samples were taken from a plating bath used in an electronic manufacturing
process, and the bath pH was determined. The sample pH values are as follows:

7.91 7.88 6.82 8.01 7.46 6.95 7.05 7.38 7.25 7.42

Use the sign test at 5% significance level to test the claim that the pH has a median
value more than 7.5.
(5 marks)

END OF QUESTIONS PAPER

© Hak Cipta Universiti Teknologi MARA CONFIDENTIAL