Sampling

• Sampling is a technique to select a part of the population under study

with a view to draw conclusions about the behaviour of or the
relationship and/or interrelationship between the elements of the
population.
Sampling: The necessity of sampling
Primary data can be collected from various sources –

1. Census –
All units of population are covered in a census survey.

2. Sample Surveys –
Only a few representative units are covered in a sample survey.
By selecting a representative sample of a population and using it to study the
relationship between different variables, we can generalize about the
population.
Universe and Population
• The universe can be defined as the aggregation of all elements
considered for sampling in a particular study.
Example: All universities
• However, the term population refers to the collection of elements
specified theoretically with specific characteristics and it has a
scientific connotation.
Example: All universities in India in 2010
Sampling Frame
• A representative sample is drawn from a sampling frame.
• A sampling frame can be defined as the specified population that has
been defined in a study from which a sample is to be selected.
• This is accomplished with respect to the objectives of the study.
Example: Consider a researcher needs to study the variables affecting
the professional effectiveness and satisfaction of managers in the steel
industry in 2012. For this researcher the sampling frame is -
List of all managers in the steel industry in 2012
Elements
• Unit of population or members of the population are known as the
elements of a sample. The elements in a population may not always
be individuals.
Example: If our population is ‘all executives in a software company
in India in 2013, -
Each executive would be considered as an element in the population.
Sampling Units
• Sampling units are elements considered or selected in a sample from
a population using a sampling procedure.
• Example: If we have considered 500 executives from a list of some
companies. All 500 selected executives are the sampling units in our
study since they have been sampled from the population using a
particular procedure. This sample selected is supposed to be
representative of the population.
Parameter and statistics
• Parameter refers to summary description of a variable with respect to
population.
• Example: We are interested in finding out the average salary of
executives in a software company. This can be referred to as
population parameter, since all the executives of the software
company is our population.
• However, if we have selected a representative sample from all the
executives in the software company, it would be known as sample
statistics.
Characteristics of Good Sampling Design
• A good sample design should give a representative sample
• Sample design should have small sampling error
• Sampling design should be economical
Need for Sampling in Empirical Research
• The following are the reasons for preferring sample surveys over
census surveys:
• The statistical tests and the formulae developed to arrive at sample
size from a finite population, the probability of arriving at correct
results is higher and hence, the preference for sample surveys.
• When the size of the population is very large, it consumes time and
hampers efficiency and quality of research. Hence, the researcher
prefers to use sample for his study.
• A researcher is constrained by budget, time and infrastructure, and
hence, a sample survey is recommended.
Classification of Sampling Techniques
• Sampling techniques could broadly be classified on the basis of the
very process and logic of selecting a unit from the population to be
included in the sample.
• There are two types of sampling:
1. Probability sampling
2. Non-probability sampling
Probability Sampling
• The techniques under this category is primarily based on the theory
of probability.
• The main focus is to assign an equal probability for each element in
the population.
• Example: If the population size is 1580. In probability sampling, the
chance of any member to be selected for the sample will be 1/1580.
It is generally believed that with this process, we get a representative
sample without any sampling bias or sampling error.
• The conclusion is that the results can be generalized to the
population.
Techniques under the Probability Sampling
• Simple random sampling
• Systematic sampling
• Stratified random sampling
• Multistage-cluster sampling
The Nature of Sampling
• The basic idea of sampling is that by selecting some of the elements
in a population, we may draw conclusions about the entire
population.
• A population element is the individual participant or object on which
the measurement is taken. It is the unit of study.
• A population is the total collection of elements about which we wish
to make some inferences.
• We call the listing of all population elements from which the sample
will be drawn the sample frame.
Why Sample?
• There are several compelling reasons from sampling:
1. Lower cost
2. Greater accuracy of results
3. Greater speed of data collection
4. Availability of population elements
What is a good sample?
Validity of a sample depends on two considerations:
• Accuracy
• Precision
Accuracy
• It is the degree to which bias is absent from the sample.
• An accurate (unbiased ) sample is one in which the underestimators
offset the overestimators.
• Systematic variance has been defined as the variation in measures
due to some known influences that cause the scores to lean in one
direction more than another.
• Increasing the sample size can reduce systematic variance as a cause
of error. However, even the large size won’t reduce error if the list
from which you draw your participants is biased.
Precision
• A second criterion of a good sample is precision of estimate.
• Researchers accept that no sample will fully represent its population in all respects.
However, to interpret the findings of research, we need a measure of how closely
the sample represents the population.
• The numerical descriptors that describe samples may be expected to differ from
those that describe populations because of random fluctuations inherent in the
sampling process. This called sampling error (or random sampling error).
• Sampling error is what is left after all known sources of systematic variance have
been accounted for.
• Precision is measured by the standard error of the estimate, a type of standard
deviation measurement; the smaller the standard error of estimate, the higher the
precision of the sample.
Concepts
How do we know we have a research
problem?
• Customer complaints
• Conversation with the company employees
• Deviation from business plan
• Success of the firm’s competitors
• Company records and reports
First step in the research problem process
• Identification / selection of the problem to be researched
Identification of a few problems and selection of one out of them,
after evaluating the alternatives against the selection criteria.
Sources of Problems:
o Reading
o Exposure to field situation
o Academic experience
o Brainstorming
o Research
o Intuition
o Consultations
o Daily experience
Identification/Selection of the problem
• Criteria for selection
Internal / Personal criteria- Researcher’s interest, researcher’s
competence, researcher’s own resource: finance and time
External Criteria or Factors- Research ability of the problem,
importance and urgency, novelty of the problem, feasibility,
facilities, usefulness and social relevance, research personnel
Definition /Formulation of the Research
Problem
1. Formulation is the process of refining the research ideas into research
questions and objectives

2. Formulation means translating and transforming selected research

problem/topic /idea into a scientifically researchable question. It
is concerned with specifying exactly what the research problem is.

3. Problem definition or problem statement is a clear, precise and succinct

statement of the question or issue that is to be investigated with the goal of
finding an answer or solution.
Defining problem results in clear research
objective
1. Symptom detection
2. Analysis of the situation
3. Problem definition
4. Statement of research objectives

There are two ways of stating a problem:

A. Posing question/s
B. Making declarative statement
Criteria of a Good Research -
• Clear and unambiguous
• Empirical
• Verifiable
• Interesting
• Novel and original
• Availability of guidance
Review of Literature
• Literature review is the documentation of a comprehensive review of
published and unpublished work from secondary sources of data in
the area of specific interest to the researcher
• The main aim is to find out problems that are already investigated and
those that need further investigation
• It is an extensive survey of all variables past studies relevant to the
field of investigation
• It gives us knowledge about what others have found out in the related
field of study and how they have done so.
Purpose of Review
• To gain a background knowledge of research topic
• To identify the concept relating to it, potential relationship between
them and to formulate researchable hypothesis
• To identify appropriate methodology, research design, methods,
measurement and technique of analysis
• To indemnify data sources used by the other researchers
• To learn how other structured their reports
Sources of Literature
• Books and journals
• Electronic database
• Govt and industry reports
• Internet
• Research dissertations/Thesis
Three categories of literature
Primary Secondary Tertiary
Reports Newspaper Indexes
Thesis Books Abstracts
Conference reports Journals Catalogues
Company report Internet Encyclopaedias
Government publications Govt Publication Dictionaries
Unpublished manuscripts Bibliographies
Difference between Amul and KMF business
model