MURANG’A UNIVERSITY OF TECHNOLOGY

SCHOOL OF PURE AND APPLIED SCIENCE

DEPARTMENT OF APPLIED SCIENCE

UNIVERSITY ORDINARY EXAMINATION

2017/2018 ACADEMIC YEAR

FOURTH YEAR SECOND SEMESTER EXAMINATION FOR THE DEGREE OF
BACHELOR OF SCIENCE IN MATHEMATICS AND COMPUTER SCIENCE

SMA 2432 – DESIGN AND ANALYSIS OF SAMPLE SURVEYS

DURATION: 2 HOURS

DATE: 27TH APRIL, 2018

TIME: 9.00 – 11.00 A.M.

Instructions to Candidates:

1. Answer Question 1 and Any Other Two questions.

2. Mobile phones are not allowed in the examination room.
3. You are not allowed to write on this examination question paper.

Page 1 of 5
 SECTION A – ANSWER ALL QUESTIONS IN THIS SECTION
QUESTION ONE
a) Explain what is a good sample in measurement terms (3 marks)
b) Define the following non-probability sampling methods:
i. Purposive sampling (1 mark)
ii. Convenience sampling (1 mark)
iii. Quota sampling (1 mark)
c) State and explain four causes of systematic bias in sampling (4 marks)
d) In a private library, the books are kept on 130 shelves of similar size. The numbers of books on
15 shelves picked at random were found to be 28, 23, 25,33, 31, 18, 22, 29, 30, 22, 26, 20, 21, 28
and 25. Estimate the total number Y of books in the library (4 marks)
e) A sample of 30 students is to be drawn from a population of 300 students belonging to two
colleges; A and B. The means and standard deviations of their marks are given below:
Total number of students 𝑦𝑖 𝑆𝑖
College A 200 30 10
College B 100 60 40

Use this information to confirm that Neyman’s allocation scheme is a more efficient scheme
when compared to proportional allocation (5 marks)
f) Suppose the following summarized information is made available
𝑛 = 25 N = 275 𝑥 = 9.2 𝑦 = 2.6
𝑥𝑖2 = 2200 𝑥𝑖 𝑦𝑖 = 500 𝑦𝑖2 = 170
Estimate R and Var (R) (5 marks)
g) A production line makes 10,000 units a day. How can the quality control department take a
systematic sample of 5% of these? (3 marks)
h) Suppose that we have a population of size N=4 whose population units are 1, 2, 3, 4 and that we
require a sample of size n=2 from the population. Assuming we use simple random sampling
without replacement (SRSWOR). Find
i. The number of possible samples (1 mark)
ii. Specify the samples (2 marks)

Page 2 of 5
 SECTION B – ANSWER ANY TWO QUESTIONS IN THIS SECTION

QUESTION TWO
a) Let Yi be the value of the characteristic under study for the ith unit of the population and Xi be
the value of the auxillary characteristic of the ith unit of the population. Show that the ratio
estimate (𝑦𝑅 ) is a biased estimator of population mean𝑌 (6 marks)
b) A daily newspaper conducts a survey of food costs by taking a simple random sample of 48 basic
foodstuffs purchased in a large supermarket. Prices (in Kenya shillings) for these items are
recorded in two separate occasions, three months apart, the earlier ones being denoted 𝑥𝑖 and the
latter 𝑦𝑖 . The sample ratio r gives an indication of change of these basic food prices over three
months period in the form of an estimate of the population ratio R of the mean prices of food.
The following results were obtained:
𝑥 = 11.41, 𝑦 = 12.07, 𝑥𝑖2 = 8431.7
𝑥𝑖 𝑦𝑖 = 8564.1 𝑦𝑖2 = 9270.6 𝑛 = 48
i. Obtain the value of r (2 marks)
ii. Estimate variance of r (3 marks)
iii. Obtain the 95% confidence interval for the population ratio R (3 marks)
c) In studying lung function in a group of 560 workers in a coal mine, an estimate was required of
mean value of some relevant measure Y. A simple random sample of 10 workers was chosen and
their values, 𝑦𝑖 determined by an appropriate test. A note was also made of their heights, 𝑥𝑖 . The
results were:
𝑦𝑖 3.0 3.5 3.3 3.1 4.1 3.2 3.7 2.9 3.9 3.4
𝑥𝑖 (cm) 173 183 170 175 160 157 168 180 178 163

From routine medical records the average height for the group of 560 workers is known to be
𝑥 = 173.2cm. Estimate 𝑌 from the data. (6 marks)

QUESTION THREE
a) Research is to be carried out in the informal sector. The sector is divided into three groups, that
is, low income earners, medium income earners and high income earners. The total amount
allocated to this research is $1500. The following data was collected after the interviews:

Page 3 of 5
 Class Number of subsectors Variance of money spent Cost per interview
taken Ni Si
Low 350 200 5
Medium 500 400 8
High 250 500 12

i. Design a proportional allocation sample (6 marks)

ii. Design an optimal allocation sample (7 marks)
iii. For the two designs in (i) and (ii), determine the estimated variances of the average amount
spent hence or otherwise state the most efficient design and explain why (7 marks)

QUESTION FOUR
a) Suppose we have a population of size N=5 whose population units are 1, 2, 3, 4, 5 and that we
require a sample of size n=2 from this population. Assuming we use simple random sampling
without replacement. Find;
i. Specify these samples (2 marks)
ii. Show that the sample mean is unbiased for the population mean using this data (4½ marks)
iii. Show that the sample variance is unbiased for the population variance using this data
(5½ marks)
iv. Find the variance of the sample mean (2 marks)
b) In a particular sector of the industry, a survey is conducted to investigate the extent of
absenteeism from duty which is not connected to illness or official holidays. A random sample of
1000 employees out of the workforce of 36000 were asked to indicate the number of days they
had failed to report to work for the previous six month. They were asked to give reasons. The
following were the results
Number of days 0 1 2 3 4 5 6 7 8 9
Number of employees 451 162 187 112 49 21 5 11 2 0

i. Determine the average number of days that were lost by the industry (3 marks)
ii. Determine an estimate of population variance, S2 (3 marks)

QUESTION FIVE
a) A videotape hire company has shops in each of the 5 regions; three regions have 12 shops and
others have 8. To estimate the total number Y of video films hired from the company in a

Page 4 of 5
 particular week, the sales manager phones 12 shops chosen by picking 3 regions at random then
making a choice of 3, 5 and 4 shops from the chosen regions. The results were as follows:
First region 260 296 182 -12 shops
Second region 156 261 130 302 241 -8 shops
Third region 196 356 264 284 -12 shops
Estimate the total number y (10 marks)
b) Let N=1000 and n=10. Assume that the random start r is selected, say, r=5. Find a systematic
sample composition (3 marks)
c) Assume that a population of size N is divisible into k groups each of size n so that N= nk. Let the
units be labeled 1, 2, … N in a fixed order of some kind.
i. Describe briefly the process of drawing a systematic sample (4 marks)
ii. Show that 𝑦𝑠𝑦 is an unbiased estimator of 𝑌 (3 marks)

Page 5 of 5