SQQSK2083 SAMPLING TECHNIQUES

CHAPTER 3 : SIMPLE RANDOM SAMPLING

DR. AYU ABDUL RAHMAN

SQS, UUM CAS

1
• In simple random sampling all the samples with the same
size are equally likely to be chosen.
• To conduct random sampling:
– assign a number to each element of the chosen
population (or use already given numbers),
– randomly select the sample numbers (members). Use a
random numbers table, or a software package.

2
• Example
– A government income-tax auditor is responsible for 1,000 tax returns.
– The auditor will randomly select 40 returns to audit.
– Use Excel’s random number generator to select the returns.
• Solution:

3
4
• Exercise:
– Let say you have a frame of 40 students who take SQQS
2083. As a researcher, you wish to select randomly 10
students to calculate the mean score of final exam.
– Use your calculator to generate a random number.
– Generate 15 random numbers.

5
 Score in Final Exam for QQS2023

Num Marks Num Marks

1 50 26 65
2 55 27 89
3 60 28 90
4 89 29 70
5 91 30 70
6 50 31 56
7 42 32 60
8 56 33 69
9 58 34 33
10 80 35 72
11 86 36 85
12 73 37 89
13 56 38 54
14 69 39 68
15 25 40 92
16 65
17 66
18 92
19 81
20 70
21 80
22 96
23 53
24 55
25 64

6
3.1 Estimation of Parameters in SRS
• In SRS, we will estimate the mean, total, variance and proportion
for the population.
• the estimation of population mean is ,
x

x
 x i

n
• The new term is estimator for total population is ,
ˆ
xi
ˆ   Nx
n N

7
 2
• Estimator for the variance sample mean, ˆ x is

2 s2  N  n
ˆ x   
n  N 
• (1 - α)% confidence interval

x Z  ˆ x
2

8
• Example 1:
– Using the sample that you generated in exercise 1,
estimate:
a.The mean score of the students
b.Error of sampling distribution/variance of the sample
means
c.Total of the population
d.95% confidence interval

9
a) The mean score of the students

b) Error of sampling distribution/variance of the sample

means

10
c) Total of the population

d) 95% confidence interval

11
• Example 2:
– In a particular city 1118 mortgages were taken out last year. A
random sample of 60 of these had mean amount RM87300 and
standard deviation RM19200. Estimate:
a.The mean amount of mortgages
b.Variance of the sample mean
c. Total of the population
d.95% confidence interval

12
• Solution:
a. Mean estimation,
b.Variance sample:

13
c. Total of the population:

d. 95% confidence interval:

14
• Exercise:
Suppose that there are 1395 collages in Malaysia. From a simple
random sample of 400 of these collages, it was found that the sample
mean enrollment during the past year in statistic courses was 320.8
students, and the sample standard deviation was found to be 149.7
students. Estimate:
i. The variance of the sample mean
ii. The total number of the students enrolled in statistic courses in
the year
iii. A 99% confidence interval for the mean of the students.

15
• Solution:

16
 • Estimation for proportion of SRS

pˆ 
 x i
x
n
2 pˆ qˆ  N  n 
 pˆ   
n  1 N 
Example:
From a simple random sample of 400 of 1395 colleges in our
population, it was found that statistics was a two- semester course in
141 of sampled colleges. Estimate the proportion of all colleges for
which the courses is two- semesters long and variance estimation
for the sample proportion

17
Solution:

18
Exercise:
– A simple random sample of 400 from a total 1395
University students in Malaysia contained 39 students
that feel statistics course made difficult and boring.
Estimate the proportion of all students that feel
statistics course made difficult and boring and estimate
the standard deviation of the sample proportion.

19
Solution:

20
3.2 Selecting Sample Size
• Selecting sample size for estimating mean
N 2
n
( N  1) D   2
B2
where D  , B 2 x
4

• Selecting sample size for estimating total

N 2
n
( N  1) D   2
B2
where D  , B 2 Var ( Nx )
4N 2
2 N x

21
Selecting sample size for estimating proportion

Npq 0.25 N
n  2
( N  1) D  pq ( N  1) pˆ  0.25
B2 2
where D  , B 2  pˆ
4

22
Example :
In a particular city, 1118 mortgages were taken last year. A
simple random sample is to be taken in order to estimate the
mean amount of the mortgages. From the previous experience
of such population it is estimated that the population standard
deviation is approximately RM20000. A 95% confidence interval
for the population mean must extend RM4000 on each side of
the sample mean. How many sample observations are needed
for this study?

23
Solution:

24
Example:
A simple random sample of the 1395 UUM students is
taken to estimate the proportion in which the statistics
course is two-semester long. Whatever the true
proportion, a 95% confidence interval must extend no
further than 0.08 on both side of the sample proportion.
How many sample observation should be taken?

25
Solution:

26
Example:
An investigator is interested in estimating the total number of ‘count
trees’ (trees larger than a specific size) on a plantation of N = 1500
acres. This information is used to determine the total volume of number
for trees on the plantation. A sample of n = 100 1-acre plots was
selected, and each plot was examined for the number of count trees.
The sample average for the n = 100 1-acre plots was 25.2 with a sample
variance of s2 = 136.
a. estimate the total number of count trees on the plantation.
b. place a bound on the error of estimation.
c. determine the sample size required to estimate τ, the total
number of tree on the plantation, with a bound on the
error of estimation of magnitude B = 500

27
Task 1:
A psychologist wishes to estimate the average reaction time to a
stimulus among 200 patients in a hospital specializing in nervous
disorders. A simple random sample of n = 20 patients was selected,
and their reaction times were measured, with
x 2.1 and s 0.4

a. Estimate the population mean and place a bound on the error of

estimation.
b. How large a sample should be taken in order to estimate μ with a
bound of 1 second on the error of estimation? Use 1 second as an
approximation of the population standard deviation

28
Solution:

29
29
30
Solution:

31
32
Example:
A country club wants to poll a random sample of its 320
members to estimate the proportion likely to attend an
early-season function. The number of sample
observations should be sufficiently large to ensure that
a 99% confidence interval for the population extends at
most 0.05 on each side of the sample proportion.
How large a sample is necessary?

33
Solution:

34
Task:
An auditor, examining a total of 800 accounts receivable of a
corporation, took a simple random sample of 60 of them. The
sample mean was $130.54, and the sample standard deviation was
$42.56. Using an unbiased estimation procedure, find an estimate
of the population mean.

Answer:

35
b. Using an unbiased estimation procedure, find an
estimate of the variance of the sample mean.

36
c. Find a 90% confidence interval for the
population

37
d. If the lower limit of confidence interval for
this population mean is 120.1826, find the
value of the significance level, α.

38
Task:
The mean amount of the 800 mortgages taken out in the city
of Melor, in the past year must be estimated. Based on the
previous experience, a real estate broker knows that the
population standard deviation is likely to be about RM25000.
If a 95% confidence interval for the population mean is to
extend RM2400 on each side of the sample mean, how many
sample observation are needed if a simple random sample is
taken?

39
40
Task:
A firm employs 189 juniors’ accountants. In a random sample of 50
of these, the mean number of hours overtime billed in a particular
week was 9.7, and the sample standard deviation was 6.2 hours.
– Estimate the variance of the sample mean.
– Estimate the total of population.
– Find a 95% confidence interval for the mean number of hours
overtimes billed per juniors’ accountant in this firm that week.

41
42
Task:
An automobile dealer has an inventory of 400 used
cars. To estimate the mean mileage of this inventory,
he intends to take a simple random sample of used
cars. Previous studies suggest that the population
standard deviation is 10,000 miles. A 90% confidence
interval for the population mean must extend 2000
miles on each side of its sample estimate. How large a
sample size necessary to satisfy this requirement?

43
44