Selecting a Sample

1
 The major learning focusses:
• The differences between sampling in
qualitative and quantitative research
• Definitions of sampling terminology
• The theoretical basis for sampling
• Factors affecting the inferences drawn from a
sample
• Different types of sampling
• The concept of saturation point
2
 Keywords:
accidental sampling, cluster sampling, data
saturation point, disproportionate sampling,
judgemental sampling, multi-stage cluster sampling,
non-random sample, population mean, population
parameters, random numbers, random sample,
sample statistics, sampling, sampling design,
sampling element, sampling error, sampling frame,
sampling population, sampling unit, sample size,
sampling strategy, saturation point, snowball
sampling, study population, stratified sampling.
3
 The differences between sampling
in quantitative and qualitative research
The selection of a sample in quantitative and qualitative
research is guided by two opposing philosophies.
• In quantitative research you attempt to select a
sample in such a way that it is unbiased and
represents the population from where it is selected.
• In qualitative research, the selection is influenced by:
easy access to the potential respondents; your
judgement that the person has extensive knowledge
about the issue of interest to you; how typical or
totally different the case is from the others.
4
 The differences between sampling
in quantitative and qualitative research

• The purpose of sampling in quantitative research is

to draw inferences about the group from which you
have selected the sample, whereas in qualitative
research it is designed either to gain in-depth
knowledge about a situation/event/episode or to
know as much as possible about different aspects of
an individual on the assumption that the individual is
typical of the group and hence will provide insight
into the group.
5
 The differences between sampling
in quantitative and qualitative research
The determination of sample size:
• In quantitative research you are guided by a
predetermined sample size, based on a number of
considerations in addition to the resources available.
• In qualitative research you do not have a
predetermined sample size but during the data
collection phase you will reach a point of data
saturation. It is assumed you have reached that point
when you are not getting new information anymore.

6
 The differences between sampling
in quantitative and qualitative research
• Considerable importance is placed on the sample size
in quantitative research, depending on the type of
study and the possible use of the findings. Studies
which are designed to formulate policies, to test
associations or relationships, or to establish impact
assessments place a considerable emphasis on large
sample size. This is based on the principle that a
larger sample size will ensure the inclusion of people
with diverse backgrounds, thus making the
sample representative of the study population.

7
 The differences between sampling
in quantitative and qualitative research
The sample size in qualitative research does not play
any significant role as the purpose is to study only one
or a few cases in order to identify the spread of
diversity and not its magnitude. Data saturation stage
during data collection determines the sample size.
In quantitative research, randomisation is used to avoid
bias in the selection of a sample and is selected in such
a way that it represents the study population. In
qualitative research no such attempt is made. You
purposely select 'information-rich' respondents who will
provide you with the information you need.
8
 Sampling in quantitative research
The concept of sampling
Let us take a simple example to explain the concept of
sampling. Suppose you want to estimate the average
age of the students in your class. There are two ways of
doing this. The first method is to contact all students in
the class, find out their ages, add them up and then
divide this by the number of students. The second
method is to select a few students from the class, ask
them their ages, add them up and then divide by the
number of students you have asked. From this you can
make an estimate of the average age of the class.

9
 Sampling in quantitative research
The concept of sampling
Similarly, suppose you want to find out the average
income of families living in a city. Imagine the amount
of effort and resources required to go to every family in
the city to find out their income!
You could instead select a few families to become the
basis of your enquiry and then, from what you have
found out from the few families, make an estimate of
the average income of all families in the city.

10
 Sampling in quantitative research
The concept of sampling

• Sampling is the process of selecting a few (a sample)

from a bigger group (the sampling population) to
become the basis for estimating or predicting the
prevalence of an unknown piece of information,
situation or outcome regarding the bigger group.
• The sample is a subgroup of the population you are
interested in.

11
 Sampling in quantitative research
The concept of sampling
This process of selecting a sample from the total
population has advantages and disadvantages.
• The advantages are that it saves time as well as
financial and human resources.
• However, the disadvantage is that you do not find
out the information about the population's
characteristics of interest to you but only estimate
or predict them. Hence, the possibility of an error in
your estimation exists.

12
 Sampling in quantitative research
The concept of sampling


You select a few sampling

units
from the study population

You collect information

from these people to
find answers to your
research questions

13
 Sampling in quantitative research
The concept of sampling
Sampling, therefore, is a trade-off between certain
benefits and disadvantages. On the one hand you save
time and resources, on the other hand you compromise
the level of accuracy in your findings. Through sampling
you only make an estimate about the actual situation
prevalent in the total population from which the sample
is drawn.
If you extract information from the total sampling
population with correct method, your findings should be
reasonably accurate.
14
 Sampling terminology

Let us, consider the examples where our aims are to

find out the average age of the class, or the average
income of families in a city. Let assume that you adopt
a sampling method to select a few students, or families
to achieve these aims. In this process there are a
number of aspects received special terminology:
• The class or families living in the city from which you
select your sample are called the population or study
population.

15
 Sampling terminology

• The small group of students or families from whom

you collect the required information to estimate the
average age of the class or average income is called
the sample.
• The number of students or families from whom you
obtain the information is called the sample size.
• The way you select students or families is called the
sampling design or sampling strategy.
• Each student or family in the sample is called the
sampling unit or sampling element.
16
 Sampling terminology

• A list identifying each student or family in the study

population is called the sampling frame.
• Your findings based on the information from your
respondents (sample) are called sample statistics.
• From sample statistics we make an estimate of the
answers to our research questions in the study
population. The estimates arrived from sample
statistics are called population parameters or the
population mean.

17
 Principles of sampling

The theory of sampling is guided by three principles.

Principle 1 - in a majority of cases of sampling there will
be a difference between the sample statistics and the
true population mean.
Principle 2 - the greater the sample size, the more
accurate the estimate of the true population mean.
Principle 3 - the greater the difference in the variable
under study in a population for a given sample
size, the greater the difference between the sample
statistics and the true population mean.
18
 Bias in the selection of a sample

Bias in the selection of a sample can occur if:

• sampling is done by a non-random method - the
selection is influenced by human choice;
• the sampling frame - list, index or other population
records - the basis of selection, does not cover the
sampling population accurately and completely;
• a section of a sampling population is impossible to
find or refuses to co-operate.

19
 Types of sampling

Sampling designs can be classified as:

• random/probability sampling designs;
• non-random/non-probability sampling designs and
• 'mixed' sampling designs.
For a sample to be called a random sample, each
element in the study population must have an equal and
independent chance of selection. Three random
designs are commonly used: simple random sampling,
stratified random sampling and cluster sampling.

20
 Simple random sampling (SRS)

SRS is the most used method of selecting a probability

sample, whereby each element in the population is
given an equal and independent chance of selection.
The simple random sample is selected In three steps:
Step 1 Identify by a number all elements or sampling
units in the population .
Step 2 Decide on the sample size.
Step 3 Select the sample size using the fishbowl draw,
the table of random numbers or a computer program.

21
 Simple random sampling (SRS)

To illustrate, let us again take our example of the class.

There are 80 students in the class. So, the first step is to
identify each student by a number from 1 to 80.
Suppose you decide to select a sample of 20 using the
simple random sampling technique (second step).
Use the fishbowl draw, the table for random numbers
or a computer program to select the 20 students (third
step). These 20 students become the basis of your
enquiry.

22
 Stratified random sampling

As discussed, the accuracy of your estimate largely

depends on the extent of variability or heterogeneity
of the study population with respect to the
characteristics that have a strong correlation with what
you are trying to ascertain (Principle 3).
It follows, that if the heterogeneity in the population
can be reduced by some means for a given sample size
you can achieve greater accuracy in your estimate.
Stratified random sampling is based on this logic.

23
 Stratified random sampling

In stratified random sampling the researcher attempts

to stratify the population in such a way that the
population within a stratum is homogeneous with
respect to the characteristic on the basis of which it is
being stratified. It is important that the characteristics
chosen as the basis of stratification are clearly
identifiable in the study population. For example, it is
much easier to stratify a population on the basis of
gender than on the basis of age, income or attitude.

24
 Stratified random sampling

Once the sampling population has been separated into

non-overlapping groups, you select the required number
of elements from each stratum, using the SRS. There are
two types of stratified sampling: proportionate
stratified sampling and disproportionate stratified
sampling. With proportionate stratified sampling, the
number of elements from each stratum in relation to its
proportion in the total population is selected, whereas
in disproportionate stratified sampling, consideration is
not given to the size of the stratum.

25
The procedure for stratified random sampling
Step 1 Identify all elements or sampling units in the sampling population.
Step 2 Decide upon the different strata (k) into which you want to stratify the population.
Step 3 Place each element into the appropriate stratum.
Step 4 Number every element in each stratum separately.
Step 5 Decide the total sample size (n).
Step 6 Decide whether you want to select proportionate or disproportionate stratified sampling and
follow the steps below.

Disproportionate stratified sampling Proportionate stratified sampling

Step 7 Determine the number of elements to Step 7 Determine the proportion of each stratum
be selected from each stratum in the study population (p)

sample size (n) elements in each stratum

=
no. of strata (k) total population size
Step 8 Select the required number of elements Step 8 Determine the number of elements to be
from each stratum with SRS technique selected from each stratum = (sample
size) x (p)
Step 9 Select the required number of elements
from each stratum with SRS technique
As this method does not take the size of the As the sample selected is in proportion to
stratum into consideration in the selection of the the size of each stratum in the population,
sample, it is called disproportionate stratified this method is called proportionate stratified
sampling. sampling.
/ 26
 Cluster sampling

Simple random and stratified sampling techniques are

based on a researcher's ability to identify each element
in a population. It is easy to do this if the total
sampling population is small, but if the population is
large, as in the case of a city, state or country, it
becomes difficult and expensive to identify each
sampling unit. In such cases the use of cluster sampling
is more appropriate.

27
 Cluster sampling

Cluster sampling is based on the ability of the researcher

to divide the sampling population into groups, called
clusters, and then to select elements within each cluster,
using the SRS technique. Clusters can be formed on the
basis of geographical proximity or a common
characteristic that has a correlation with the main
variable of the study (as in stratified sampling).
Depending on the level of clustering, sometimes
sampling may be done at different levels. These levels
constitute the different stages (single, double or
multiple) of clustering.
28
Non-random/non-probability sampling designs in
quantitative research

Non-probability sampling designs are used when the

number of elements in a population is either unknown
or cannot-be individually identified. There are five
commonly used non-random designs, each based on a
different consideration:
• quota sampling;
• accidental sampling;
• judgemental sampling or purposive sampling;
• expert sampling; and
• snowball sampling.
29
 Quota sampling

The main consideration directing quota sampling is the

researcher's ease of access to the sample population.
In addition to convenience, you are guided by some
visible characteristic that is of interest to you. The
sample is selected from a location convenient to you as
a researcher, and whenever a person with this visible
relevant characteristic is seen that person is asked to
participate in the study. The process continues
until you have been able to contact the required
number of respondents.
30
 Quota sampling

Let us suppose that you want to select a sample of 20

male students in order to find out the average age of
the male students in your class. You decide to stand at
the entrance to the classroom, as this is convenient,
and whenever a male student enters the classroom,
you ask his age.
This process continues until you have asked 20 students
their age. The advantages of using this design are: least
expensive; no need of a sampling frame, inclusion of
the type of people you need.
31
 Quota sampling

The disadvantages are: the findings cannot be

generalised to the total sampling population; the most
accessible individuals might not be truly representative
of the total sampling population.
You can make your sample more representative of your
study population by selecting it from various locations
where people of interest to you are likely to be
available.

32
 Accidental sampling

Accidental sampling is based on convenience in accessing

the sampling population. Whereas quota sampling
attempts to include people possessing an obvious/visible
characteristic, accidental sampling makes no such attempt.
You stop collecting data when you reach the required
number of respondents you decided to have.
Accidental sampling is common among market research
and newspaper reporters. It has the same advantages and
disadvantages as quota sampling but, in addition, some
people contacted may not have the required information.
33
 Judgmental or purposive sampling

The primary consideration in purposive sampling is

your judgment about who can provide the best
information to achieve the objectives of your study.
You as a researcher only go to those people who in
your opinion are likely to have the required
information and be willing to share it with you.
This type of sampling is extremely useful when you
want to construct a historical reality, describe a
phenomenon or develop something about which only
a little is known.
34
 Expert sampling

In the case of expert sampling, your respondents must

be known experts in the field of interest to you.
• When you use it in qualitative research, the number
of people you talk to is dependent upon the data
saturation point.
• In quantitative research you decide on the number
of experts to be contacted.
You first identify persons with known expertise, seek
their consent for participation, and then collect the
information.
35
 Snowball sampling

Snowball sampling is the process of selecting a sample

using networks. To start with, a few individuals in a
group or organisation are selected and the required
information is collected from them. They are then
asked to identity other people in the group or
organisation, and the people selected by them become
a part of the sample. Information is collected from
them, and then these people are asked to identity
other members of the group and, in turn, those
identified become the basis of further data collection.

36
Snowball sampling

37
 Snowball sampling

This sampling technique is useful if you know little

about the group or organisation you wish to study, as
you need only to make contact with a few individuals,
who can then direct you to the other members of the
group. It is useful for studying communication patterns
or decision making within a group.
The disadvantages arise from the fact, that the sample
depends on the choice of individuals, selected at the
first stage. If they belong to a particular faction or have
strong biases, the study may be biased.
38
 Systematic sampling design: a 'mixed' design

Systematic sampling has been classified as a 'mixed'

sampling design because it has the characteristics of
both random and non-random sampling designs.
In systematic sampling the sampling frame is first
divided into a number of segments called intervals.
The selection of subsequent elements from other
intervals is dependent upon the order of the element
selected in the first interval.

39
 Systematic sampling design: a 'mixed' design

If in the first interval the chosen is the fifth element,

the fifth element of each subsequent interval will be
chosen.
Notice that from the first interval the choice of an
element is on a random basis, but the choice of the
elements from subsequent intervals is dependent upon
the choice from the first, and hence cannot be
classified as a random sample.

40
 Systematic sampling design: a 'mixed' design

The procedure used in systematic sampling is:

Step 1 Prepare a list of all the elements in the study
population (N).
Step 2 Decide on the sample size (n).
Step 3 Determine the width of the Interval (k) = total
population/sample size.
Step 4 Using the SRS, select an element from the first
interval.
Step 5 Select the same order element from each
subsequent interval.
41
 Systematic sampling design: a 'mixed' design

Suppose there are 50 students in a class and you

want to select 10 students using the systematic
sampling technique. To determine the width of the
interval you calculate: 50/10 = 5.
This means that from every five you need to select
one element. Using the SRS technique, from the first
interval (1-5 elements), select one of the elements.
Suppose you selected the third element. From the
rest of the intervals you would select every third
element.
42
 Sampling in qualitative research

As the main aim in qualitative enquiries is to explore

the diversity, sample size and sampling
strategy do not play a significant role in the selection
of a sample. If selected carefully, diversity
can be extensively and accurately described on the
basis of information obtained even from one
individual. All non-probability sampling designs -
purposive, judgemental, expert, accidental and
snowball - can also be used in qualitative research
with two differences:
43
 Sampling in qualitative research

① In quantitative studies you collect information from

a predetermined number of people but, in qualitative
research, you do not have a sample size in mind. Data
collection based upon a predetermined sample size
and the saturation point distinguishes their use in
quantitative and qualitative research.
② In quantitative research you are guided by your
desire to select a random sample, whereas in
qualitative research you are guided by your judgement
-who is likely to provide you with the 'best' information.
44
 The concept of saturation point in qualitative
research
In qualitative research data is usually collected to a
point where you are not getting new information or it is
negligible - the data saturation point. This stage
determines the sample size.
It is important for you to keep in mind that the concept
of data saturation point is highly subjective. It is you
who decide when you have attained the saturation
point in your data collection.
Objectively, the saturation point depends on how
diverse is the studying situation or phenomenon.
45
 The concept of saturation point in qualitative
research
The greater the diversity, the greater the number of
people from whom you need to collect the information
to reach the saturation point.
The concept of saturation point is more applicable to
situations where you are collecting information on a
one-to-one basis. Where the information is collected in a
collective format such as focus groups, community
forums or panel discussions, you strive to gather as
diverse and as much information as possible.

46
 For You to Think About

• Familiarise yourself with the keywords listed at the

beginning of this presentation.
• Consider the implications of selecting a sample
based upon your choice as a researcher and how you
could make sure that you do not introduce bias.
• In the absence of a sampling frame for employees of
a large organisation, which sampling design would
you use to select a sample of 200 people? Explain
why you would choose this design and the process
you would undertake to ensure that the sample is
representative.
47
References
Ranjit Kumar (2011), Research Methodology, third edition,
SAGE Publications Ltd.
Alan Bryman and Emma Bell (2011), Business Research Methods,
third edition, Oxford University Press.
Chris Welman, Fanie Kruger and Bruce Mitchell (2010), Research
Methodology, third edition, ninth impression, Oxford University
Press.

48
Thank you

49