Research Methodology for Beginners

Module 3 : Sampling and Methods of Data Collection

Objectives of the Module:

Explain a sampling design appropriate for a research study

Explain tool, design and procedure for collection of data

Knowledge Analysis:

Prerequisites: Essential Questions:

A student has Knowledge of : 1. What are the sources of

1. Qualitative and Quantitative data and how they are
methods of research selected?
2. Types and steps of different 2. How to collect the data for
methods research different type of research?

Instruction (Referencing):

Content Mode of Transaction

1. Concept of sampling and other concepts 1. Lecture with
related to sampling Presentation
2. Reading and reflective
2. Probability and non-probability samples,
thinking
their characteristics and implications 3. Assignment & Project
3. Tools of data collections, their types,
attributes and uses (questionnaire,
observation, interviews, scales and tests
etc

Learning hours : 12

Dr. Mahesh M. Joshi 1 © SRTM University, Nanded

 Research Methodology for Beginners

Module 3 : Sampling and Methods of Data Collection

INTRODUCTION

Research requires the accurate, reliable and valid information to solve the problem
in a system way, to verify the proposed hypothesis and draw out a generalizable
conclusions. The required information or pieces of information may be in the
different form like facts, numbers, symbols, etc. which are referred as ‘Data’. It is
gathered for reference and analysis with some specific purpose or in relevance of
research problem.

The types of data may be defined by several way on different basis.

Types based on nature of data:

1. Qualitative: The data which express trait, attribute, characteristic,

merit, preference or class of an element of population is called as
qualitative data. It is not perfectly quantifiable but observed
subjectively. Examples: Colours, goodness etc.
2. Quantitative: the data which show quantification of things or their
dimensions and express in numbers is called as qualitative data.
Examples : Height, weight, temperature etc.

There are two subtypes of quantitative data based on nature of variables:

1. Continuous: the data belonging to variable which shows gradual

changes and can be divided to finer level is called continuous data.
For example, the age may be measured finer level from years,
months, days, hours and beyond.
2. Discrete: the data belonging to variable which cannot be divided
to and expressed in perfect numbers is discrete data. For example,
the number of students will be always expressed in perfect
numbers like 20, 30, 12 etc. but never like 14.5, 16.58 etc.

Types based on the Sources:

1. Primary Data: It is the data collected by researcher with specific

purpose and which not used by others (It is first hand).
2. Secondary Data: It is the data that is readily available to use , may
be collected by others for other purpose.

Dr. Mahesh M. Joshi 2 © SRTM University, Nanded

 Research Methodology for Beginners

Module 3 : Sampling and Methods of Data Collection

Both the types have their own advantages and disadvantages. An

investigator has to fix the proportion of these resources according the
need to meet the purpose of the research.

Types based on level/scale of measurement

1. Nominal Scale: The data which can be classified only or describes

the name of category is called as nominal level data.

For example: Gender wise list of the student.

2. Ordinal Scale: The data which may be arranged in a rank or order

(ascending or descending) is called ordinal level data.

For example: Results in grades: A, B,C,D and E

3. Interval Scale: The data which show order and equal intervals in set
of values is called interval level data. The beginning point is
hypothetical (not absolute) zero.

For example: measurement of temperature

4. Ratio Scale: The data which have have intervals in the data with
perfect ratio and begins with a true zero value is called as ratio scale.

For example: Measurement of height, weight, age etc.

The data is gathered from various sources and it is gathered by using different
tools. Thus, researchers need the knowledge of:

1. Nature of required data and sources of data

2. Sources of data and methods of selecting sources of data
3. The tools and techniques of data collection.

On this backdrop, present module introduces with two important components of

research methodology i) Sampling and ii) Tools of data collection.

CONCEPT OF UNIVERSE, SAMPLE AND SAMPLING:

Generalizability is one of the most important characteristic of research. It shows

the applicability of research beyond the data and limited context of the study. The

Dr. Mahesh M. Joshi 3 © SRTM University, Nanded

 Research Methodology for Beginners

Module 3 : Sampling and Methods of Data Collection

concept of sampling is based on the idea of selected representatives of things or

subjects and draw conclusions applicable to the larger group of things or subjects.

Before studying the concept of sampling it is essential to understand the concept of

universe and population.

Universe:

Universe is set of entire units of objects (physical things) or subjects (human

being) selected for the study. For example, in a study of football players in India,
the set of entire football players is the universe of study.

It is not possible to involve the universe in above mentioned example for the study.
The researcher has to think on accessible portion of the universe.

Population:

The accessible portion of the universe is called as population of the study.

Sample:

Sample is single, independent unit of an object or a subject selected for study. It is

the portion of population that represents the population. It is set units of objects
(physical things) or subjects (human being) selected from population. Thus, it is
sub-set of universe and population. Samples are reliable and valid source of
information because they have all characteristics of universe and represent the
universe.

Sampling:

The process of selection of samples is called as sampling.

Sampling Design:

The plan of selection of samples that defines details of population, procedure of

sampling and size of sampling etc. is called as sampling design.

CHARACTERISTICS OF A GOOD SAMPLE:

A good sample should possess the following characteristics:

S: Systematic, Scientific, sizable

M : Measurable

Dr. Mahesh M. Joshi 4 © SRTM University, Nanded

 Research Methodology for Beginners

Module 3 : Sampling and Methods of Data Collection

A: Attainable, accurate

R: Representative, Relevant, Reliable

T: Testable

The selection may be affected by bias or errors. Thus, researcher must be cautious
about it.

SIGNIFICANCE OF SAMPLING:

Sampling saves the time, efforts and resources. It make feasible, practical and
empirical to the study. It limits the size of data and make it possible to handle,
Analyze and interpret the data, testing of hypothesis. The conclusions are
applicable to the population.

The process of selection is not simple. It is complex because it is difficult to select a

true, accurate and perfect representative group from population. If the population
is heterogeneous then specialized knowledge is required for sampling. There is
chance of influence of bias in sampling which may reduce accuracy, reliability and
validity of research.

TECHNIQUES OF SAMPLING:

There are two main techniques of sampling based on the principle of probability.

1. Probability Sampling
2. Non-Probability Sampling.

The probability sampling offers equal chance to each unit of the population to be
selected as sample and Non Probability sampling provides chance to specific and
limited units of population to be selected as sample to fulfill some specific,
predetermined purpose of the research.

There are various types of probability and non-probability sampling.

TYPES OF PROBABILITY SAMPLING:

Following are the main types of probability sampling

1) Simple random sampling

2) Systematic sampling

Dr. Mahesh M. Joshi 5 © SRTM University, Nanded

 Research Methodology for Beginners

Module 3 : Sampling and Methods of Data Collection

3) Random Stratified sampling

4) Cluster sampling

5) Multi stage sampling

Simple Random Sampling:

This method is generally applicable to homogeneous population. It offers equal

chance to entire elements of population to be selected. Lottery method is used for
selection of sample. This is simple method because it doesn’t need any specialized
knowledge.

Systematic Sampling:

The samples are derived after an equal interval from previous member. It is
applicable to homogeneous and finite population. All the units are equal in this
sampling because they are member of homogeneous population.

Random Stratified Sampling:

When the population is heterogeneous then it is necessary that sample also should
maintain same heterogeneity. This sampling method offer equal chance to each
member of small homogeneous group to be selected as sample within the entire
population. Small homogeneous existing within the population are called strata and
sampling method is called as random stratified sampling.

Cluster Sampling (Area Sampling):

Cluster sampling is used when population is distributed in different areas. It is

essential that sample should represent the characteristics of entire population.
When the selection of sample is on location basis then it is applied. Thus, this type
is also known as area sampling.

Multistage Sampling:

This type is applicable when the population is distributed in different stages. The
samples are randomly selected from different stages.

Dr. Mahesh M. Joshi 6 © SRTM University, Nanded

 Research Methodology for Beginners

Module 3 : Sampling and Methods of Data Collection

TYPES NON-PROBABILITY SAMPLING:

There are four main types of non-probability sampling:

1. Purposive Sampling
2. Convenience Sampling
3. Quota sampling
4. Snowball Sampling

Purposive Sampling:

This method is used to serve the specific purpose with the samples having specific
characteristics. It doesn’t offer the equal chance to the members of population for
being selected as sample but it is only limited to the some particular type of
members.

Convenience Sampling:

The selection of the samples conveniently available to serve the purpose of

research is called as convenient sampling. When the population is not countable or
widely sprayed and not easily available for study or required for longitudinal study
the convenient sampling is applied.

Quota Sampling:

The sample selection having predetermined proportion of subgroups existing in the

population is called quota sampling. The quota may be defined by two ways i) equal
may be allotted for each subgroup of population and ii) quota may be allotted with
equal to the proportion of each subgroup in the population.

Snowball Sampling:

The snowball sampling is process of selecting samples with the help of other
samples who are acquainted to them. The selection begins with few samples or
small group of samples and they help the researcher to identify and select available
respondents who meet the purpose of research.

Dr. Mahesh M. Joshi 7 © SRTM University, Nanded

 Research Methodology for Beginners

Module 3 : Sampling and Methods of Data Collection

TOOLS OF DATA COLLECTION

The first part of the module is on sampling and this second part belongs to tools of
data collection. The sampling frame is significant for knowledge of nature of sample
which is source of data. It is necessary to employ the appropriate tool and
technique to draw out the information relevant to research and its objectives. The
tool and technique of data collection varies according to purpose of research,
nature of data, nature of sample etc.
The tools may be enlisted as follows :
1. Observation
2. Interviews
3. Questionnaires
4. Rating Scales
5. Inventories & Tests

OBSERVATION :
Observation is one of the basic method of data collection. It may be defined as the
purposeful, careful of watching of something to draw out factual statements with
adequate evidences. Now a day various devices are available for observation. Thus,
the recording or noting of the observation became smooth and easy. There are
several types of observation according to nature of observation. they are like :
1. Structured and Unstructured

2. Participant and Non-participant

Structured and Unstructured Observation:

When it is required to specific aspect of the process, occurrence etc the structured
observation is preferred and when the researcher is unknown to process,
occurrence etc. unstructured observation may be preferred. The structured
observation doesn’t provide flexibility to observer. However, the degree of flexibility
is optimal in the unstructured observation.
Participant and Non-Participant Observation:
In some situation it may be required that the observer must be the part of situation
to collect factual data the participant observation is more suitable. The non-
participant is employed when the participation of observer may influence the
situation and data could not be gathered in natural setting.

Dr. Mahesh M. Joshi 8 © SRTM University, Nanded

 Research Methodology for Beginners

Module 3 : Sampling and Methods of Data Collection

INTERVIEW:
Interview is also one of the direct method of data collection. It is different than
conversation in various aspects. It is method to collect the responses to oral
enquiry. Thus, it is expected that an interview must be planned, systematic and
purposeful.
There are various type of interviews according to purpose, nature and process of
interview.
Structured Interview:
Structured interview is comprised of predetermined selected questions and
sometimes the responses also may be restricted to limited options. The structured
interviews are useful to collect the information on specific points and avoid
unnecessary information which may be given by the respondent.
Semi-Structured Interview:
The interview having flexibility in both questions and responses is semi structured
interview. It provides some flexibility to interviewer as well as respondents.
Unstructured Interview:
The unstructured interview provides high degree of flexibility to both interviewer
and interviewee in questioning and responses.
QUESTIONNAIRE:
Questionnaire is tool collect the responses in written form to the set of
predetermined questions. It is a device employed on respondents competent to
give responses in written form. It is tool which may be employed to the
respondents sprayed over wide jurisdiction.
There are two important type of the questionnaire based on the types of questions:
i. Closed Questionnaire :
The closed questionnaire is comprised of the questions having restricted
responses. Respondents has to choose one of the option from the list of
predetermined responses.
This type is useful to avoid the unnecessary information which may be
provided by respondents
ii. Open Questionnaire:
The responses to the questions in this type are not restricted to
predetermined responses. The respondents can note their responses with
explanations or information in detail. It is useful when the subjective and
in-depth information is required for research.

Dr. Mahesh M. Joshi 9 © SRTM University, Nanded

 Research Methodology for Beginners

Module 3 : Sampling and Methods of Data Collection

RATING SCALE:
Rating scale is also an enquiry form employed to collect the judgements, opinions
regarding some situation, statements etc. Rating allows the quantification of
responses. For example, Goodness of the something may be asked and respondent
has to choose one of the option from set of responses like : A, B, C, D, E which
shows gradual increase or decrease of goodness.
SUMMARY:
The data is gathered from various sources and it is of different types. Thus,
researchers need the knowledge of Nature of required data, Sources of data,
methods of selecting sources of data and the tools and techniques of data
collection.
The concept of sampling is based on the idea of selected representatives of
things or subjects and draw conclusions applicable to the larger group of things or
subjects.
The probability method gives equal chance of selection to each member of
population to be selected as sample. The different probability methods are
applicable according to constitution of population.
The non- probability methods of sampling allows only those members to be
sample who are competent to serve the predetermined criterions and objectives of
the research.
The tools are important for collection of reliable and valid data. The
researcher has to device the tools according to nature of problem, population and
nature of required data.

Dr. Mahesh M. Joshi 10 © SRTM University, Nanded