The Meaning

3.1 Population refers to a group of individuals, events, text or objects that have one or more
characteristics in common that are of interest to the researcher.
3.2 A sample is a group of items taken from the population for examination. This means that a
sample should represent the population. A sample is simply a small part or quantity, which has
been selected to represent the whole population or universe. As it is obvious even to the rookie
researcher, it is most often difficult to study an entire population especially when that population
is very large. For example, it is difficult to study the total population, of a typical kenyan
university, village, clan, town or city. While it may be difficult or even easy for one to do this in
some kinds of studies or populations, science has offered us a means of studying large
populations without recourse to studying them individually. A scientifically selected sample is
normally regarded as a good representation of the population. According to Sobowale (1983:38),
once the first element of a sample is picked through a table of random numbers (TRN) all other
elements of that sample become known. A sample can either be “probability or non-
probability based.

3.3 TYPES OF SAMPLING DESIGNS

a) Probability Sampling is one in which every element; unit or portion has a known probability
of being members in the sample. Examples are:
i. Simple Random Sampling
Each subject has an equal chance of being selected or included in the research study. E.g if you
would like to select a sample of 20 individuals from a population of 100, each individual would
be given a number. The numbers are then placed in a container and one is blindfolded to pick
one number at a time until the required sample of 20 is reaching.
ii. Systematic random Sampling
It is a modification of simple random which ordinarily is less time consuming and easier to
implement.one is required to come up with a sampling frame first- a list of all elements in the
population then obtain a sampling interval by dividing the population size by the required size.
SI=population size
Sample size
Then select the starting point in the sampling frame by choosing the initial number.
 iii. Stratified Sampling
It is used where the population is not homogenous. Elements may be grouped in different
categories called strata then the population is grouped according to characteristics. All the
characteristics are important to the researcher. The researcher then selects the sample by either
simple or systematic sampling proportionate to the existence of the population.
iv. Clustered Sampling
It is used where the population is sparsely distributed. The population is grouped into clusters
which become units of sampling. A random sampling technique is used to select a sample of
clusters then the elements of the clusters are investigated.
v. multi-stage sampling
It is mainly used when drawing samples from large and diversified population. The researcher
groups up the population with the already defined administrative boundaries.
Stages
1. Break down the population into naturally occurring clusters or standard regions.
2. Divide the sample among regions according to their population.
3. Select a smaller number of districts or villages at random with each selected
region.
4. Select individual elements from within each sampled clusters using the random,
stratified or systematic random sampling. (Individual households)
5. Individuals in each household are selected at random

b) Non probability
1. The purposive technique is adopted when you require a particular characteristic from a
sample e.g. Particular sex, age, socio-economic group, ethnic profession etc. those who
fail to meet the criterion of the study are eliminated. It is selected with the knowledge that
it is not representative of the population.
2. Convenient sampling
It consists of collecting units that are convenient to the researcher. They are a collection of
readily available and accessible subjects for the study.it is also referred to as volunteer or
accidental sampling e.g a radio program may ask a few questions to radio listeners and tell them
to take the answer to certain place. Such a sample may refer to a convenient because only those
who have radios and who listen to them will answer those questions. A researcher may select
anyone who is convenient e.g a roommate or a neighbour. Volunteer samples have been shown
to be inappropriate in scientific research even though electronic media uses them in getting views
about the program, they are not representative and the results cannot be generalised.
3. Snow ball
Subjects with the desired characteristics are identified using purposive technique. With few
identified subjects they will name others that they have required characteristics until the
researcher gets the numbers required for the study

3.4 Reasons for Sampling

In all we have said so far we have alluded to some reasons why we sample. Whether a
population is finite (that is determinable) or infinite (indeterminable or countless), the process of
drawing a sample from it can be difficult, expensive and time consuming. If it can be so for the
sampling process how much more for studying the whole population. So we decide on sampling
for various reasons namely,
(1) There are similarities among the elements that make up a population we wish to study. For
this reason a study of a few of the elements within the population can give us sufficient
knowledge of what obtains in the entire population that enables us to generalize after observing
an adequate sample , for the rest of the population.
(2) In most studies, it is practically impossible to carry out a complete and comprehensive study
of the population because of the nature and pattern of distribution or dispersion of the elements
of the population. In this circumstance, the only practical way to estimate the population
characteristics is to draw a sample. For an individual it is near impossible for him/her to study
the whole population of say Nairobi city. Within the time frame and resources available to
him/her this might seem a mission impossible.
(3) Arising from the above, because sampling enables us to deal with a part of the population, in
an era of unavailable or limited resource; it is cheaper and reasonable to study a sample rather
than an entire population.
(4) With a sample we are more thorough and more able to provide thorough supervision than if
we had to study the entire population.
(5) It also enables us to obtain quicker result in our study than if we had had to cover the whole
population.
(6) Interestingly sampling provides greater accuracy because personnel of higher quality can be
recruited and given intensive training for the study than when working with the whole
population.
Overall, sampling makes possible the conduct of otherwise impossible studies by selecting
representative units from the population, so that the result obtained can be used to draw inference
on the entire population.

3.5 Sampling Frame

A sampling frame is the “list of elements from which a probability sample is selected”. (Babbie,
1973:89). It is the real list of the units to be sampled from which the sample or some stage of the
sample is selected. There is need for the appropriate selection of samples so as to get the
appropriate information required for describing the population from which the sample has to be
drawn. A sampling frame could be the list of schools in a Local Education Authority, List of
students in a class, list of houses on a street, list of phone numbers on telephone directory etc.
from which a researcher wants to draw the sample for research. A sample selected without an
adequate sampling frame is not likely to be a true representative sample of the population of
interest.

3.6 sampling error

ASSIGNMENT
Distinguish between research questions and hypotheses. If you were to carry out a research study
in an area of journalism, which approach would you utilize to achieve your objectives: RQ or H?
Give reasons for your answer

REFERENCES/FURTHER READINGS
Ajala, O. V. (1996). Scholarly Writing Guide For Researchers. Ibadan: MayBest Publishers.

Kerlinger, F. N. (1973). Foundations of Behavioural Research.New York: Holt, Rinehart and

Winston Inc.
Tejumaiye, A. J. (2003). Mass Communication Research.(An Introduction).Ibadan: Sceptre
Prints Ltd.

Wimmer, R. D. and Dominick J. R. (1987).Mass Media Research.An Introduction.Belmont,

California: Wadsworth PublishingCompany.