2. SAMPLING…
• Sampling is the method of studying a whole population on the basis of the study of
samples drawn from it. A sample is a representative subset of a whole population.
• It represents the entire population in respect of the specific characteristics under
investigation. Study of the sample gives information about the whole population. This
is called statistical inference.
• Sampling involves three principal steps, namely
(a) selection of samples
(b) collection of information about them and
(c) making inference about the whole population.
3. SAMPLING ADVANTAGES AND DISADVANTAGES
• It is relatively less expensive.
• As the number of enumerators required is
less, more efficient and better trained
personals can be employed and this will result
in the improvement of the quality of the data.
• Since the number of enumerators are less,
more sophisticated instruments can be used.
• If destructive tests are involved in the
collection of information, sampling alone can
be adopted. For example, in a study of the
toxicity of poisonous chemicals on a particular
breed of animals, census method is
unacceptable.
• It requires the services of experts,
otherwise incorrect or misleading results
will be obtained.
• In this method, selection of appropriate
methods of sampling is necessary.
• In case the units of population are
spread over a large area, this method
cannot be used.
• In case the size of samples is small,
sampling does not provide true
representation of the population.
4.
5. sampling techniques
• Sampling techniques are of mainly two types
• Random/Probability sampling and Non Random/Non Probability sampling
Random sampling
• Here, the selection of sample units is absolutely a matter of chance or probability. For this reason
it is also termed chance selection or probability sampling.
• It is the most commonly used sampling method. In it every member of the population has an
equal chance of being selected.
Advantages of random sampling
(i) It does not require detailed information about the population for its effectiveness
(ii) It provides estimates which are essentially unbiased and have measurable precision
(iii) Evaluation of the relative efficiency of various sample designs is possible only when probability
sampling is applied.
6. Methods of random sampling
• There are two main kinds of random sampling, namely simple or unrestricted random sampling
and restricted random sampling.
• Simple random sampling :This is the random sampling method in which all items of the
population get an equal chance of being included in the sample. The selection is free from
personal bias. To ensure randomness of selection, either the lottery method or table of random
numbers is used.
a) Lottery method: This is the random sampling method in which all the items of a population are
numbered or named on identical paper slips and then such slips are randomly selected in lots.
Selection is blind fold with replacement until the desired number of units are obtained.
b) Tables of random numbers: These are tables that consist of a sequence of randomly chosen
digits from 0 to 9,arranged in the form of all possible combinations. Each digit has a probability of
0.1 to appear in a particular position. So, approximately equal frequencies of all combinations may
be obtained. Tables of random numbers can be used to select units at random from a population.
8. • Restricted random sampling: This is the type of random sampling in which certain
restrictions are imposed while sampling.
• There are mainly three types under this, stratified sampling, systematic sampling and
multistage sampling
Stratified sampling: The population is first divided into homogenous groups called strata,
then a specific number of random samples drawn from each stratum, and finally all the
samples thus selected pooled together. E.g. one sample at random from each plot of a field.
9. • Systematic sampling: systematic random sampling is also called quasi random sampling.
Here the population is arranged in order, the first item is selected at random and further
items are selected at specified intervals.
10. • Multi-stage sampling or cluster sampling: This is the sampling procedure carried out in
several stages. In this case the population is divided into several groups, called
clusters, and a desired sample is selected from them to represent the whole population.
• In multi-stage sampling the population is first divided into several first level sampling
units. From them first stage samples are obtained by a suitable method. Then, each
sampling unit is divided into second level sampling units and from them second stage
samples are obtained. In this manner further samples may be obtained, if necessary.
11. Non-random sampling
• Non-probability sampling is defined as a sampling technique in which the researcher
selects samples based on the subjective judgment of the researcher rather than random
selection
• It is a sampling method in which not all members of the population have an equal chance
of participating in the study
Advantages of non-random sampling
• Non-probability sampling is most useful for exploratory studies like a pilot survey
(deploying a survey to a smaller sample compared to pre -determined sample size).
• Researchers use this method in studies where it is impossible to draw random probability
sampling due to time or cost considerations.
12. Judgement sampling: Here the choice of sample items depends exclusively on the discretion
of the investigator. The investigator uses his judgement in the choice and includes those
items in the sample which he thinks are most typical of the universe with regard to the
characteristic under investigation.
Convenience sampling: In this, each units are selected only for convenience. A unit selected
in this way is called a chunk. The results obtained following convenience sampling can hardly
be representative of the population. They are generally biased and unsatisfactory. It is often
used for making pilot studies.
13. • Quota sampling: Here, quotas are set up according to some specific characteristics, such as
so many in each of several flower colour groups, so many in each duration group, etc.
• There are two types of quota sampling methods
1) Controlled Quota Sampling: If the sampling imposes restrictions on the
researcher’s/Statisticians choice of sample, then it is known as controlled quota sampling. In
this method, the researcher can be able to select the limited samples.
2) Uncontrolled Quota Sampling: If the sampling does not impose any restrictions on the
researcher’s/Statisticians choice of sample, then it is known as uncontrolled quota sampling.
In this process, the researcher can select the samples of their interest.
14. • Snowball sampling: Snowball sampling is where research participants recruit other
participants for a test or study. It is used where potential participants are hard to find.
It’s called snowball sampling because (in theory) once you have the ball rolling, it picks
up more “snow” along the way and becomes larger and larger