Chap-5Sampling Design & Strategies

1
Chap -5
Sampling Design & Strategies
2
Learning Objectives:
At the end of this Chapter, students are expected
to demonstrate competence in their capability to
independently develop sampling design.
3
1. Census and Sample Survey

 All items in any field of inquiry constitute a ‘Universe’ or ‘Population.’
 A complete enumeration of all items in the ‘population’ is known as a census inquiry.
 It can be presumed that in such an inquiry, when all items are covered, no element of
chance is left and highest accuracy is obtained.
 But in practice this may not be true.
 Besides, this type of inquiry involves a great deal of time, money and energy.
 Therefore, when the field of inquiry is large, this method becomes difficult to adopt
because of the resources involved.
4
Steps in Sample Design

While developing a sampling design, the researcher must pay
attention to the following points:
I)Type of Universe
 The universe can be finite or infinite.
 In finite universe the number of items is certain, but in case of an infinite universe the number
of items is infinite, i.e., we cannot have any idea about the total number of items.
 The population of a city, the number of workers in a factory and the like are examples of finite
universes,
 Whereas the number of stars in the sky, listeners of a specific radio programme, throwing of a
dice etc. are examples of infinite universes.
ii) Populations: desired, defined, and excluded
 In any scientific research study it is important to have a precise

description of the population of elements (persons, organizations,
objects, etc.) that is to form the focus of the study.
 In most studies this population will be a finite one that consists of
elements which conform to some designated set of specifications.
 These specifications provide clear guidance as to which elements
are to be included in the population and which are to be excluded .
6
 The population or universe represents the entire group of units which
is the focus of the study.
 Thus, the population could consist of all the persons in the country, or
those in a particular geographical location, or a special ethnic or
economic group, depending on the purpose and coverage of the study.
 A population could also consist on non-human units such as farms,
houses or business establishments.

iii) Representativeness
 The notion of ‘representativeness’ is a frequently
used, and often misunderstood, notion in social
science research.
 A sample is often described as being representative if certain

percentage frequency distributions of element
characteristics within the sample data are similar to
corresponding distributions within the whole population .
a) Sampling Unit
 A decision has to be taken concerning a sampling unit before
selecting sample.
 Sampling unit may be a geographical one such as state, district, village,

etc., or a construction unit such as house, flat, etc., or it may be a social
unit such as family, club, school, etc., or it may be an individual.
 The researcher will have to decide one or more of such units

that he has to select for his study.
b) Source List
 It is also known as ‘sampling frame’ from which sample is to be
drawn.
 It contains the names of all items of a universe (in case of finite
universe only).
 If source list is not available, researcher has to prepare it.
 Such a list should be comprehensive, correct, reliable and
appropriate.
 It is extremely important for the source list to be as
representative of the population as possible.
10
c) Size of Sample
 This refers to the number of items to be selected from the
universe to constitute a sample.
 This is one major problem in front of a researcher.
 The size of sample should neither be
excessively large, nor too small.
 It should be optimum.
 An optimum sample is one which fulfills the
requirements of efficiency, representativeness, reliability
and flexibility.
11
(d) Parameters of Interest

 In determining the sample design, one must consider the
question of the specific population parameters which are
of interest.
For Instance,
 We in estimating the
may be interested
proportion of persons with some
characteristic in the population, or we
may be interested in knowing some average or the
other measure concerning the population.
12
e) Budgetary Constraint
 Cost considerations, from practical point of view, have

a major impact upon decisions relating to not only the
size of the sample but also to the type of sample.
 This fact can even lead to the use of a non-probability
sample.
13
f) Sampling Procedure
 Finally, the researcher must decide the type of sample

he will use i.e., he must decide about the technique to
be used in selecting the items for the sample.
 In fact, this technique or procedure stands
for the sample design itself.
14
Criteria of Selecting a Sampling Procedure
 The two costs involved in a sampling analysis are namely:

 the cost of collecting the data and
 the cost of an incorrect inference resulting from the data.
 Researcher must keep in view the two causes of incorrect inferences
viz., systematic bias and sampling error.
 Systematic bias results from errors in the sampling procedures, and
it cannot be reduced or eliminated by increasing the sample size.
15
Usually a systematic bias is the result of one or more

of the following factors:
a) Inappropriate sampling frame

 If the sampling frame is inappropriate i.e., a biased representation of the
universe, it will result in a systematic bias.
b) Defective measuring device

 If the measuring device is constantly in error, it will result in systematic bias.
 In survey work, systematic bias can result if the questionnaire or the interviewer is
biased.
 Similarly, if the physical measuring device is defective there will be systematic bias in
the data collected through such a measuring device.
16
C)Non-Respondents
 If we are unable to sample all the individuals initially
included in the sample, there may arise a systematic
bias.
 The reason is that in such a situation the likelihood of
establishing contact or receiving a response from an
individual is often correlated with the measure of
what is to be estimated.
17
d) Indeterminacy Principle
 Sometimes we find that individuals act differently when kept
under observation than what they do when kept in non-
observed situations.
 For Instance,
 if workers are aware that somebody is observing them in
course of a work study on the basis of which the average
length of time to complete a task will be determined and
accordingly the quota will be set for piecework, they
generally tend to work slowly in comparison to the speed
with which they work if kept unobserved.
18
e) Natural bias in the reporting of data

 Natural bias of respondents in the reporting of
data is often the cause of a systematic bias in
many inquiries.
 There is usually a downward bias in the income
data collected by government taxation
department,
 Whereas we find an upward bias in the income
data collected by some social organization.
19
Characteristics of a Good Sample Design

 From what has been stated above, we can list down the characteristics of a good sample design asunder:
1) Sample design must result in a truly representative sample.
2) Sample design must be such which results in a small sampling error.
3) Sample design must be viable in the context of funds available for the research study.
4) Sample design must be such so that systematic bias can be controlled in a better way.
5) Sample should be such that the results of the sample study can be applied, in general, for the universe with a
reasonable level of confidence.

20
Types of Sample Design

Sampling is divided into two types:
a) Probability sampling: In probability sample, every

unit in the population has equal chances for being
selected as a sample unit.
b) Non–probability sampling: In non-probability
sampling, units in the population have unequal or zero
chances for being selected as a sample unit.
21
A) Probability Sampling Techniques:
1) Simple Random sampling
2) Systematic sampling
3) Stratified random sampling
4) Cluster sampling
5) Multistage sampling
22
1)Simple Random Sampling

 Simple random sample is a process in which every item of the population has equal
probability of being chosen.
 There are two methods used in random sampling –
a)Lottery Method
 Take a population containing 4 departmental stores: A, B, C & D.
 Suppose we need to pick a sample of two stores from the population using simple
random procedure.
 Combinations are AB,AD, AC, BC, BD, CD.
 write down 6 sample combination on six identical pieces of paper, fold the piece
of paper so that they cannot be distinguished.
 Put them in a box. Mix it and pull one at random.
23
(b) Using Random Number Table
 A random number table consists of a

group of digits that are arranged in
random order, i.e.
 Any row, column, or diagonal in such a table
contains digits that are not in any systematic
order.
24
2) Systematic Random Sampling

There are 3 steps:
 Sampling interval K is determined
 One unit between the first and Kth unit in the population list is
randomly chosen.
 Add Kth unit to the randomly chosen number.
Example:
 Consider 1000 households, from which we want to select 50 units.
 Calculate K =1000/50=20
 To select the first unit, we randomly pick one number between 1 to
20 say 17. So our sample is starting with 17, 37, 57…………..
Please note that only first item was randomly selected.
25
3) Stratified Random Sampling

 A probability sampling procedure in which simple random sub-samples
are drawn from within different strata that are more or less equal on
some characteristics.
Stratified sampling is of two types

a) Proportionate stratified sampling
 The number of sampling units drawn from each stratum is in
proportion to the population size of that stratum.
b) Disproportionate Stratified Sampling:
 The number of sampling units drawn from each stratum is
based on the analytical consideration, but not in proportion to
the population size of that stratum.
26
Reason for Stratified Sampling

 Sometimes business professionals want information about the
component part of the population.
Assume there are 3 stores.
Each store forms a strata and sampling from
within each strata is selected.
The result might be used to plan different
promotional activities for each store strata.
27
Stores size No. of stores Percentage of stores
Large stores 2000 20
Medium stores 3000 30
Small stores 5000 50
Total 10,000 100

28
 Suppose we need 12 stores, and then choose 4 from

each strata.
 Choose 4 stores at random.
 If there was no stratification, simple random sampling from the population
would be expected to choose:
 2 large stores (20 percent of 12)
 about 4 medium stores (30 percent of 12)
and
 about 6 small stores (50 percent of 12).
29
Stratified sampling can be carried out with
a) Same proportion across strata called proportionate

stratified sample
b) Varying proportion across strata called

disproportionate stratified sample.
30
Proportionate and Disproportionate Sampling
Stores size No. of stores Sample Sample

(Population) Proportionate Disproportionate
Large 2000 20 25
Medium 3000 30 35
Small 5000 50 40
Total 10000 100 100

31
Illustration:
A survey is planned to analyze the perception of people towards corruption practices in Sub
cities of Addis Ababa. Population consists of various strata, viz, Bole, Yeka, Gulele, Akaki,
and Arada. Assume total population is 10,000. Bole, Yeka, Gulele, Akaki, and Arada consists
of 6,000, 2,000, 1,000, 500 and 500 respectively. Determine the sample size of each stratum by
applying proportionate stratified sampling,
if the sample size required is 200.
Solution: Total population, N=10,000
 Population in the strata of Bole N1=6000, n1= N1/N x200= 120
 Population in the strata of Yeka N2=2000, n2= N1/N x200= 40
 Population in the strata of Gulele N3=1000, n3= N1/N x200=20
 Population in the strata of Akaki N4=500, n4= N1/N x200=10
 Population in the strata of Arada N5=500, n5= N1/Nx200=10

 Total…………………………………………………………………………………………………………200
32
4) Cluster Sampling
The following steps are followed:
a) Population is divided into clusters
b) A simple random sample of few clusters
selected
c) All the units in the selected cluster is
studied
33
 Cluster sampling is a sampling technique used when "natural" but

relatively homogeneous groupings are evident in a statistical population.
 It is often used in marketing research .
 In this technique, the total population is divided into these groups (or
clusters) and a simple random sample of the groups is selected.
 Then the required information is collected from a simple random sample
of the elements within each selected group.
 This may be done for every element in these groups or a subsample of
elements may be selected within each of these groups.
 A common motivation for cluster sampling is to reduce the total number
of interviews and costs given the desired accuracy.
 Assuming a fixed sample size, the technique gives more accurate results
when most of the variation in the population is within the groups, not
between them.
34
 The population within a cluster should ideally be as

heterogeneous as possible, but there should be homogeneity
between cluster means.
 Each cluster should be a small-scale representation of the total
population.
 The clusters should be mutually exclusive and collectively
exhaustive.
 A random sampling technique is then used on any relevant
clusters to choose which clusters to include in the study.
 In single-stage cluster sampling, all the elements from each of the
selected clusters are used.
 In two-stage cluster sampling, a random sampling technique is applied
to the elements from each of the selected clusters...
35
 The main difference between cluster sampling

and stratified sampling is that in cluster sampling
the cluster is treated as the sampling unit so
analysis is done on a population of clusters (at least
in the first stage).
 In stratified sampling, the analysis is done on
elements within strata.
 In stratified sampling, a random sample is drawn
from each of the strata, whereas in cluster sampling
only the selected clusters are studied.
36
 The main objective of cluster sampling is to

reduce costs by increasing sampling efficiency.
 Thiscontrasts with stratified sampling where

the main objective is to increase precision.
 There also exists multistage sampling, here

more than two steps are taken in selecting
clusters from clusters.
37
 One version of cluster sampling is area

sampling or geographical cluster sampling.
 Clusters consist of geographical areas.
 Because a geographically dispersed population can be
expensive to survey, greater economy than simple random
sampling can be achieved by treating several respondents
within a local area as a cluster.
 It is usually necessary to increase the total sample size to
achieve equivalent precision in the estimators, but cost
savings may make that feasible.
38
 Suppose we have a population of 20,000 units

from which we want to select 500 units.
 Choosing a sample of that size is a very time consuming
process, if we use Random Numbers table.
 Suppose the entire population is divided into 80
clusters of 250 units, we can choose two sample
clusters (2x250=500) easily by using cluster
sampling. The most difficult job is to form clusters.
In marketing, the researcher forms clusters
so that he can deal each cluster differently.
39
Area Sampling
 This is a Probability sampling.
 This is a special form of cluster sampling
Example 1:
 If someone wants to measure coffee sale in retail stores, one
might choose a city locality and then audit coffee sales, in all
retail outlets in those localities.
 The main problem in area sampling is the non-availability of
shop list selling coffee in a particular area.
 Therefore, it would be impossible to choose a probability
sample from these outlets directly.
40
5) Multistage Sampling
 The name implies that sampling is done in several stages.
 This is used with stratified /cluster designs .
An illustration of double sampling is as follows.
 Management of newly opened club is soliciting for membership.

 Therefore during the first round all corporate are sent details so that those
who are interested may enroll.
 Having enrolled, the second round concentrates on, how many are
interested to enroll for various entertainment activities that club is offering
such as Billiards club, indoor sports, swimming, and gym etc.
 After getting this information, you might stratify the interested respondents.
41
What are the Advantages and Disadvantages

of probability sampling?
 The advantage of probability Sampling is that:

 It is unbiased.
 Quantification is possible in probability sampling.
 Less knowledge of universe is sufficient.
 The disadvantages of probability sampling are that:
 It takes time.
 It is costly
 More resources are required to design and execute than non-probability design.
42
B. Non-Probability Sampling Techniques

1) Deliberate sampling
2) Shopping Mall Intercept Sampling
3) Sequential sampling
4) Quota sampling
5) Snowball sampling
6) Panel samples
43
Non- Probablity Sampling

a)Deliberate or Purposive Sampling
 This is also called judgment sampling.
 The investigator uses, his discretion in selecting sample observations from the universe.
 As a result, there is an element of bias in the selection.
 From the point of the investigator, the sample thus chosen may be a true representative of the
universe.
 However, the units in the universe do not enjoy equal chance of getting included in the sample.
 Therefore, it cannot be considered as a probability sampling.
Example:
 Test market cities are selected based on judgment sampling, because these
cities are viewed as a typical cities matches certain demographical
characteristics.
44
b)Shopping Mall Intercept Sampling
 This is a non-probability sampling method.

 In this method, respondents are recruited for individual
interviews at fixed locations in shopping malls.
Example:
 The researcher may wish to compare responses of two or more
TV commercials for two or more products.
 Mall samples can be informative for this kind of studies.
45
2) Sequential Sampling
 This is a method in which sample is formed on the basis of a series of
successive decisions.
 They aim at answering the research question on the basis of accumulated
evidence.
 Sometimes, a researcher may want to take a modest sample, look at the
results.
 Thereafter decide if more information is required for which larger samples are
considered.
 If the evidence is not conclusive, after a small sample is taken, more samples
are required.
 If still inconclusive, larger samples are taken.
 At each stage a decisions made about whether more information should be
collected or the evidence is now sufficient to permit a conclusion.
46
3)Quota Sampling
 Quota sampling is a frequently used type of non-probability sampling.
 It is sometimes misleadingly referred to as ‘representative sampling’ because numbers
of elements are drawn from various target population strata in proportion to the size of
these strata.
 While quota sampling places fairly tight restrictions on the number of sample elements per
stratum, there is often little or no control exercised over the procedures used to select elements
within these strata.
 For Example,
 Either judgment or convenience sampling may be used in any or all of the strata.
 Therefore, the superficial appearance of accuracy associated with proportionate
representation of strata should be considered in the light that there is no way of
checking either the accuracy of estimates obtained for any one stratum, or the
accuracy of estimates obtained by combining individual stratum estimates.
47
4)Snowball Sampling
 This is a non-probability sampling.

 In this method, the initial group of respondents is selected randomly.
 Subsequent respondents are selected based on the opinion or
referrals provided by the initial respondents.
 Further, referrals will lead referrals thus, leading to a snowball
sampling.
 The referrals will have demographic and psychographic
characteristics that are relatively similar to the person referring
them.
48
5) Panel Samples
 Panel samples are frequently used in marketing research.
 To give an example, suppose that one is interested in
knowing the change in the consumption pattern of
households.
 A sample of households is drawn.
 These households are contacted to gather information on
the pattern of consumption, subsequently, say after a
period of six months, the same households are approached
once again and the necessary information on their
consumptions collected.
49
6)Convenience Sampling
 A sample of convenience is the terminology used to
describe a sample in which elements have been selected
from the target population on the basis of their
accessibility or convenience to the researcher.
 Convenience samples are sometimes referred to as
‘accidental samples’ for the reason that elements may be
drawn into the sample simply because they just happen
to be situated, spatially or administratively, near to
where the researcher is conducting the data collection.
50
!! !
o u
Y
n k
h a
T

Chap-5Sampling Design & Strategies

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chap-5Sampling Design & Strategies

Uploaded by

Copyright:

Available Formats

1

1. Census and Sample Survey

Steps in Sample Design

ii) Populations: desired, defined, and excluded

 In any scientific research study it is important to have a precise

 The population or universe represents the entire group of units which

is the focus of the study.

those in a particular geographical location, or a special ethnic or

economic group, depending on the purpose and coverage of the study.

 A population could also consist on non-human units such as farms,

houses or business establishments.

 A sample is often described as being representative if certain

 Sampling unit may be a geographical one such as state, district, village,

 The researcher will have to decide one or more of such units

(d) Parameters of Interest

 Cost considerations, from practical point of view, have

 Finally, the researcher must decide the type of sample

Criteria of Selecting a Sampling Procedure

 The two costs involved in a sampling analysis are namely:

Usually a systematic bias is the result of one or more

a) Inappropriate sampling frame

b) Defective measuring device

e) Natural bias in the reporting of data

Characteristics of a Good Sample Design

1) Sample design must result in a truly representative sample.

2) Sample design must be such which results in a small sampling error.

reasonable level of confidence.

Types of Sample Design

a) Probability sampling: In probability sample, every

A) Probability Sampling Techniques:

1) Simple Random sampling

3) Stratified random sampling

1)Simple Random Sampling

(b) Using Random Number Table

 A random number table consists of a

2) Systematic Random Sampling

3) Stratified Random Sampling

Stratified sampling is of two types

Reason for Stratified Sampling

Stores size No. of stores Percentage of stores

Large stores 2000 20

Medium stores 3000 30

Small stores 5000 50

Total 10,000 100

 Suppose we need 12 stores, and then choose 4 from

Stratified sampling can be carried out with

a) Same proportion across strata called proportionate

b) Varying proportion across strata called

Proportionate and Disproportionate Sampling

Stores size No. of stores Sample Sample

Total 10000 100 100

 Population in the strata of Bole N1=6000, n1= N1/N x200= 120

 Population in the strata of Yeka N2=2000, n2= N1/N x200= 40

 Population in the strata of Gulele N3=1000, n3= N1/N x200=20

 Population in the strata of Akaki N4=500, n4= N1/N x200=10

 Population in the strata of Arada N5=500, n5= N1/Nx200=10

 Cluster sampling is a sampling technique used when "natural" but

 The population within a cluster should ideally be as

 The main difference between cluster sampling

 The main objective of cluster sampling is to

 Thiscontrasts with stratified sampling where

 There also exists multistage sampling, here