Population and Sampling

Introduction
• This chapter explores the ways by which data can be gathered for a
research project, and the chapter deals with research populations and
sampling methods.

• The methodological framework of the research project is comprised of an

account of all of the ways and means by which the research was actually
carried out.

• The population used in the research, the sample selected from that
population, and the means used to gather data for the research project
are all fundamental aspects of this framework.

• The model of the research process, (outlined in Chapter One), shows that,
at this stage in the research process we have
decided on the research methodology to be used in
the research project, and we have now come to the
stage of defining the population of the research, deciding
whether to work with the entire population or with a sample
of that population.
 The Population of the Study
• The population of a study is all of the individuals, items or
units relevant to the study.

• The population can be comprised of individuals, groups,

organisations, documents, campaigns, incidents, and so on.

• The population of the research is also called the universe.

• As the term ‘the universe’ implies, the

population of a study is comprised of all of
the units or individual belonging to that
population.
 Examples of research populations
The researcher might decide to compare levels of disposable income among:

• EU and non-EU undergraduate students at the University of Liverpool;

• or among undergraduate students at university in Liverpool;

• or among undergraduate students at university in England;

• or among undergraduate students at university in the UK.

• The researcher may decide to focus on the population of undergraduates in UK

universities, or the population of undergraduates in the University of Westminster,
or in the University of Edinburgh, or in Cardiff University.

• The researcher may decide to focus on the population of

undergraduates in all three universities.

• All of these populations are valid. The population of the

study is every person relevant to the population of the study,
as the population of the study is defined by the researcher.
 Questions of Scope
• The researcher defines the population of the study precisely, in order to
ensure that the research project is do-able, in order to ensure that it is
researchable.

• The researcher can only do what it is possible to do, and so the researcher
limits the scope of the research.

• The resources necessary in order to ensure that the research project is

researchable include the amount of time available for the research, the
amount of money required to conduct the research, and the level of
access to the necessary data available to the researcher.

• In designing the research project, the researcher might decide, for

example, to situate the research in one University, or
in more than one University. The student might chose
to situate the research in one Business School or
Faculty within a particular University or College, or
in more than one Business School or Faculty.
• These are questions of scope, questions of how big
the study or the research project is to be.
 Samples and Sampling in Research
• In social science research, it is often the case that the entire
population of the research is too big, and as a result beyond
the scope of the researcher.

• In such situations, the researcher clearly defines the

population of the research, and then selects from that
population a sample to study.

• The proposed research is then carried out on or with the

sample, instead of the entire population.

• Some research is carried out using entire

populations and some is carried out using
sample populations or samples of
populations.
 Populations and Samples
• Figure 10.4 (see the following slide) illustrates the
population, the sample and the unit the individual
or the case.

• As can be seen from the figure, the population is all

of the units in the figure.

• The sample is a small sub-set of the

population.
• The population is made up all many
individual units, cases or individuals.
 The Issue of Representation
• When using a sample of a population in a research project, the researcher
must clearly describe the sample.

• Then s/he must explain why that sample was selected and clearly describe
the sampling method, the means by which that sample was selected.

• In describing the sampling method, the researcher aims to establish how

representative the sample is of the research population.

• The key issue in sampling is this notion of representation.

• The concept of representation relates to the degree to

which a sample drawn from a population can be said
to be representative of the population.
 Inclusion and Exclusion Criteria

• Another issue to be addressed is that of the inclusion and exclusion

criteria.

• The inclusion criteria are the criteria potential participants must meet in
order to be included in the study.

• Exclusion criteria are the criteria on which participants will be excluded

from participation in the study.

• Outlining clear inclusion and exclusion

criteria is a good way of achieving clarity in terms of
what potential participants would be serve the
research.
 Probability Sampling
• Using probability sampling, the sample selected from the population is
claimed by the researcher to be representative of the population.

• It is of fundamental importance that the sample selected be

representative of the population of the study.

• The researcher using a probability sampling technique wants to claim that

the findings of research conducted with the sample are generalisable to
the entire population of the study.

• Probability sampling is based on the theories of mathematics of

probability.

• Probability sampling techniques include simple

random sampling, stratified sampling, systematic
sampling and cluster sampling.
• If used properly, probability sampling techniques yield
precise results while working with samples a fraction
the size of the original populations of research.
 The Sampling Frame
• The basic rule of probability sampling holds that each member of the population
has an equal probability of being selected for inclusion in the sample.

• As this is the case, the researcher, in order to engage in probability sampling, must
have a complete list (or map, or chart), of every member of the population.

• The sample is drawn from this list.

• This list is known as a sampling frame. A sampling frame is a list of element from
which a sample may be drawn, also called working population.

• A sampling frame is a list or a map or a chart in which

every member of the population of the study is
represented.
• Each member or item in the sample is randomly
selected from the population for inclusion in the study,
using the sampling frame.
 Random sampling and non sampling errors

• Random sampling error is the difference between the

sample result and the result of a census conducted using
identical procedures. Random sampling error occurs
because of chance variation in the scientific selection of
sampling units.

• Random sampling error is a function of sample size. As

sample size increases, random sampling error decreases.

• Systematic (non-sampling) errors result from non-sampling

factors, primarily the nature of a study’s design and the
correctness of execution.
 Simple Random Sampling
• Simple random sampling involves selecting a sample at random from a sampling
frame.

• Let us say that you want to study the population of your class, and there are
thirty students in your class.

• As it is a simple thing to get a complete list of the names of the students in your
class (a sampling frame), it is possible to engage in simple random sampling with
this population.

• The first thing to do is to make a list on a sheet of paper of all of the names of
the people in your class.

• Then tear off each name one by one, and place each of them into a hat or a box.

• Then select one name at a time.

• You are now engaging in simple random sampling.

 Systematic Sampling
• Systematic sampling involves selecting items at systematic or regular
intervals from the sampling frame.

• For example you might be working in a housing estate trying to establish

which brand of washing powder is used in each house.

• Your sampling frame is made up of all of the houses in the housing

estate.

• Suppose there are 500 houses in the estate and you begin at a random
starting point and then sample every third house, or every fifth house,
or whatever interval of house you decide on, until you
complete your sample.

• You are engaging in systematic sampling.

 Stratified Sampling
• A Stratified Sample is a sample selected based on some known
characteristic of the population, a characteristic which will have an impact
on the research.

• Using stratified sampling the researcher divides, or stratifies, the sample

selected for use in the research using the characteristic which s/he knows
will have an impact on the research.

• See the example ‘A study of religious practice among your classmates’ for
a simple and good explanation of these sampling methods. In the
example, the
researcher uses first a simple random
sampling technique, then a systematic
sampling technique, before finally deciding that a
stratified sampling technique in really required for this
particular study with this particular population.
 Non-probability sampling
• In non-probability sampling, the sample is selected to represent the population, but it
cannot be said to be representative of the population, in any statistical sense.

• The emphasis in non-probability sampling is on the capacity of a relatively small number of

cases to clearly and comprehensively illustrate the phenomenon under investigation.

• It often happens with social science research projects that it is not possible to produce a
complete list of the population; when this is the case, it is not possible to develop a
sampling frame.

• For example, a researcher might be asked to examine brand loyalty among consumers of
Cosmopolitan magazine. It wouldn't be possible to compile a complete list of consumers of
Cosmopolitan magazine.

• Without a complete sampling frame, it is not possible to engage in probability sampling, as

without a sampling frame it is not possible to guarantee that every member of the
population has an equally likely chance of being included in the study.

• The sampling approach used in such circumstances is non-

probability sampling. Non-probability sampling techniques include
judgmental sampling, quota sampling, snowball sampling and
convenience sampling.
 Judgemental or Purposive Sampling
• Using a judgmental or purposive sampling technique the
researcher decides, or makes a judgement, about who to
include in the research.

• The criterion for inclusion in the research is the capacity of

the participant to inform the research.

• Each person, or unit, chosen to be included must have a

contribution to make to the research.

• People chosen to be included in such a

sample would be key informants on the topic
under investigation.
Two different non-probability sampling techniques

Quota Sampling: Using a quota sampling technique the

researcher develops a sample of participants for the research
using different quota criteria.

Convenience Sampling: Using a convenience sampling technique

the researcher engages those participants in the research it is
easiest to include, for example people in the newsagents,
people in the supermarket and so on. The researcher knows
how many people to
include in the sample, then s/he continues to
engage people in the research until the
sample has been filled.
 Snowball Sampling

• Using a snowball sampling technique the researcher finds one participant in

the research, s/he conducts the research with that participant, and then
s/he asks that participant to recommend the next participant.

• Participants must fit the inclusion criteria for the research project.

• The researcher goes through the procedure with the second participant and
when finished, asks that participant to recommend another participant to be
included in the research.

• The researcher continues in this manner,

conducting the research with participants and
then asking each participant to recommend
the next participant, until the sample is
complete.
 Saturation Point

• In some qualitative research projects the research is allowed to unfold, and so the
population or the samples used are sometimes not defined from the outset.

• A frequently asked question in such circumstances is, when does the researcher
stop engaging participants in the research?

• The answer is when the researcher reaches saturation point.

• Saturation point is reached when the researcher, though continuing to explore

the phenomenon with participants in the research, no longer hears any new
thoughts, feelings, attitudes, emotions, intentions, etc.

• This is saturation point, the researcher is ‘saturated’ with

knowledge on the topic and continuing to engage
participants would not be useful, necessary, or ethically
sound, as engaging further participants in the research will
not add in any way to the knowledge generated.
 Internet sampling
• Internet surveys allow researchers to reach a large sample rapidly—both
an advantage and a disadvantage.

• Sample size requirements can be met overnight or in some cases almost

instantaneously.

• There are now so many Internet surveys, frequent Internet users may be
more selective about which surveys they bother answering.

• Another disadvantage of Internet surveys is the lack of computer

ownership and Internet access among certain segments of the population.

• A sample of Internet users is representative only of Internet users, who

tend to be younger, better educated, and more affluent than the general
population.
 Determining the sample size
• Slovin Formulae

n = N/(1+Nxe2)
n – sample size
N – population size
e = margin of error - 1- level of confidence
For e.g 95% level of confidence – e = 1-0.95=0.05