Professional Documents
Culture Documents
CSIT745 - Research Methodology
CSIT745 - Research Methodology
Module-I
Notes
ity
Key learning Outcomes
At the end of this module, participant will be able to:
rs
●● Identify different approaches of research
●● Elaborate planning a research project
●● Analyze Application of Research
ve
Strucute:
Unit-1.1: Introduction to Research Methodology
1.1.1 Definition & Objectives of Research
1.1.2 Characteristics of Research
ity
Unit Objectives:
At the end of this unit, the participant will be able to:
rs
●● Identify objectives, significance and types of research
ve
What is research?
Authors and management gurus have defined research in different ways. Usually,
a research is said to begin with a question or a problem.Research is defined as the
generation of new concepts, methodologies, and understandings through the creation
of new knowledge and/or the creative application of existing knowledge. This could
ni
include synthesising and analysing previous research to the point where it produces
new and innovative results. By applying research we are able to find out the solutions of
a problem with the application of systematic and scientific methods.You could talk about
experimentation or innovation. You could use the word “risk” to describe the element of
danger that comes with discovery. It is possible that investigation will lead to analysis. It
U
is possible that you will conduct tests to prove your hypothesis. You could simply state
that this work is unique and never seen. You could discuss what new knowledge will be
gained because of your work.
You could talk about a new method or a new data source that will result in a
ity
breakthrough or a small improvement over current practise. You could state that it is a
prerequisite for development in the sense of “research and development.”
●● Clifford Woody:
Research comprises “defining and redefining problems, formulating hypotheses or
suggested solutions; collecting, organizing and evaluating data; making deductions and
reaching conclusions; and finally, carefully testing the conclusions to determine whether
they fit the formulated hypotheses”.
(c
Research Objectives
General Objectives
of a study’s goal. In other words, by the end of your studies, you will have a general
idea of what you want to accomplish. For example, if you want to investigate an Notes
ity
organization’s contribution to environmental sustainability, your broad goal could be to
investigate sustainable practises and the organization’s use of renewable energy.
Specific Objectives
Specific objectives define the primary aim of the study. In most cases, general
rs
objectives serve as the foundation for identifying specific goals. In other words, specific
objectives are defined as general objectives that have been broken down into smaller,
logically connected objectives. They assist you in defining the who, what, why, when,
and how of your project. It’s much easier to develop and carry out a research plan once
you’ve identified the main goal.
ve
Take, for example, a study of an organization’s contribution to environmental
sustainability. The specific goals will be as follows:
●● Determine how the organisation has changed its practises and adopted new
solutions throughout its history.
ni
●● To determine the impact of new practises, technology, and strategies on
overall effectiveness.
●● The study should be condensed. This means that a researcher’s findings should
be made available to other researchers so that they don’t have to repeat the same
research.
●● The findings should be repeatable. This asserts that previous research findings
should be able to be confirmed in a new environment and different settings with a
m
ity
●● The research findings should be presented in formats that are most useful
to administrators, decision-makers, business managers, or members of the
community.
rs
Although the names sound similar, both Research methods and Research
methodology are different, as explained below:
Research Methods
ve
The various procedures, schemes, steps, and algorithms used in research are
known as research methods. The term “research methods” refers to all of the methods
used by a researcher during a research study. They’re primarily planned, scientific, and
value-agnostic. Observations, theoretical procedures, experimental studies, numerical
schemes, statistical approaches, and so on are all examples of these. We can use
research methods to collect samples, data, and come up with a solution to a problem.
ni
Business and scientific research methods, in particular, demand explanations based on
collected facts, measurements, and observations, rather than solely on reasoning. They
only accept explanations that can be verified through experiments.
U
Research Methodology
A systematic approach to solving a problem is known as research methodology. It
is a science that studies how research should be conducted. Research methodology
is essentially the procedures by which researchers go about their work of describing,
explaining, and predicting phenomena. It can also be defined as the study of methods
ity
ity
Unit Objectives:
At the end of this unit the participant will be able to:
rs
etc.
●● Define hypothesis formulation and its types
●● Describe various hypothesis errors
ve
1.2.1 Research Criteria and Research Process
Research Criteria
●● The research’s purpose should be clearly defined, and common concepts should
be used.
●●
●●
ni
The research procedure should be sufficiently described. detail to allow another
researcher to continue the research for further advancement while maintaining the
integrity of what has already been accomplished.
The research’s procedural design should be meticulously planned to produce
U
objective results.
●● The researcher should be completely honest about any flaws in the procedure
design and estimate their impact on the findings.
●● The data analysis should be sufficient to reveal its significance, and the analysis
ity
methods used should be appropriate. The data’s validity and reliability should be
double-checked.
●● Conclusions should be limited to those that are supported by the research data
and for which the data provide an adequate foundation.
●● If the researcher is experienced, has a good research reputation, and is a person
m
Research Process
In the early decades human inquiry was primarily based on the examination of
)A
one’s own conscious thoughts and feelings that means the observation of any one
and understanding through the logical discussion to seek the truth. This procedure
was accepted for a millennium and was a well-established conceptual framework for
understanding the world. The knowledge seeker was an integral part of the inquiry
process. With time, this part was changed. The Scientific method introduced several
major components in research procedure like Objectivity.
(c
ity
influence the research, it should be free of the researcher’s or management’s personal
or political biases.
For example, we may find out that our topic is too broad and needs to be narrowed,
sufficient information resources may not be available, what we learn may not support
our thesis or the size of the project does not fit the requirements.
rs
There are main nine steps of research process that are followed at the time of
designing a research project. They are as follows.
ve
Step 2: Development of an Approach to the Problem
ni
Step 6: Report Preparation and Presentation
Advantages of deductive:
Notes
ity
Inductive: Inductive approach or inductive reasoning, starts with observations
rs
and theories are proposed which are related to the end of the research process as
a result of the observations. It involves the search for pattern from observation and
the development of explanations – theories – for those patterns through series of
hypotheses. In inductive studies, at the beginning of the research neither theories nor
hypotheses would apply and the researcher is free to alter the direction for the study
ve
after the research process had commenced.
Inductive approach does not imply disregarding theories when formulating research
questions and objectives. This approach helps to generate meanings from the data set
collected in order to identify patterns and relationships to build a theory. This approach
is mainly based on learning from experience. Previous patterns, resemblances and
ni
regularities are observed in order to reach conclusions or to generate theory.
U
ity
The figure below illustrates the main differences between abductive, deductive and
inductive reasoning:
)A
(c
ity
Abductive approach starts with ‘surprising facts’ or ‘puzzles’ and the research
process is devoted in their explanation. ‘Surprising facts’ or ‘puzzles’ may emerge when
a researcher encounters with an empirical phenomena that cannot be explained by the
existing theories. In that approach, researcher searches for the ‘best’explanation among
many alternatives to choose. Researcher can combine both, numerical and cognitive
rs
reasoning for explaining ‘surprising facts’ or ‘puzzles’.
ve
people. Open-ended questions are asked( to gather as much information as possible
from the sample )in a manner that encourages answers that lead to another question or
a group of questions. The below methods are used for qualitative research:
1. One-to-one interview
2. Focus groups
3. Ethnographic research
4. Content/Text analysis
ni
U
5. Case study research
helps to obtain more accurate results. This research uses close-ended questions
because the researchers are typically looking to gather statistical data. It involves use
of data collection tools likes online surveys, questionnaires, and polls .There are various
methods of deploying surveys or questionnaires. Online surveys helps surveyor to
reach large number of people or smaller focus groups for different types of research
that meet different goals.
m
When researcher wants to bring about a positive change in the lives of the research
subjects, it is sometimes described as emancipatory. It is not a neutral stance.
Researchers want their research to directly or indirectly result in some kind of reform,
for which they involve the group being studied in the research at all stages, so as to
avoid further marginalising them.
The researchers may adopt a less neutral position than that which is usually
(c
required in scientific research. This might involve interacting informally or even living
amongst the research participants (the co-researchers). The searching of the research
can be reported in more personal terms, often using the precise words of the research
participants. Notes
ity
1.2.3 Hypothesis Formulation & Its Types
What is Hypothesis
A hypothesis is a statement about the research problem and may or may not be
rs
true. It is a tentative and testable relationship between at least two variables that require
research methodology for testing its authenticity. You are required to test the hypothesis
and ascertain the truth in way of Research. Theory and hypothesis are completely
different. Unlike hypothesis, a theory is a tested, well-substantiated, complete
explanation for a set of proven factors.
ve
The term “hypo” means “subject to verification” and “thesis” means “statement
about the solution of a problem”. Thus, “hypothesis” is “ a tentative statement, which is
subject to verification and states the solution of a problem”. Hypothesis offers a solution
of the research problem that must be verified empirically
Illustration:
ni
A renowned asset management company is under the process of launching a new
Mutual Fund. The fund is aimed at catering to the wealth-building needs of the Indian
U
middle-class population. The company wants to conduct a detailed market research on
the popular choices of investment options in India. It recruits a research team, which
comes up with the research problem:
H0: The Indian middle-class population does not invest in Mutual Funds
H1: At least 20% of the Indian middle-class population invests in Mutual Funds
H2: Less than 20% of the Indian middle-class population invests in Mutual Funds
m
Note that all three hypotheses have been phrased in a manner so that any one is
correct. However, note all three cannot be correct or incorrect simultaneously. Also note
that in order to make an effective decision, you need to select the first act if H1 is true
and the second act if H2 is true. The decision of launching or not launching the Mutual
)A
Fund depends on whether H1 or H2 is true. The hypothesis H0 suggests that the Indian
middle-class population is completely indifferent towards Mutual Funds
Thus, the asset management company chooses to launch the mutual fund if H1
is true, i.e. if it is established from the market research that at least 20% of the Indian
ity
Formulation of Hypothesis
Variables in a Hypothesis:
rs
Let us now study the types of variables in hypotheses.
ve
ni
It can be clearly understood by an example described below
(cause) .
Its value depends upon the value of
an independent variable .
The variable changes due to
modifications in an independent
variable.
(c
ity
studied by researchers to find out if
it changes the relationship between
the independent and dependent
variables.
Control Age group intelligence Since all variables in a phenomenon
cannot be studied simultaneously, a
rs
few must be neutralised to prevent
them from having a moderating
effect on the relationship between
the independent and dependent
variables .
ve
Intervening Learning Attitude Fields of Prior A factor that affects the relationship
process of interest learning between the independent and
students of of dependent variables but cannot be
students students seen, measured, or manipulated
ni
Steps of formulating a hypothesis: There are basically 5 steps of formulating.
Step 1: Define variables: List all variables of the research problem to determine
what do we want to test.
Step 2: Study all variables in depth: Classify all listed variables to determine the
U
independent and dependent variables
Step 4: Identify study population: Study the whole group of individuals (population)
ity
Summary
●● A hypothesis is a statement about the research problem and may or may not be
true.
)A
●● In type II error, you wrongly reject the alternative hypothesis and wrongly accept
Notes the null hypothesis
ity
●● A hypothesis comprises at least two of the following variables - independent,
dependent, moderator, control, and intervening
Activity
Consider a scenario where you have been asked to conduct Research on
rs
a problem of your interest. Structure haveto be tested in terms of an appropriate
hypothesis. Also state the possible errors and recommendations associated with the
hypothesis.
Types of Hypothesis
ve
Depending on the daily research work hypothesis classified into several categories
likes
ni
1. Higher rate of crime against women in the society due to the higher rate of
unemployment. Here the dependent variable is (Higher rate of crime against women
in the society) and the independent variable (higher rate of unemployment).
2. Higher the use of fertilisers, higher would be the agricultural productivity of farms
U
Identify the dependent and independent variable.(do yourself)
1. Lower the use of fertiliser, modern equipment and high quality seeds, lower
ity
that there are less parametersinvolved.This is primarily done to determine whether the
varibales are really ordinary or have come up due to errors in sampling.
If we take an example then we can understand easily like the literacy rate does
not affect crime rate in the society . Here in theoretically we can say the independent
variable is literacy rate and the dependent variable is “crime rate in the society”, but
practically there is no relation between two variables because no changes have been
occurred.
(c
Ex2: Illiteracy has nothing to do with the rate of female infanticide in the society .
(Test yourself)
Alternative hypothesis:
Notes
ity
If the hypothesis under test is proved wrong and rejected, then another
hypothesis(Alternative hypothesis) is formulated to replace the wrong one and tested to
obtain the desired results . Just like
H0 (Null Hypothesis): Advertisements do not affect the TRP (Television Rating Point)
of soaps
rs
H1 (Alternative Hypothesis): Advertisements decrease the TRP of soaps
A null hypothesis is used for a reverse strategy, i.e. to prove it wrong so that you may
confirm that there definitely is a relationship between the variables . Just like
H0 (Null Hypothesis): Academic achievements are not related to intelligence
ve
H1 (Alternative Hypothesis): Intelligence enhances academic achievements
ni
because a minor error in a step may corrupt the entire process
decisions. Note that chances of major errors lie in the choice
common errors involved in the choice of hypothesis are Type1 &
describe bellow.
extremely carefully
and lead to wrong
of hypothesis. The
Type 2 , which are
U
Error Explanation
Type I H0 wrongly rejected
H1 wrongly accepted
ity
Illustration:
A renowned channel of children’s conducts a market research to find out the need
for advertisement. The research team constructs the below hypotheses:
m
Note: that in case the null hypothesis H0 is true, the company should advertise. On
the contrary, in case the alternative hypothesis H1 is true, there is no need to advertise.
A hypothesis must be framed very carefully. There are high possibilities of errors
during each step of constructing a hypothesis. Note that hypotheses are constructed
in such a way that they cannot both be correct or incorrect simultaneously. To avoid
(c
erroneous decisions, we are required to select the first action (first hypothesis), if the
first action is true, and the second one (second hypothesis), if the second action is true.
ity
universe. The first chance of error lies with the first step of constructing a hypothesis -
determining the relationship between the states of nature and the recommended action.
The second chance of error may occur if the research findings stand erroneous w.r.t the
state of nature.
The below table would help us understand the relationship between the states of
rs
nature and the recommended action. This comprises two plausible states of nature and
two alternative courses of action.
ve
H0 True No error Type I error
H1 True Type II error No error
From the above table, it is evident that, given that the relationship between
the two states of nature and the two actions is true, there is no error if we arrive at
ni
the conclusion that H0 is true. Likewise, there would be no error if we arrive at the
What would happen if the team accepts the wrong hypothesis? Would the
company take a wrong decision? Let me explain with the help of the two scenarios that
U
may arise.
wrongly accepted
H1 H1 wrongly rejected and H0 Type II Unnecessary costs
wrongly accepted incurred due to
advertisements
Our research area and the methodology what we choose must address the
following questions: What we plan to accomplish, why we want to do it and how we are
going to do it.
ity
A research proposal is comprised of:
●● Title
●● Introduction
●● Literature review
rs
●● Methodology
●● Plan - time frame and schedule of activities (Gantt chart)
●● Budget
●● Details of research team (signed CV)
ve
Title
The name of a research proposal must be crisp and describeable. Try to make it
informative but catchy. An effective title not only increases the reader’s interest, but also
brings him/her favourably towards the proposal.
Introduction
ni
Through the introduction researcher have to produce the required skeletal structure
for the issues in research. How to frame the research problem is perhaps the biggest
U
problem in proposal writing.
The introduction typically begins with a general statement of the problem area, with
a focus on a specific research problem, to be followed by the rational or justification for
the proposed study. The introduction generally covers the following elements:
ity
●● State the research problem, which is often referred to as the purpose of the
study.
●● Provide the objectives of the study. The objectives identified should be general
as well as specific.
●● Identify the rationale of the proposed study and clearly indicate why it is worth
doing.
m
Literature Review
Notes
ity
The intention of the literature review is to provide adequate background
information.
Review should be brief, and indicate relevant related Research that had or is being
conducted (references should be included) and the review committee have idea about
the various on goingprojects.
rs
The literature review serves several important functions:
●● Gives credits to those who have laid the groundwork for the proposed
Research.
●● Demonstrates about the knowledge of the research problem.
ve
●● Demonstrates about understanding of the theoretical andResearch issues
related to the research question.
●● Shows your ability to critically evaluate relevant literature information.
●● Indicates the ability to integrate and synthesize the existing literature.
●● Provides new theoretical insights or develops a new model as the conceptual
●●
ni
framework for our Research.
Convinces our reader that our proposed Research will make a significant and
substantial contribution to the literature (i.e., resolving an important theoretical
issue or filling a major gap in the literature).
U
Methodology
The Research Committee depends on methodology to determine how we plan to
tackle our research problem. It will provide our work plan and describe the activities
necessary for the completion of our project.
ity
calculation.
●● Study instruments - What kind of measuring instruments or questionnaires do
you use? Why do you choose them? Are they valid and reliable?
●● Data collection - How do you plan to carry out your study? What activities are
involved? How long does it take?
(c
●● Data analysis and interpretation – this includes plans for processing and
coding data, computer software to be used (eg Statistical Package for Social
Sciences / SPSS, EPI-INFO, etc), choice of statistical methods, confidence
ity
●● Ethical considerations – It is necessary to submit our research proposal to
the Ethical Committee where we work and also where we plan to conduct our
Research. Depending where we are working and the type of Research we are
planning to conduct.
Plan
rs
Planning for the research proposal should include the time frame and activity
schedule for the proposed Research.
ve
●● purchasing and obtaining relevant consumables and facilities needed to
conduct the study
●● conduct of study
●● analysis of data
●● writing up of project report
ni
The activity schedule is essential for effective monitoring of project. It should list the
time frame for major activities, and include milestones. A most effective way of plotting
the activity schedule is by using the Gantt Chart
U
Budget It is essential to request for an o Provide a total and yearly
adequate budget for the study you breakdown of the budget
are planning to conduct: needed
Give appropriate estimates of costs depending on the different areas, eg: travel
and transportation, consumables, salaries, services, rentals, equipment, utilities,
ity
repairs, etc.
Provide adequate justification, especially for costly items
Research team:
●● Identify all expertise required for your Research
m
ity
Introduction
Do you know that we, human beings, are the most intelligent living beings on
earth? Thanks to our stellar intelligence, we can utilise the knowledge that has been
preserved or accumulated over ages. Human knowledge comprises three equally
crucial phases –
rs
●● namely preservation,
●● transmission,
●● and advancement.
Research helps in advancement of knowledge so that an updated knowledge
ve
reservoir is created and transmitted for the benefit of mankind. Human beings build
upon the recorded and accumulated knowledge of the past and this constant endeavour
of adding to the vast reservoir of knowledge in every possible field makes advancement
of human race possible. We, as a researcher, need to ensure that considerable work
has already been done on topics related to our field of investigation. We are required
to be familiar with all previous projects, Research, and theory related to the research
updating existing knowledge pools, related to specific disciplines, which add on to and
enrich fields of Research.
that solves the problem sufficiently without the need of further investigation - this
initiative avoids the replication of Research
●● Review of Literature serves as prominent sources for hypothesis - researchers can
formulate research hypotheses based on available studies
●● Review of Literature suggests data sources, methodology, and statistical
techniques apt for the solution of the research problem
(c
Sources of Literature
Notes
ity
What documents would you consider the sources of literature?
Sources Examples
Lists and indices of books, textbooks, and Cumulative Book Index
handbooks Book Review Index
Books Review Digest
rs
Subject Guide to Books
National Union Catalogue
P e r i o d i c a l s ( p u b l i c a t i o n i s s u e d i n Yearbook
successive volumes, at regular intervals) Documents
ve
Almanacs
Cumulative Book Index
International abstracts
Journals
Newspapers
Magazines
ni
International Index to Periodicals
Encyclopaedias (reservoirs of concise The Encyclopaedia Britannica
U
and convenient sources of information on World Book Encyclopaedia
various subjects; these include biographies
The Encyclopaedia of Science Fiction
and illustrations)
The Atlas of the Prehistoric World
Encyclopaedia of Cryptography and
ity
Security
Encyclopaedia of Public Health
Encyclopaedia of General Science
Apart from these, newspapers, dissertations, abstracts, theses, online libraries, etc.
serve as good sources of review of literature.
m
a) Get an overall view by referring to a generic source like a good textbook. This will
help us grasp the theoretical concepts like the meaning, nature, and importance of
the concept. We will also get a clear idea about the variables that we would use to
formulate the hypothesis of our research problem.
b) On developing a clear insight about the research problem, we have to review the
(c
empirical researches related to our research problem. During this phase, our chief
concern would be to gather specific details of the field of investigation by using good
handbooks, recommended encyclopaedias, and good abstracts for this purpose.
c) Conduct systematic and thorough Research on study materials gathered from online
Notes and offline libraries. Start collecting references from the educational index. Use a
ity
high degree of precision to copy or type such references.
d) Take systematic, consistent, comprehensive, and accurate notes on cards. Record
references with complete bibliographic information. Record references on the front
side of these cards and take notes on the reverse sides.
rs
1.2.7: Errors in Research
Various types of errors in Research are describe bellow.
1. Population Specification
ve
Population specification errors occur when the researcher does not understand
whom they should survey. Because there are morethan one people whomay consume
the product, but only one who purchases it, or they may miss a segment looking to
purchase in the future.
How to avoid this: In the beginning analysis of customers buying the items have
ni
to be done. This is necessary as it will allow us to identify the appropriate custom\mers
and ways to satisfy them.
Example: Suppose that we collected a random sample of 5000 people from the
m
of consisting of parts or people that are similar to each other) of the population being
studied and sampled from and by the size of the sample.
How to avoid this: Since it cannot be avoided completely, then we should have
multiple people reviewing our sample to account for an accurate representation of
our target population. We can also increase the size of our sample so we get more
participants to survey.
(c
3. Selection
When a sample selected by a non-probability method then selection error is
ity
an inherent bias. This can also occur when respondents who are not relevant to the
study participate, or when there’s a bias in the way participants are put into groups.
rs
some degree of resemblance in characteristics to those of the desired population.
How to avoid this: It may be controlled by going extra lengths to get participation.
A usual survey process includes initiating pre-survey contact requesting cooperation,
actual surveying, and post-survey follow-up. If a response is not received, a second
survey request follows, and perhaps interviews using alternate modes such as
ve
telephone or person-to-person.
4. Non-responsive
Non-response error occur when we can get some sample difference between the
original sample and selected sample.It may occur when either the potential respondent
a) Non-contact errors: Happen due to the inability to reach the respondent. This
may be because the respondent is not at home.
U
(b) Refusal errors: Occur when the respondent does not respond to a particular
item or to multiple items of the questionnaire such as monthly household
income, Religion, Sex and Politics are topics that may obtain item refusal.
Example: If we conduct a telephone surveys, then some respondents are
ity
inaccessible because either they are not at home for the initial call or call-backs.
Remaining have moved or are away from home for the period of the survey. Not-at-
home respondents are typically younger with no small children, and have a much
higher proportion of working wives than households with someone at home. People
who have moved or are away for the survey period have a higher geographic mobility
than the average of the population. Most of the surveys can anticipate errors from non-
m
contact of respondents. By using online surveys we can avoid this error through e-mail
distribution, thus eliminating not-at-home respondents.
How to avoid this: When collecting responses, ensure our original respondents are
participating, and use follow-up surveys and alternates modes of reaching them if they
)A
don’t initially respond. We can also use different channels to reach our audience like in
person, web surveys, or SMS.
5. Measurement
The measurement method can have its issues as well. Thes issues include the
difference between the data produced and the data required by the analyst. There are
(c
How to avoid this: Check all measurements more than one times for accuracy and
Notes ensure our observers and measurement takes are well trained and understand the
ity
parameters of the experiment.
While not all of these errors can be completely avoidable, recognizing them to
reduce the complexity. Next time we’re starting a research project, use this blog as a
checklist to ensure you’re doing everything you can to avoid these common mistakes.
rs
Surrogate Information Error
That kind of error arises from either the inability or the unwillingness of the
respondent to provide the needed information, in other words, information is
ve
obtained from substitutes rather than original sample. It is also defined as the non-
correspondence of the information being sought by the researcher and that is required
to solve the problem.
Experimental Error:It can be defined as the non correspondence of the “true impact
of” and the “impact at-tribute to” the independent variable
For example
ni
Conduct a study trying to find out reasons for cancer may conclude that as all
the patients under observation used to eat bread therefore eating bread is the cause
U
of cancer. After that When an experiment is conducted, the researcher attempts
to measure the impact of one or more manipulated independent variable on some
dependent variablewhile controlling the impact of exogenous variable.
ity
m
)A
(c
ity
Unit Objectives:
At the end of this unit the participant will be able to:
rs
1.3.1 Pure &Applied
Applied Research
Applied research is a type of study that aims to solve a specific problem or offer
novel solutions to issues that affect a person, a group, or a society. Because it involves
ve
the practical application of scientific methods to everyday problems, it is often referred
to as a scientific method of inquiry or contractual research.
ni
Because of its direct approach to finding a solution to a problem, applied research
is sometimes considered a non-systematic inquiry. It’s a type of follow-up research that
digs deeper into the findings of pure or basic research in order to validate them and
U
use them to develop innovative solutions.
Pure research or fundamental research are other terms for basic research.
Between the late 19th and early 20th centuries, the concept of basic research arose as
a means of bridging the gaps in science’s societal utility.
(c
data in order to improve one’s understanding, which can then be used to propose
Notes solutions to a problem.
ity
Basic Research Example in Education
How does human retentive memory?
rs
1.3.2 Causal & Correlational Research
Explanatory research, also known as causal research, is used to determine the
scope and nature of cause-and-effect relationships. Causal research can be used to
assess the effects of specific changes on existing norms, processes, and so on.
ve
Causal studies examine a situation or a problem in order to explain the patterns of
relationships between variables. In studies using a causal research design, the most
common primary data collection method is experiments.
●●
●●
●●
ni
To determine the effects of foreign direct investment on Taiwanese economic
growth.
To investigate the impact of rebranding initiatives on customer loyalty.
To determine the nature of the impact of work process re-engineering on
U
employee motivation levels.
Did u know? Conceptual Research is related to certain abstract ideas or theories
that are often applied by philosophers to develop new concepts or to rework on the
existing ones.
ity
Cross-sectional study
The longitudinal and cross-sectional studies are both observational studies. This
)A
means that researchers record data about their subjects without tampering with the
research environment. We would simply measure the cholesterol levels of daily walkers
and non-walkers, as well as any other.
ity
in the same age groups. We might even create gender subgroups. We would not,
however, consider past or future cholesterol levels because they would be outside the
scope. We’d only examine cholesterol levels at a single point in time.
rs
educational level in relation to walking and cholesterol levels, for example, with little or
no extrapolation.
Cross-sectional studies, on the other hand, may not provide conclusive evidence
of cause-and-effect relationships. This is because such studies provide a snapshot of
a single moment in time and do not consider what occurs before or after the snapshot.
ve
As a result, we can’t say for sure whether our daily walkers had low cholesterol levels
before starting their exercise routines or if the daily walking behaviour helped to lower
cholesterol levels that were previously high.
Longitudinal Study
ni
A longitudinal study is observational, just like a cross-sectional one. As a result,
researchers do not interfere with their subjects once more. A longitudinal study, on the
other hand, involves researchers making multiple observations of the same subjects
over a long period of time, sometimes many years.
U
A longitudinal study has the advantage of allowing researchers to detect changes
in the characteristics of the target population at both the group and individual level. The
important thing to remember is that longitudinal studies go beyond a single point in
time. As a result, they can create event sequences.
ity
In general, the design should be driven by the research. However, the progression
of the research can sometimes aid in determining which design is best. Longitudinal
studies take longer to complete than cross-sectional studies.
)A
Experimental
Experimental research is a type of study that employs a scientific approach
to manipulate one or more control variables of the research subject(s) and then
(c
measure the impact of the manipulation on the subject. It is well-known for allowing the
manipulation of control variables.
Even though it can be difficult to execute, this research method is widely used in a
variety of physical and social science fields. They are far more common in information
Notes systems research than in library and information management research within the
ity
information field.
rs
Semi Experimental: The prefix quasi means “similar to.” As a result, quasi-
experimental research is research that resembles experimental research but isn’t
actually experimental. Participants are not randomly assigned to conditions or orders
of conditions, despite the fact that the independent variable is manipulated (Cook &
Campbell, 1979). 1st The directionality problem is eliminated in quasi-experimental
ve
research because the independent variable is manipulated before the dependent
variable is measured. However, because participants are not assigned at random, there
is a chance that other differences exist between conditions. Thus, quasi-experimental
research does not eliminate the problem of confounding variables.
ni
school program which assured the students of higher grades. We divided two similar
groups of children who attend different schools, one of which implements the new
program while the other does not.
U
In field settings where random assignment is difficult or impossible, quasi-
experiments are most likely to be conducted. They’re frequently used to assess the
efficacy of a treatment, such as psychotherapy or an educational intervention. There are
numerous types of quasi-experiments, but we will focus on a few of the most common
here.
ity
Non-Experimental:
Non-experimental research is defined as research in which no control or
independent variable is manipulated. Researchers in non-experimental research
measure variables as they occur naturally, without any further manipulation.
When the researcher doesn’t have a specific research question about a causal
m
●● The subject of the study is a causal relationship, but the independent variable
cannot be changed.
●● The study is broad and exploratory in nature.
●● The study focuses on a variable-to-variable non-causal relationship.
●● Only a limited amount of information about the research topic is available.
(c
Descriptive research
Descriptive research is defined as a research method that describes the
Amity Directorate of Distance & Online Education
Research Methodology 27
ity
The descriptive research method primarily focuses on describing the nature of a
demographic segment, without focusing on “why” a particular phenomenon occurs.
In other words, it “describes” the subject of the research, without covering “why” it
happens.
rs
preference, etc. are examples of descriptive research. Example: AA group of market
researchers aims at identifying the impulse buying trends among various households
pan India. The researchers would focus on collecting data related to “what is the
impulse buying pattern of Indian consumers” and the scope of their research would
be limited to that. The research does not explain the underlying reasons behind such
ve
impulse buying practices or “why” such buying pattern exists. Here, the scope of the
research is just to report the existence of such buying trends and not why do people
resort to impulse buying. This is, hence, an ideal example of descriptive research.
Exploratory Research
ni
Exploratory research is the investigation of a problem that has not previously been
studied or thoroughly investigated. Exploratory research is usually done to gain a better
understanding of the problem at hand, but it rarely yields a conclusive result.
Exploratory research is used by researchers when they want to learn more about
U
an existing phenomenon and gain new insights into it in order to formulate a more
precise problem. It starts with a broad concept, and the research findings are used to
uncover related issues to the research topic.
data or insight. The results of this research, also known as interpretative research or
grounded theory approach, provide answers to questions like what, how, and why.
The exploratory research conducted after product development will be the focus
of our attention. It’s known as the beta testing stage of product development for tech
products.
)A
For example, if a new feature is added to an existing app, product researchers will
want to see how well the feature is received by users. The research is not exploratory if
the feature added to the app is something that already exists.
If Telegram adds a status feature to its app, for example, the app’s beta research
stage is not exploratory. This is because this feature is already available, and they can
easily obtain sufficient information from WhatsApp.
(c
When it comes to a new feature, such as Snapchat filters when they first launched,
the research is instructive. A focus group of beta testers is used to conduct exploratory
research in this case.
Amity Directorate of Distance & Online Education
28 Research Methodology
ity
Unit Objectives:
At the end of this unit, you will learn:
rs
●● Types of Research Models
●● Stages of Research Model
●● Heuristic Research Model
ve
●● Simulation Research Model
●● Data Considerations while analyzing Data for a Research
problem or develop new ideas for research. Qualitative research finds its use mostly
in the humanities and social sciences, in subjects such as anthropology, sociology,
education, health sciences, history, etc. Qualitative research helps to visualize how
people can experience the world. While there are many approaches to qualitative
research, they are less desirable as they are flexible and focus on retaining rich Notes
ity
meaning when interpreting data.
rs
finds a wide use in the natural and social sciences: biology, chemistry, psychology,
economics, sociology, marketing, etc.
ve
These steps are: (1) choosing a topic, (2) defining the problem, (3) reviewing the
literature, (4) formulating a hypothesis, (5) selecting a research method, (6) collecting
data, (7) analysing the results, and (8) sharing the findings.
Other authors may identify more or fewer steps, but the fundamental model
remains the same. Validity and reliability are two important aspects of research. Validity
ni
refers to whether or not the research actually measures what it claims to. The degree
to which research produces consistent or dependable results is referred to as reliability.
Sociologists use six different research methods to conduct their studies: (1)
surveys, (2) participant observation, (3) secondary analysis, (4) documents, (5)
U
unobtrusive measures, and (6) experiments. Resources, access to subjects, the
purpose of the research, and the researcher’s background all play a role in how
sociologists choose their research methods.
as mentioned before, at a rate of 40 to 60% per year. Simply storing this huge amount
of knowledge (data) is not going to be all that useful unleass thera re used for any
particulr research and this can be the reason why organizations are looking at options
like data lakes and big data analysis tools that can help them in handling big data to a
)A
great extent.
●● Nature, scope and object of enquiry: This constitutes the most important
factor affecting the choice of a particular consideration of data. Data is to be
considered on its nature, scope and object of enquiry, when the data is available
for research, checking its sources and utilization in the research it will be judged
Notes and considered.
ity
●● Time Factor: Time is important as there can be huge data which is available but
there will certainy be a deadline for the research hence, firstly the data collection
needs to be done timely and post the collection there has to be enough time for
the considerating and testing.
●● Precision Required: it is another factor whiclethe collection of data is ongoing
rs
and post the collection. While considering the data the right precision is required
so that it follows the same wave length of the reseach program.
ve
HeuristicResearch Model
There are a wide range of qualitative research models available and one of the
lesser-knownmodels is the Heuristic research model. This research model was
developed by Clark Moustakas (an American psychologist and researcher). The name,
Heuristic was derived from the Greek work ‘Heuriskein’ (which means discover, find).
The research model has six phases
●●
●●
Initial engagement
Immersion ni
U
●● Incubation
●● Illumination
●● Explication
●● Creative synthesis
ity
instead of cognitive resources, which affects message intake and processing. The
elaboration likelihood model, or ELM, is very similar to the HSM. Both models were
developed primarily in the early to mid-1980s, and they share many of the same
concepts and ideas.
)A
maintenance, for example. When we consider the effect on downstream units, the effect
of such a delay may be amplified (or absorbed).
‘Entities’ (e.g. machines, materials, people, etc.) and ‘activities’ are two types of
entities in a simulation model (e.g. processing, transporting, etc.). It also includes an Notes
ity
explanation of the logic that governs each activity. A processing activity, for example,
can only begin when a certain quantity of working material, a person to operate the
machine, and an empty conveyor to transport the product are all available. Once an
activity has begun, the time it will take to complete it is calculated, which is frequently
done using a sample from a statistical distribution.
rs
Summary
ve
●● Explaining research process
●● Identifying different approaches of research
●● Elaborating planning a research project
●● Analyzing Application of Research
Exercise
ni
1. The purpose of research is to find solutions through the application of ......................
and ...................... different methods.
U
a) Synthesizing and Analyzing
b) Applying and interpreting
c) Both and b
d) none of the above
ity
a) Market research
b) Product research
c) Demand research
Notes
ity
d) none of the above
5. Planning, organising, staffing, communicating, __________, ________ and
_________are all management functions
a) coordinating
b) motivating
rs
c) controlling
d) all of the above
Answers:
ve
1. a) Synthesizing and Analyzing
2. b) Studying performance appraisal techniques
3. d) Revising monetary policies
4. a) Market research
5. d) all of the above
ni
U
ity
m
)A
(c
Module-2: Sampling
Notes
ity
Key learning outcomes
At the end of this module the participant will be able to:
rs
●● Define Data Collection and its Methods
●● Explain Questionnaire Designing
●● Describe Measurement and Scaling
●● Analyze Sampling
ve
Structure:
Unit-2.3: Sampling
2.3.1 Sampling Plan
2.3.2 Sampling Frame
m
ity
Unit Objectives:
At the end of this unit, you will learn:
rs
●● Tabulating and Validating the Collected Data
ve
Data or information is of two types:
●● Primary Data
●● Secondary Data
Information gathered through original or first-hand research is referred to as
ni
primary data. Surveys and focus group discussions, for example. Secondary data, on
the other hand, is information that has already been gathered by someone else. For
instance, internet research, newspaper articles, and company reports.
Primary Data: It is a term for data collected at source. This type of information
is obtained directly from first hand sources by means of surveys, observations and
experimentation and not subjected to any processing or manipulation and also called
primary data. Primary data means original data that has been collected specially for
the purpose in mind. It means someone collected the data from the original source
)A
first hand. Primary data has not been published yet and is more reliable, authentic and
objective.
Secondary data: It refers to the data collected by someone other than the user
i.e. the data is already available and analysed by someone else. Common sources
of secondary data include various published or unpublished data, books, magazines,
newspaper, trade journals etc.
(c
ity
below-
rs
a. Structured (descriptive) and unstructured (exploratory) observation- When
a observation is characterized by careful definition of units to be observed,
style of observer, conditions of or observation and selection of pertinent data
of observation it is a structured observation. When there characteristics are not
thought of in advance or not present. it is a unstructured observation.
ve
b. Participant, Non-participant and disguised observation- When the observer
observes by making himself more or less, the member of the group he is
observing, it is participant observation but when the observer observes by
detaching himself from the group under observation it is non participant
observation. If the observer observes in such manner that his presence is
ni
unknown to the people he is observing it is disguised observation.
c. Controlled (laboratory) and uncontrolled(exploratory) observation- If the
observation takes place in the natural setting it is a uncontrolled observation but
when observation takes place according to some pre-arranged plans ,involving
U
experimental procedure it is a controlled observation.
Limitations:
●● Expensive methodology.
●● Information provided is limited.
m
Interview Methods:
This method of collecting data involves presentation of oral verbal stimuli
)A
and deeply in terms of oral- verbal responses. It can be achieved by many types of
interviews like
Indirect oral examination- The interviewer has to cross examine other persons who
are suppose to have a knowledge about the problem.
ity
Unstructured interviews- It does not follow a system of pre-determined questions
and is characteirzsed by flexibility of approach to questioning.
rs
Clinical interviews- It is concerned with broad underlying feeling and motives or
individuals life experience which are used as method to collect information under this
method at the interviewer direction.
ve
Non directive interview- The interviewer`s function is to encourage the respendent
to talk about the given topic with a bare minimum of direct
Advantages-
●● More information and in depth can be obtained.
ni
●● Samples can be controlled.
●● There is greater flexibility under this method
●● Personal information can as well be obtained.
Telephonic Interview:
It requires the interviewer to collect information by contacting respondents on
telephone and asking questions or opinions orally.
m
2. Questionnaire
In this method a ouestionnaire is sent (mailed) to the concerned respondents who
are expected to read, understand and reply on their own and return the questionnaire. It
consists of a number of questions printed or typed in a definite order on a form or set of
)A
forms. It is advisable to conduct a ‘pilot study’ which is the rehearsal of the main survey
by experts for testing the questionnaire for weaknesses of the questions and techniques
used.
ity
●● The physical appearances-quality of paper, colour etc must be good to attract
the attention of the respondent
Advantages
●● Free from bias of interviewer.
rs
●● Respondents have adequate time to give answers
●● Respondents are easily and conveniently approachable
●● Large samples can be used to be more reliable.
ve
Limitations
●● Low rate of return of duly filled questionnaire.
●● Control over questions is lost once it is sent.
●● It is inflexible once it is sent.
●● Possiblitty of ambiguous omission of replies.
●● Time taking and slow process
3. Schedules
ni
This method of data collection is similar to questionnaire method with difference
U
that schedule are being filled by the enumerations specially appointed for the purpose.
Enumerations explain the aims and objects of the investigation and may remove any
misunderstanding and help the respondents to record answer. Enumerations should be
well trained to perform their job,he/she should be honest hardworking and patient. This
type of data is helpful in extensive enquiries however it is very expensive.
ity
●● Publications of government.
●● Technical and trade journals.
●● Reports of various businesses, banks etc.
●● Public records.
)A
●● Stastistical or historical
Unpublished data may be found in letters, diaries, unpublished biographies or
work. Before using secondary data it must be checked for the following characteristics-
1. Reliability of data- Who collected the data? From what source? Which method?
Time? Possibility of bias? Accuracy?
(c
2. Suitability of data- The object scope and nature of the original enquiry must be
studies and then carefully scrutinize the data for suitability.
3. Adequency- The data is considered inadequate if the level of accuracy achieved
in data is found inadequate or if they are related to an area which may be either
Notes narrower or wider than the area of the present enquiry.
ity
2.1.3 Types of Data Collection Methods
There are various methods to collect the two sources of data (Primary and
Secondary) as mentioned and explained below:
Primary data is gathered from first-hand experience and has never been used
rs
before. The data gathered through primary data collection methods is highly accurate
and specific to the research’s purpose.
Quantitative and qualitative data collection methods are the two types of primary
data collection methods.
ve
Quantitative Methods:
Smoothing Techniques ni
services for the future using patterns.
Smoothing techniques can be used when the time series lacks significant trends.
U
They get rid of the random variation in historical demand. It aids in the identification
of patterns and demand levels in order to forecast future demand. The simple moving
average method and the weighted moving average method are the two most common
methods for smoothing demand forecasting techniques.
ity
Barometric Method
Researchers use this method, also known as the leading indicators approach, to
predict future trends based on current events. When past events are used to forecast
future events, they are referred to as leading indicators.
Qualitative Methods:
m
Surveys
Surveys are used to gather information about the target audience’s preferences,
opinions, choices, and feedback on their products and services. Most survey software
)A
You can also save time and effort by using a pre-made survey template. By
changing the theme, logo, and other elements, online surveys can be tailored to fit
the brand of the company. They can be distributed via a variety of channels, including
email, website, offline app, QR code, social media, and so on. You can choose the
channel based on the type and source of your audience.Survey software can generate
(c
various reports and run analytics algorithms to uncover hidden insights once the data
has been collected. A survey dashboard can show you statistics such as response rate,
completion rate, demographic filters, export and sharing options, and so on. Integrating
survey builder with third-party apps can help you get the most out of your online data
collection efforts. Notes
ity
Polls
One single or multiple choice questions is asked in a poll. You can use polls when
you need to get a quick pulse on the audience’s feelings. It is easier to get responses
from people because they are short in length.
rs
Online polls, like surveys, can be integrated into a variety of platforms. After
the respondents have responded to the question, they can see how their responses
compare to those of others.
Interviews
ve
The interviewer asks the respondents questions either face-to-face or over the
phone in this method. In face-to-face interviews, the interviewer asks the interviewee
a series of questions in person and takes notes on the answers. If meeting the person
is not possible, the interviewer can conduct a telephonic interview. When there are
only a few respondents, this method of data collection is appropriate. If there are many
Delphi Technique
ni
participants, repeating the same process is too time-consuming and tedious.
Market experts are given the estimates and assumptions of forecasts made
U
by other industry experts in this method. Based on the information provided by other
experts, experts may reconsider and revise their estimates and assumptions. The final
demand forecast is based on the consensus of all experts on demand forecasts.
Focus Groups
ity
A focus group is a small group of people (around 8-10 members) who meet to
discuss the problem’s common areas. Each person expresses his or her viewpoint
on the subject at hand. The discussion among the group members is moderated by a
moderator. The group comes to an agreement at the end of the discussion.
Questionnaire
m
●● Published Sources
(c
●● Unpublished Sources
Published Sources:
International, governmental and local agencies are the ones to publish statistical
ity
International Publications
We have international institutions and bodies like I.M.F, I.B.R.D, I.C.A.F.E and
U.N.O who occasionally publish on occasional reports on statistical and economic
matters.
rs
Reports on different subjects are published by several departments of the Central
and State Governments regularly. They collect all the additional information. Important
publications among these are: The Reserve Bank of India Bulletin, Census of India,
Statistical Abstracts of States, Agricultural Statistics of India, Indian Trade Journal, etc.
ve
Semi-Official Publications
Example: Municipal Corporations, District Boards, Panchayats, etc. that will publish
reports relating to different matters of public concern.
ni
Indian Statistical Institute (I.S.I), Indian Council of Agricultural Research (I.C.A.R),
Indian Agricultural Statistics Research Institute (I.A.S.R.I), etc., publish the findings of
their research programs.
U
5. Publications of various Commercial and Financial Institutions:
6. Reports of various Committees and Commissions appointed by the Government:
Such as the Raj Committee’s Report on Agricultural Taxation, Wanchoo
Committee’s Report on Taxation and Black Money, etc. are also important sources of
ity
secondary data.
Unpublished Sources
There are different examples of these source of data like records maintained by
various government and private offices, the theses of the numerous research scholars
)A
value.” This data can be erroneous in different respects due to biases and prejudiced
mindset of the information collectors along with the sample size being inadequate,
mistakes in definition, mathematical errors and substitution issues. Even without
error, such data still can be unsuitable for enquiry purpose. E According to Prof. Simon
Amity Directorate of Distance & Online Education
Research Methodology 41
ity
and the users also, for the most part, tend to accept a series particularly one issued by
a government agency at its face value without enquiring its reliability”.
rs
This is possible by judging the scope and nature of the present enquiry with the
original one. For example, if we are looking for trend in retail prices while the data
provided is meant for wholesale prices, then it is of no use.
Adequacy of Data
ve
Once it is ensured that the data is suitable for investigation, it should be checked
for the purpose of present analysis. Geographical area for the original enquiry can be
studied in this respect along with the time for which we are getting the data. In the
above example, if we want to study the retail price trend of India, and acquired data will
cover only the retail price trend in the state of UP, then it would not serve the purpose.
Reliability of Data
ni
This issue concerns whether research findings can be applied to a larger group
than those who participated in the study. To put it another way, would similar results
U
have been obtained if a different group of respondents or a different set of data points
had been used? Is the information obtained from these 40 people sufficient to conclude
how the entire sales force feels about company policies, for example, if 40 salespeople
out of a 2,000-person corporate sales force participate in a research study focusing
on company policy? Would the results be the same if the study was repeated with 40
ity
different salespeople?
The main goal of reliability is to ensure that the data collection method produces
consistent results. This can be measured in some types of research by having different
researchers use the same methods to see if the results can be replicated. If the results
are similar, the data collection method is most likely reliable. The scientific research
method includes ensuring that research can be replicated and produces similar results.
m
While editing primary data, the following considerations should be borne in mind:
Meaning of Tabulation
Tabulation is a systematic and logical presentation of numeric data in rows and
ity
and interpretation.
In other words, the method of placing organized data into a tabular form is called
as tabulation. It may be complex, double or simple depending upon the nature of
categorization.
rs
Tabulation of Data Collected
The objectives of tabulation of collected data are as follows:
ve
●● To simplify the data collected
●● Conserves space
●● Ease of comparison
●● Summation of items
●●
●●
ni
Enables easy detection of errors and omissions
Facilities Statistical Computation
Data validation means checking the accuracy and quality of source data before
using, importing or otherwise processing data. Different types of validation can be
ity
For example, an email question will automatically check if the data entered is a
valid email. A phone number question can check whether the phone number has the
right number of digits, based on its country code.
m
Types of Validity
Notes
ity
1. Content validity:
The extent to which the items’ content adequately represents the universe of all
relevant items under investigation. Is it true that samples are representative of the
population/universe?
2. Criterion Validity
rs
The extent to which each criterion can be measured correctly. For instance,
consider a family’s income.
3. Construct Validity -
ve
The construct validity of a scale or test refers to how well it measures the construct.
For example, a doctor might assess the effectiveness of a painkiller. Each day, he
tries to assess the level of pain by asking his patients to rate pain on a 1-10 scale.
Whether its pain or numbness, he’s measuring it.
ni
U
ity
m
)A
(c
Unit-2.3: Sampling
Notes
ity
Unit Objectives:
At the end of this unit, participants will be able to learn:
●● Describe sampling
●● Analyze sampling plan and sampling frame
rs
●● List steps involved in sampling process
●● Identify different sample selection methods
●● Describe probability and non-probability sampling techniques
ve
●● Identify sampling and non-sampling errors
It is a subset of population
Bryman & Bell– ni
“Segment of population that is selected for investigation.”
U
“Some of the elements of population”
Cooper & Schindler–
ity
m
sample can be applied to the entire population, the sample should be representative of
the population.
We obtain a sample of the population for a variety of reasons, including the fact
that it is rarely practical and almost never cost-effective.
ity
●● Inaccessibility of some populations: Access to some populations is so difficult
that only a sample can be used. Prisoners, people with severe mental illness, and
disaster survivors are just a few examples. And so on. The inaccessibility could be
due to a lack of funds, time, or simply access.
●● Destructiveness of observation: Sometimes just observing a product’s desired
characteristic destroys it for its intended use. Quality control is a good example of
rs
this. For example, a fuse must be destroyed to determine its quality and whether it
is defective.
●● As a result, if you tested all of the fuses, they’d all blow up.
●● Accuracy and sampling: A sample of the study population may be more accurate
ve
than the entire population. A population that has been incorrectly identified can
provide less reliable data than a sample that has been carefully selected.
ni
A sampling plan is a term widely used in research studies that provide an outline
on the basis of which research is conducted. It tells which category is to be surveyed,
what should be the sample size and how the respondents should be chosen out of the
population. Sampling plan is the base from which the research starts and includes the
U
following major decisions:
as possible.
procedure i.e., which method can be used such that every object in the population
has an equal chance of being selected. Generally, the researchers use the probability
sampling to determine the objects to be chosen as these represents the sample more
accurately.
A sampling frame is a list or database from which a sample can be used. In market
research terms, a sampling frame is a database of potential respondents that can be
drawn from, to invite to take part in a given research project.
ity
The following are the series of steps that are involved in the sampling process:
rs
●● Execute the sampling process
ve
●● Probability Sampling
●● Non-Probability Sampling
●●
●●
●●
Simple random sampling
Systematic Sampling
Stratified Sampling ni
U
●● Cluster Sampling
●● Multistage Sampling
chance of selection
Sample size
m
= 25/591
= 0.042 or 4.2%
)A
(c
Systematic Sampling
Notes
ity
A probability sample drawn by applying a calculated skip interval to a sample
frame.
rs
ve
Population sample frame (N)
Skip interval (k) = ------------------------------------------------
Sample Size (n)
= 591/25
= 23.64 = 23
a gap of 23 numbers.
k = 23
ni
Select any number randomly between 1 – 23, and then select rest 24 numbers with
U
Select any number randomly between 1 – 23, and then select rest 24 numbers with
a gap of 23 numbers.
ity
m
Stratified Sampling
A probability sampling technique in which the population is divided into different
)A
sub-homogeneous groups or strata and samples are randomly selected from such sub-
groups or strata.
Sub groups-
(c
MPM = 59 DBM = 92
DIEM = 37 MCA = 59
Notes
ity
MCM = 30
rs
ve
Cluster Sampling
ni
A probability sampling technique in which the population is divided into several
small sub groups and some groups selected randomly for study.
DIEM = 37 MCA = 59
MCM = 30
ity
Notes
ity
rs
ve
2.3.5 Non - Probability Sampling
Convenience Sampling
A non-probability sampling technique where researcher use any readily available
individuals as participants
ity
Least reliable
purpose of the study. Purposive sampling may involve studying the entire population of
some limited group or a subset of a population). As with other non-probability sampling
methods, purposive sampling does not produce a sample that is representative
of a larger population, but it can be exactly what is needed in some cases - study of
organization, community, or some other clearly defined and relatively limited group
(c
Quota Sampling
A purposive sampling where relevant characteristics are used to stratify the
samples.
ity
MBA = 272 46% 12
MMM = 42 7% 2
MPM = 59 10% 2
DBM = 92 16% 4
DIEM = 37 6% 2
rs
MCA = 59 10% 2
MCM = 30 5% 1
Total = 591 100% 25
ve
Snowball Sampling
A non- probability sampling in which subsequent participants are referred by
current sample elements.
ni
These initial individuals refer others who are similar to them and so on. Like a
snowball gathers subjects as its rolls along.
Non-Sample Errors
Non-sample errors can be classified into:
●● Non-response Error
●● Response Error
)A
Non-Response Error
A non-response error occurs when units selected as part of the sampling procedure
do not respond in whole or in part.
Response Error
(c
A response or data error is any systematic bias that occurs during data collection,
analysis or interpretation.
ity
●● Respondent Error (example: lying, forgetting, etc.,)
●● Interviewer Errors
●● Recording Errors
●● Poorly designed questionnaires.
●● Measurement errors.
rs
ve
ni
U
ity
m
)A
(c
Notes
Module-3: Experiments
ity
Structure:
rs
3.1.2: Strategies
3.1.3: Pre-Experimental Research Design
3.1.4: Quasi-Experimental Research Design
3.1.5: True Experimental Research Design
ve
3.1.6: Factorial Experimental Design
3.1.7: Designing Engineering Experiments
3.1.8: Causal Attributions
3.1.9: Statistical Control
3.2.6: Orthogonality
3.2.7: Guidelines for designing experiments
3.2.8: False Positives
m
)A
(c
ity
Unit Objectives:
Objectives: From this chapter, we can learn about
rs
●● What is the basic principle for an experiment.
●● How an engineering experiment will be conducted.
●● The Attributes
ve
Experimental research is conducted with a scientific approach while keeping a
set of variables constant and measuring the other set of variables as the subject of an
experiment. A highly effective quantitative research technique, experimental research is
particularly useful when the researcher does not have enough information to arrive at
meaningful conclusions and have to carry out experiments to prove facts. Experimental
ni
research, thus, helps a researcher make important decisions. It is considered
successful only when the researcher assures that a change in the dependent variable is
wholly due to a change in the independent variable
3.1.1 Objectives
U
The results of experimental research should help a researcher establish a causal
relationship between the dependent and independent variables in a given phenomenon.
Effects are observed, during an experiment, due to the underlying cause. Experimental
research is conducted when:
ity
●● There exists a constant and consistent relationship between the cause and the
effect
●● Time is a crucial factor for establishing a relationship between cause and effect
●● The factors that drive the success of experimental research are:
●● Control group (a group of research participants that are familiar with the
m
3.1.2 Strategies
It is the purpose of experimental design to obtain a maximum of information with
minimal time. Experiments mainly planned by knowledgeable people who can:
(c
ity
The preparation of good quality glass bottles could be influenced by the chemical
glass composition and the resulting transition temperature, the raw material suppliers,
the kind and amount of cullet used, the process parameters, the experience, salary, and
motivation of the personnel involved in the production, the day of production within a
weeks work (e.g., weekend or workday), and many other influences.
rs
c. Select thesignificant influences andprimary causes without oversimplifying or
overcomplicating the problem.
An experienced engineer may estimate that, for example, that the day of
production does not influence the quality of glass bottles significantly, despite the
ve
different working atmosphere on weekends. On the other hand, the engineer may
recognize the glass bottle annealing schedule and the transition temperature among
the most important influences on the quality that must not be neglected. The transition
temperature, however, may be deemed not to be a primary cause because it largely
depends on the chemical glass composition. Therefore, the transition temperature as a
possible variable could be replaced by the glass composition.
ni
d. Express the influences in numerical variables.
Most influences can be easily expressed in meaningful numbers, while qualities
such as the experience of personnel should be simplified and categorized in levels like
U
1 (little experience) to 3 (much experience). Similarly, the type of raw material supplier
may be categorized.
f. predict whether the expected effects of the variables are linear, independently
of the values of the variables.
It could occur that a cooling rate change of 1oC/min is expected to influence the
glass bottle strength by 5-20%, but at the same time, a cooling rate change of 2oC/
)A
min is expected to influence the glass bottle strength for 20-50%, i.e., the effect of the
cooling rate is probably not linear.
the glass not as feldspar but as hydroxide. The influence of cullet and alumina may be
traced back to the glass composition as a primary cause.
h. Set practical upper and lower bounds on the variables, the desired properties,
and also on the variable combinations if variable interactions exist. Notes
ity
Variables seldom can be modified completely arbitrarily. For example, the cooling
rate during glass bottle annealing can not exceed the limits of safe operation of the
machine, and the chemical glass composition is subject to financial and environmental
restrictions. It may also not be reasonable to investigate the strength of glass bottles
below a certain limit where even careful handling leads to breakage. If variable
rs
interaction exists between the glass bottle cooling rate and the percentage of cullet
used for glass melting (i.e., the glass composition), it is likely that the full range of all
possible cooling rate - glass composition combinations should not be investigated
because undesirably low strength data may result or technical or financial limitations
exist.
ve
i. apply experimental design matrix types accordingly. The more information that
can be gathered beforehand, the less intensive work will be required for the
final experimental plan.
The mean of many measurements is more reliable than one single or a few
measurements. In other words: the error of the mean decreases with an increasing
ni
number of measurements as depicted. Similarly, the effect of a variable is best
determined by comparing means against means. When analyzing experimental
findings, two types of errors are possible: an effect could be declared although there
is none in fact (error type 1), or a real effect could go undetected (error type 2). To
U
defend properly against both types of errors, one must run n = 16 • s2/D2 experiments
at each variable level (= variable value) and use a statistical significance test to
compare the mean of the results. For example, if it would be desirable to find out if
a change in the glass composition decreases the annealing point for 10oC (D = 10)
and the measurement precision is 5oC (s = 5), one would need to run 16 • 52 / 102
ity
= 4 experiments with each of the two glass compositions and compare the mean of
the results for statistical significance. If several variables need to be considered, the
number of experiments required for a thorough investigation could thus become very
large. A proper experimental design can reduce the effort significantly.
ity
●● Used in cases where random assignment is either irrelevant or not required
Ex:We choose two similar groups of children who attend different schools, one of
which implements the new program while the other does not
rs
●● Most accurate form of experimental research design
●● Depends on statistical analysis to prove or disprove a hypothesis
●● Only type of experimental research design that can establish a cause-effect
relationship within a group/a few groups
ve
●● Commonly used in Physical Sciences
ni
the interactingeffect of the several variables.
Combinations Sales
m
R1P1 S1
R1P2 S2
R2P1 S3
)A
R2P2 S4
R3P1 S5
R3P2 S6
S1 TO S6 represents the sales resulting out of each variable. The data gathered
will provide
(c
2. Is the display at the entrance more effective than the display at the usual
location? Notes
ity
Also, theresearch will tell us about the interaction effect of the two variables.
rs
3. No display and the regular price applicable
4. Display at the entrance with regular price applicable
ve
In engineering,experiments are a natural part ofthe decision-making process.
Just like for an example, a civil engineer is investigating the effects of different
curing methods on the mean compressive strength of concrete. The experiment
consistsof several test specimens of concrete by using each and every proposed
ni
method of curing and then testing the compressive strength of each specimen. The
data collected from this experimentis used to determinewhich curing method is perfect
for providing maximum mean compressive strength.
If there are only two curing methods of interest, this experiment could be
U
designed and analyzed using the statistical hypothesis methods for two samples. The
experimenter has a single factor of interest—curing methods—and there are onlytwo
levels of the factor. If the experimenter is interested in determining which curing
methodproduces the maximum compressive strength, the number of specimens to test
can be determinedfrom the operating characteristic curves, and the t-test can beused to
ity
Many single-factor experiments require that more than two levels of the factor be
considered.
For example, the civil engineer may want to investigate five different curing
methods.
m
●● Improved process.
●● Reduced variability in the process and conformance to nominal or
targetrequirements
(c
These methods are also useful in engineering design activities to develop new
Notes products and improve existing ones. Some typical applications of statisticallydesigned
ity
experiments in engineering design include
rs
●● Determination of key product design parameters that affect product
performance
The use of experimental design in the engineering design process can result in
products that are easier to manufacture, products that have better field performance
ve
and reliability than their competitors, and products that can be designed, developed,
and produced in less time.
ni
Subsequent experiments are applied on these critical variables to refine
the collected information and determine which adjustmentsare required to
improve the process.
U
●● Finally, the experimenter optimized which levels of the critical variables result
inthe best process performance.
●● Every experiment involves a sequence of activities like as:
1. Conjecture—the original hypothesis to motivate the experiment.
2. Experiment—the test performed to investigate the conjecture.
ity
described the various ways of causal attributionexamination with the help of research
design, respective data collection and strategies analysis. The OECD-DAC definition of
impact makes it clear that an impact evaluation must establish what has been the cause
of observed changes: “Positive and negative, primary and secondary long-term effects
produced by a development intervention, directly or indirectly, intended or unintended.” Notes
ity
1. Causal attribution is defined by OECD-DAC as: “Ascription of a causal link between
observed (or expected to be observed) changes and a specific intervention.”
2. This definition does not require that changes are produced solely or wholly by the
programme or policy under investigation
3. In other words, it takes into consideration that other causes may also have been
rs
involved, for example, other programmes/policies in the area of interest or certain
contextual factors (often referred to as ‘external factors’). Evaluations produce
stronger and more useful findings if they not only investigate the links between
activities and impacts but also investigate links along the causal chain between
ve
activities, outputs, intermediate outcomes and impacts.
Sole causal attribution: Here the programme or policy is both necessary and
sufficient to produce the impacts, independent or relatively independent of contextual
factors or other interventions.
ni
U
In practical view, science programmes or policies are not sufficient to produce the
intended impacts alone; that’s why they’re used various alternative ways to achieve the
intended impacts. So it’s called a rarely useful model of cause and effect for impact
evaluation.
ity
This is a very common situation, where a programme will only work effectively if
favourable conditions are present and/or unfavourable conditions are removed. For
example, if a programme is found only to work in a place where there is transparency
)A
Notes
ity
rs
Alternative (or multiple) causal paths: Here, the programme or policy is just one
of several possible ways to achieve the impacts. A particular programme or policy can
produce the impacts, but they can also come about through other interventions and/or
external factors.
ve
This can happen when participants are able to access services through an
alternative provider or when different programmes by different providers are intended to
produce the same impact. For example, a number of programmes might aim to reduce
infant mortality – some by improving nutrition, some by reducing waterborne diseases
and some through immunization.
ni
U
ity
logical,evidence-based process.
●● Using a combination of these strategies can usually help to increase the
strength of the conclusions.
It is important that all impact evaluations systematically undertake causal
attribution. But there must also be recognition of what can be realistically achieved by
a single evaluation – especially when there is limited time for iterative data collection
(c
and analysis. A single impact evaluation can add to the evidence base, but even where
there is a reasonable degree of confidence that the programme or policy caused the
observed impacts, this does not mean that the evaluation has examined all aspects of
the intervention or how it will work in other settings or at other times. A synthesis of
findings across impact evaluations thus provides stronger evidence than can a single Notes
ity
evaluation about thegeneralizability of the effects observed.
Main points
Causal attribution investigates the causal links between a programme or other
intervention and observed changes.
rs
L attribution is an essential element of impact evaluation.
There are several strategies for examining causal attribution, all of which benefit
from being based on a sound theory of change.
The ‘best fit’ strategy for causal attribution depends on the evaluation context as
ve
well as what is being evaluated.
ni
if those responsible for making decisions act on the information the chart reveals. In
other words, control charts are tools that can be used, misused, or not used at all.
Their effectiveness is directly related to the understanding of and proper application by
the people. The control chart is a graphical representation of this monitoring process.
U
Control charts are all basically like the conceptual chart with the vertical axis; a scale
is established consistent with the data being collected. The horizontal axis reflects the
data sequence or time interval. As data is obtained, it can be plotted relative to these
two axes.
ity
m
)A
The two major divisions in control charts result from the fact that there are two
types of data - variable data and attribute data.
Variables Charts
_
Amity Directorate of Distance & Online Education
62 Research Methodology
ity
_
X - S Average and Standard Deviation
Attributes Charts
p Fraction Nonconforming
rs
np Number of nonconforming Units
c Number of Nonconformities
ve
Requirements for Using Control Charts
Before applying control chart techniques to any process or operation, certain
ni
essential requirements must be satisfied. First, the people in decision-making positions
should understand and support efforts to continually improve processes and quality
through the use of statistical methods such as control charts. When this has been
achieved, the following criteria must be satisfied for a business to really experience
process improvement using control charts
U
●● Understand the process being studied. Understanding is necessary in order
to make decisions regarding chart selection, sample size, sampling frequency,
characteristics to be monitored and actions that can be taken to reduce variability.
●● Determine what characteristics are to be monitored. Usually, these would be the
ity
ones most critical for the satisfactory functioning of the process or product or
perhaps the ones with the tightest specifications. They could be the characteristics
most often found to be nonconforming or the ones most expensive to repair. A
technique such as Pareto analysis is often used to select a single character to be
charted.
●● Define and understand the measurement system. The data on the chart is going
m
to be no better than the measuring processor device. The required accuracy must
be established so that proper measuring equipment can be selected. Measuring
equipment should be reviewed for accuracy and repeatability and calibrated
regularly.
)A
●● Select a chart type to best accomplish the purpose. Often this is dictated by the
cost of measurement and the loss incurred if important changes go undetected
when and if they should occur. Also, the type of information needed may dictate
which chart to use.
●● In general, use an attribute chart if
●● Measurements are not possible, as in visual inspection.
(c
ity
●● Plans are to use 100% inspection.
Use a variables chart if
rs
common sample size is five. The subgroup of a sample mustbe selected to allow a
minimum opportunity for variation within the group. However, the larger the sample size,
the more normal the distribution of subgroup averages even if the population is non-
normal. The economy is a big factor to determine the sample sizes.
ve
Whereas the attribute charts work with a much larger sample size (30 to several
hundred or more). Since the attribute data may be available from valid preexisting
reports or studies,thenthe time and expense have already been incurred.
ni
In manufacturing plants, when unfortunately arise some type of process trouble,
then control charts can be invaluable trouble-shooting aids for the engineer as well as
for the operators. Control charts provide information mainly in four primary areas likes:
●● Process control charts are tools that help us analyze data, mainly distinguish
between variability as a result of common causes or specialcauses.
●● If only common causes arepresent, then Processes are in control.
●● If special causes arepresent, then Processes are out of control.
m
●● The two general categories of control charts are Variables Control Charts and
Attribute Control Charts.
●● Tables ofc onstants have been developed for use with control charts to utilize the
)A
●● Enter the heading information such as the name of the part, the part number,
Notes department, machine number, plant identification, an operation being performed,
ity
etc.,into the chart.
●● Record the information of the sample, likes the date, the sample was taken, the
time, the initials of the checker, and the shift on which the samples were taken.
●● Entered the measured data in particular places.
●● Calculate the statistic average, the median, the range, the standard deviation, or
rs
otherstatistics being monitored.
●● Establish the scales for the graphs becausescales should be used to do the
plotting. As a general rule, the range of values for the central tendency measure
should be either the product specification or twice the difference in the highest and
ve
lowest subgroup statistics.
●● Plot thedata.
●● Calculate and plot the central lines for both portions of the chart - central tendency
and variability. These should be drawn in as solid lines andlabelled.
ni
●● Calculate and plot the control limits. These are calculated
using the specific formulas for the statistics being monitored.
Thecontrollimitsshouldbedrawninasdashedlineson the chart andlabelled.
●● Interpret the chart.
U
Control Limit Calculations
As you remember from the section “Introduction to Control Charts”, the upper and
lower control limits are, in reality, simply the ±3σ limits for the sample statistic. Let’s look
at the formulas which_
ity
Summary
ity
●● Strategies applied to design an experiment.
●● Casual attribution applied in experiments.
●● Statistical control on experiments.
Questions:
rs
●● What is the factorial design?
●● Discus Quasi experiment with example
●● Discus Pre experiment with example
●● Discus True experiment with example
ve
●● Difference between the above three experiments.
ni
U
ity
m
)A
(c
ity
Unit Objectives:
●● Discus about the Validity function.
●● Discus about Reliability function.
●● Discus about the Replication function.
rs
●● Designing Engineering Experiments
3.2.1 Validity
ve
Validity refers to how accurately a method measures what it is intended to
measure. If research has high validity, that means it produces results that correspond
to real properties, characteristics, and variations in the physical or social world.High
reliability indicates that measurement is valid, but if a method is not reliable, thenit is
probablynot valid. For example,if the thermometer shows different temperatures each
time, even though you have carefully controlled conditions to ensure the sample’s
ni
temperature stays the same, the thermometer is probably malfunctioning, and therefore
However, reliability on its own is not enough to ensure validity. Even if a test is
reliable, it may not accurately reflect the real situation.
construct validity.
Content The measurement covers A test to measure a class of students’ level
all aspects of the concept of Bengali contains reading, writing and
being measured. speaking components, but no listening
)A
3.2.2 Reliability
Notes
ity
Reliability is a measure of the stability or consistency of every test. We can also
think of it as the ability for a test or research findings to be repeatable. For example,
a medical thermometer is a reliable tool that would measure the correct temperature
each time it is used. In the same way, a reliable math test will accurately measure
mathematical knowledge for every student who takes it, and reliable research findings
can be replicated over and over. There are basically two types of reliability Internal and
rs
External reliability.
External reliability means that our test or measure can be generalized beyond
ve
what we’re using it for.
For example, a claim that individual tutoring improves test scores should apply to
more than one subject (e.g. to English as well as math). A test for depression should be
able to detect depression in different age groups, for people in different socio-economic
statuses, or introverts.
Coefficient Reliability
ni
A reliability coefficient is a measure of how well a test measures achievement. It is
the proportion of variance in observed scoresattributable to true scores (the theoretical
U
“real” score that a person would get if a perfect test existed).
The several different coefficients are calculating through the several methods of
test; these are test-retest, parallel forms and alternate-form:
●● A simple correlation between two scores from the same person is one of the
simplest ways to estimate a reliability coefficient. If the scores are taken at
different times, then this is one way to estimate test-retest reliability; Different
forms of the test given on the same day can estimate parallel forms reliability.
●● Pearson’s correlation can be used to estimate the theoretical reliability
coefficient between parallel tests.
m
●● For high stakes tests (e.g. college admissions), > 0.85. Some authors suggest
this figure should be above .90.
●● For low stakes tests (e.g. classroom assessment), > 0.70. Some authors
suggest this figure should be above 0.80
(c
3.2.3 Replication
Replication
ity
For example, if we want to compare the height of two plant varieties by only taking
one plant height measurement and observing a difference of 20 cm. On the other hand,
if multiple plants of each variety are measured, and the height differences always turn
out somewhere around 10 cm, the observed difference is less likely due to chance. The
difference is strong relative to the variability between the measurements.
rs
ve
ni
U
A comparison between two groups
3.2.4 Randomization
After careful identification of blocks, other factors may still influence the
experimental goal, like mouse age and sex differences and various genetic
backgrounds. Try to balance these factors, and we can use randomization techniques.
Randomization reduces confounding effects (distortion of the association between
m
3.2.5 Blocking
Notes
ity
We want to do experiments within a homogeneous group of experimental units.
Those particular homogeneous groups called blocks try to reduce the variability
between the units and increase the meaning of differences between conditions.
rs
estimated and eliminated. If all control conditions are measured on one day and all
treatment conditions on another day, then it is not possible to disentangle the day effect
from the treatment effect and, in the worst case, the data become inconclusive. As a
general rule, at least some “common conditions” are essential to assess potential block
effects.
ve
As an example, assume there are six treatment conditions we want to apply to
mice (the experimental units), but we can fit only five mice per cage (i.e. block). In this
case, not all treatments can be applied simultaneously in each cage/block. We can
apply four identical treatments to each of the cages and only alternate the fifth condition
each time.Now, the “cage effect” can be estimated by computing the mean of the
ni
differences between the four treatments that are identical, as given by the formula. A
priori ,the conditions E and F are not directly comparable since they were measured on
mice from two different cages. However, the replicated treatments allow computation
of a “cage effect” that corresponds to the average difference between the identical
U
conditions measured in the two cages. Then, the difference between E and F can be
computed as E − F − “cage effect”.
ity
m
)A
Illustration of batches and how to correct them. All but two treatments have been
applied to mice in two different cages (= batches). The batch/cage effect can now be
computed based on the treatments that are shared between the cages.
3.2.6 Orthogonality
(c
Notes
ity
rs
The design on the left is balanced because it has even levels.
ve
continuous variable).
For example, let’s say you had four cells in an ANOVA: three cells have ten subjects,
and the fourth cell has nine subjects. Technically this is a non-balanced (and therefore
non-orthogonal) design. However, the missing subject in one cell will have very little
impact on results. In other words, you can treat this semi-unbalanced design as
orthogonal.
The term orthogonality comes from matrix algebra. Here two vectors are
orthogonal if the sum of the cross-element products is zero. For example, the cross
products of these two vectors are zero:
)A
Orthogonality is present in a model if any factor’s effects sum to zero across the effects
of any other factors in the table.
ity
●● Recognition of Problem and statement of the problem (Pre-experimental planning)
●● Factor screening; Optimization; Confirmation; Discovery; Robustness
●● Selection of the response variable(Pre-experimental planning) Choice of factors,
levels, and ranges(Pre-experimental planning) Controllable; Uncontrollable; noise
factors;
rs
●● cause-and-effect diagram
●● Choice of experimental design Performing the Statistical experiment analysis of
the graphical data methods; empirical model
●● Conclusions and recommendations
ve
ni
U
ity
A false positive is that where we get a positive result for a test when we should
have received negative results. For that, sometimes it’s called a “false alarm” or “false
positive error.” It’s usually used in the medical field, but it can also apply to another area
like software testing.
●● The covid-19 test is positive, but in fact, the patient is not a covid-19 patient.
●● A cancer screening test comes back positive, but the patient doesn’t have the
disease.
●● A prenatal test comes back positive for Down’s syndrome when the
Notes patientfactor does not have the disorder.
ity
●● Virus software on our computer incorrectly identifies a harmless program as a
malicious one.
False positives can be worrisome, especially when it comes to medical tests.
Researchers are consistently trying to identify reasons for false positives in order to
make tests more sensitive.
rs
A related concept is a false negative, where you receive a negative result when you
should have received a positive one. For example, a pregnancy test may come back
negative even though the patient is pregnant.
ve
If a test for a disease is 99% accurate and we receive a positive result, what are
the odds that the patient actually have the disease?
If we said 99%, we might be surprised to learn we’re wrong. If the disease is very
common, our odds might approach 99%. But the rarer the disease, the less accurate
the test and the lower the odds that the patient actually has the disease. The difference
ni
can be quite dramatic. For example, if we test positive for a rare disease (one that
affects, say, 1 in 1,000 people), our odds might be less than percent of actually having
the disease! The reason involves conditional probability.
U
Summary:
●● Discuss validity, replication, reliability
●● What is blocking
●● Guidelines for designing experiments
ity
●● False Positives
ity
d. a & b
3. Recognition of Problem and statement of the problem is called
a. Pre experimental design
b. Post experimental deign
c. Quasi experimental design.
rs
d. None
4. If research has high validity, that means
1. it produces results that correspond to real properties
ve
2. it produces results that correspond to real characteristics
3. it produces results that correspond to all variations.
4. All the above
5. Attribute charts work with
a. larger sample size
b.small sample size
c. random sample size
ni
U
d. none
6. applying control chart techniques to any process or operation must be satisfied
a. the people in decision-making positions
b. Understand the process being studied
ity
8. Here, the programme or policy is both necessary and sufficient to produce the
impacts it is
a. Alternative (or multiple) causal paths
b. Joint causal attribution
c. Sole causal attribution
(c
d. None
ity
A . conjecture, experiments, analysis, conclusion.
b. experiments, conjecture, analysis, conclusion.
c. analysis, conjecture, experiments, conclusion
d. none
10. With the help of factorial design, the researcher can test...........
rs
a. two variables at a time
b. more than two variables at a time
c. Two or more variables at a time
ve
d. one variable at a time
same as it.
ni
2. Quasi-Experimental Research DesignResembles true experimental research as the
Ans:
)A
1. A,
2. D
3. a
4. d
5. a
(c
6. d
7. a
8. c
Amity Directorate of Distance & Online Education
Research Methodology 75
9. a
Notes
ity
10. c
True / False
1. T
2. F
rs
3. T
4. T
5. F
6. T
ve
7. F
8. T
9. T
10. T
ni
U
ity
m
)A
(c
ity
Structure:
rs
4.1.2: Introduction to Hypothesis
ve
4.1.5: Type I & Type II Error(with example)
ity
Unit Objectives:
Learn about single factor experiments
●● What is hypothesis
●● Types and Error of hypothesis
rs
●● Degree of freedom.
ve
response (dependent variable). Then we carried out an experiment where the levels of
the factor are varied. Such kind of experiments is known as a single-factor experiment.
Thereare various types of designs available to carry out such experiment. The most
popular ones are completely randomized design, randomized block design, Latin
square design, and balanced incomplete block design.
ni
The CRD is the simplest design. Suppose there are v treatments to be compared.
All experimental units are considered the same, and no division or grouping among
U
them exist.
●● In CRD, the v treatments are allocated randomly to the whole set of experimental
units, without
●● making any effort to group the experimental units in any way for more
ity
- the treatment one is replicated, say three times and is given to 3 experimental
)A
units,
- the treatment two is replicated, say five times and is given to 5 experimental
units,
- treatmentthree is replicated, say six times and is given to 6 experimental units
and
- finally, treatment four is replicated [20-(6+5+3)=] six times and is given to the
(c
ity
CRD is often inefficient.
CRD is more useful when the experiments are conducted inside the lab.
CRD is well suited for the small number of treatments and for the
homogeneous experimental material.
rs
Divide the entire experimental material or area into a number of experimental
units, say n
Fix the number of replications for different treatments in advance (forgiven total
ve
number of available experimental units).
No local control measure is provided as such except that the error variance can
be reduced by choosing a homogeneous set of experimental un
Procedure Let the v treatments are numbered from 1,2,...,v and I n be the number
of replications required for i the treatment such that
Σ ni=n.
n=i
ni
U
●● Select n1 units out of n units randomly and apply treatment 1 to these n1
●● Select n2 units out of ( n- n1) units randomly and apply treatment 2 to these n2
units
●● Continue with this procedure until all the treatments have been utilized.
ity
v
yij: Independently distributed following N (ai μ,σ2 ) withΣniai.
)A
n=i
µ: overall mean
αi : i th treatment effect
ity
Treatment observations totals averages
1 y11y12 · · · y 1 ny 1 y¯ 1.
2 y21y22 · · · y 2 ny 2. y¯ 2
..........···
rs
.........
Analysis of Variance
ve
●● Statistical Model (Factor Effects Model): yij = µ + τi + ǫij i = 1, 2 . . . , a
j = 1, 2, . . . ,ni
µ - grand mean;
iid
ˆμ = y.
ity
●● E(MSE) =σ2
Example
Twelve lambs are randomly assigned to three different diets. The weight gain (in
two weeks) is recorded. Is there a difference between the diets?
rs
Diet 1 8 16 9
Diet 2 9 16 21 11 18
Diet 3 15 10 17 6
ve
●● N = 12, ∑∑yij = 156 and y/= 156/12 = 13.
●● n1 = 3, y1. = 33, y/1. = 11; n2 = 5, y2. = 75, y/2. = 15; n3 = 4, y3. = 48 andy/3. = 12.
●●
ᴧ
τ1 =
y/1– y/ = 11 − 13 = −2; Similarly, ˆτ2 = 15 − 13 = 2 and
●● ˆτ3 = 12 − 13 = −1.
●● SST = ∑i∑j (yij– y/)2 = 246.
●●
●● ni
SSTreatment = 3 * (−2)2 + 5 * (2)2 + 4 * (−1)2 = 36.
SSE = 246 − 36 = 210;
U
MSE = ˆσ2 = 210/(12 − 3) = 23.33
●● F0 = (36/2)/(210/9) = 0.77;
ity
What is a Hypothesis :
A hypothesis is a statement about the research problem and may or may not be
true. It is a tentative and testable relationship between at least two variables that require
research methodology for testing its authenticity. You are required to test the hypothesis
)A
and ascertain the truth in the way of research. Theory and hypothesis are completely
different. Unlike a hypothesis, a theory is a tested, well-substantiated, complete
explanation for a set of proven factors.
The term “hypo” means “subject to verification”, and “thesis” means “statement
about the solution of a problem”. Thus, “hypothesis” is “ a tentative statement, which is
subject to verification and states the solution of a problem”. Hypothesis offers a solution
(c
Illustration:
Notes
ity
A renowned asset management company is in the process of launching a new
Mutual Fund. The fund is aimed at catering to the wealth-building needs of the Indian
middle-class population. The company wants to conduct detailed market research on
the popular choices of investment options in India. It recruits a research team, which
comes up with the research problem:
rs
The team constructs a trio of hypotheses, namely
H0: The Indian middle-class population does not invest in Mutual Funds
H1: At least 20% of the Indian middle-class population invests in Mutual Funds
ve
H2: Less than 20% of the Indian middle-class population invests in Mutual Funds
Note that all three hypotheses have been phrased in a manner so that anyone is
correct. However, note all three cannot be correct or incorrect simultaneously. Also,
note that in order to make an effective decision, you need to select the first act if H1
is true and the second act if H2 is true. The decision of launching or not launching the
ni
Mutual Fund depends on whether H1 or H2 is true. The hypothesis H0 suggests that
the Indian middle-class population is completely indifferent towards Mutual Funds
Thus, the asset management company chooses to launch the mutual fund if H1
is true, i.e. if it is established from the market research that at least 20% of the Indian
ity
Formulation of Hypothesis
●● Simple Hypothesis
●● Complex Hypothesis
●● Empirical Hypothesis
●● Null Hypothesis
●● Alternative Hypothesis
●● Logical Hypothesis
(c
●● Statistical Hypothesis
Simple Hypothesis
Notes
ity
In a simple hypothesis,there exists a relationship between two variables; one
is called an independent variable or cause, and the other is a dependent variable or
effect. For example
rs
Complex Hypothesis
In a complex hypothesis,there exists a relationship among more variables (more
than two dependent and independent). For example
ve
1. Smoking and other drugs lead to cancer, chest infections etc.
2. The high rate of unemployment,poverty, illiteracy leads to crimes like dacoit, Robbery,
Rape, prostitution & killing etc.
ni
Working hypothesis is mainly applied to a particular field. At the time of formulation,
it is an assumption, but when it is readyfor a test to become an empirical or working
hypothesis.
Null Hypothesis
U
It iscontrary to the positive statement of the empirical hypothesis, which has no
relationship between dependent and independent variable. It is denoted by ‘h0”.
Alternative Hypothesis
ity
Generally,many hypotheses are selected but select one from them which is more
workable and most efficient. That hypothesis is introduced later on due to changes in
the old formulated hypothesis. It is denoted by “h1”.
Logical Hypothesis
Here hypothesis is verified logically. J.S. Mill has given four cannons of this
m
Statistical Hypothesis
A hypothesis may be logical or illogical but can be verified statistically, called a
)A
statistical hypothesis.
Testing Hypothesis
The test statistic is a function or estimator of the sample observations upon which
(c
the statistical decision will be based. The rejection region (RR) specifies the values
of the test statistic for which the null hypothesis is rejected in favour of the alternative
hypothesis.
●● If for a particular sample, the computed value of the test statistic falls in RR, we
reject the null hypothesis Ho and accept the alternative hypothesis H1. Notes
ity
●● If the value of the test statistic does not fall into the rejection (critical) region, we
accept Ho. The region, other than the rejection region, is the acceptance region.
Significance Level
The significance level is the critical probability in choosing between the null and the
rs
alternative hypotheses. It is the probability level that is too low to warrant the support of
the null hypothesis.
ve
it is true.
Error Explanation
ity
Illustration:
A renowned channel of children’s conducts market research to find out the need for
)A
the contrary, in case the alternative hypothesis H1 is true, there is no need to advertise.
A hypothesis must be framed very carefully. There are high possibilities of errors
during each step of constructing a hypothesis. Note that hypotheses are constructed
Amity Directorate of Distance & Online Education
84 Research Methodology
in such a way that they cannot both be correct or incorrect simultaneously. To avoid
Notes erroneous decisions, we are required to select the first action (first hypothesis), if the
ity
first action is true,and the second one (second hypothesis), if the second action is true.
rs
determining the relationship between the states of nature and the recommended action.
The second chance of error may occur if the research findings stand erroneous w.r.t the
state of nature.
The below table would help us understand the relationship between the states of
nature and the recommended action. This comprises two plausible states of nature and
ve
two alternative courses of action.
H0 True H1 True
H0
H1
True
True
ni No error
Type II error
Type I error
No error
From the above table, it is evident that, given that the relationship between
U
the two states of nature and the two actions is true, there is no error if we arrive at
the conclusion that H0 is true. Likewise, there would be no error if we arrive at the
conclusion that H1 is true.
What would happen if the team accepts the wrong hypothesis? Would the
company take the wrong decision? Let me explain with the help of the two scenarios
ity
ed
H1 H1 wrongly rejected Type II Unnecessary
and H0 wrongly accept- costs incurred
ed due to advertise-
)A
ments
parameter under the null hypothesis, but not both. We formulate null and alternative
hypotheses for a one-tailed test as follows:
ity
Alternative Hypothesis H1 : μ < μ0 or μ > μ0
Inthe two-tailed test, the values of the parameter being studied under the
alternative hypothesis are allowed to be greater or less than the values of the
parameter under the null hypothesis.
rs
Null Hypothesis H0 : μ = μ0
Alternative Hypothesis H1: μ ≠ μ1
ve
4.1.7: Degree of Freedom
Definition
In a statistical calculation, the degrees of freedom represent the values involved
in a calculation have the freedom to vary. The degrees of freedom can be calculated
to help ensure the statistical validity of chi-square tests, t-tests and even the more
ni
advanced f-tests. These tests are commonly used to compare observed data with data
that would be expected to be obtained according to a specific hypothesis.
For example, let’s suppose a vaccine trial is conducted on a group of patients, and
U
it is hypothesized that the patients receiving the vaccine would show increased heart
rates compared to those that did not receive the vaccine. The results of the test could
then be analyzed to determine whether the difference in heart rates is considered
significant, and degrees of freedom are part of the calculations. Because degrees
of freedom calculations identify how many values in the final calculation are allowed
to vary, they can contribute to the validity of an outcome. These calculations are
ity
dependent upon the sample size, or observations, and the parameters to be estimated,
but generally, in statistics,
Here N is the number of values in the data set (sample size). For example, we
assume
Let the data set is Dwith the values: 15, 30, 25, 10
= (15+30+25+10)/4
=(80)/4
Notes
ity
= 20
This indicates that, in this data set, three numbers have the freedom to vary as
rs
long as the mean remains 20.
ve
A confidence interval is the mean of our estimate plus and minus the variation
in that estimate. Exactly this is the range of values that we expect our estimate to fall
between if we redo our test within a certain level of confidence.
ni
construct a confidence interval with a 95% confidence level, then we are confident that
95 out of 100 times, the estimate will fall between the upper and lower values specified
by the confidence interval.
In statistical test:
U
Confidence level = 1 − a
So if we use an alpha value of p < 0.05 for statistical significance, then our
confidence level would be 1 − 0.05 = 0.95, or 95%.
We can use confidence intervals for many kinds of statistical estimates, likes:
ity
Proportions
Population means
These are all point estimates, and don’t give any information about the variation
around the number. Confidence intervals are useful for communicating the variation
around a point estimate.
)A
Example: Variation around an estimate we survey 100 Indian and 100 Americans
about their television-watching habits, and find that both groups watch an average of 35
hours of television per week.
However, the Indian people surveyed had a wide variation in the number of hours
watched, while the Americans all watched similar amounts.
Even though both groups have the same point estimate (average number of hours
(c
watched), the Indian estimate will have a wider confidence interval than the American
estimate because there is more variation in the data.
ity
Bayesian statistics are a mathematical procedure that applies probabilities
to statistical problems.In Bayesian statistics, the interpretation of probability is a
description of how certain some statement, or proposition, is true.
rs
●● If the probability is 0.5, then we are in as uncertain state as we would be about a
fair coin toss.
If the probability is 0.95, then we’re quite sure the statement is true, but it wouldn’t
be too surprising to us if we found out the statement was false.
ve
ni
0......................probability..........................1
U
The above figure can say that probability can be used to describe degrees of
certainty or how plausible some statement is. 0 and 1 are the two extremes of the scale
and correspond to complete certainty. However, probabilities are not static quantities.
When we get more information, our probabilities can change.
ity
It might sound like there is nothing more to Bayesian statistics than just thinking
about a question and then blurting out a probability that feels appropriate. For example,
we may be on “Who Wants to Be a Millionaire?” and not know the answer to a question,
so we might think the probability that it is A is 25%. But if we call our friend using “phone
a friend”, and our friend says, “It’s definitely A”, then we would be much more confident
that it is A! our probability probably wouldn’t go all the way to 100%.
m
We will now look at a simple example to demonstrate the basics of how Bayesian
statistics works.
●● We start with some probabilities at the beginning of the problem are called prior
)A
probabilities.
●● And how exactly these get updated when we get more information? These
updated probabilities are called posterior probabilities.
●● To make all these clearer, we will use a table that we will call a Bayes’ Boxto help
us calculate the posterior probabilities easily.
Assume there are two balls in a bag, where:
(c
ity
white, and another is black so that we can label our two competing hypotheses BB and
BW. So, at the
beginning of the problem, we know that one and only oneof the following
statements/hypotheses is true:
rs
BW: One ball is black and the other is white.
ve
and observes its colour. The result of this experiment is
D: The ball that was removed from the bag was black.
A Bayesian analysis starts by choosing some values for the prior probabilities with
ni
our two competing hypotheses BB and BW, and we need to choose some probability
values to describe how sure we are that each of these is true. Since we are taking two
hypotheses, then there will be two prior probabilities, one for BB and one for BW. For
simplicity, we will assume that we don’t have much of an idea that is true, and so we will
use the following prior probabilities:
U
P (BB) = 0.5
P (BW) = 0.5.
The above two hypotheses are mutually exclusive (they can’t both be true) and
ity
exhaustive (one of these is true, it can’t be some undefined third option). The choice of
0.5 for the two prior probabilities describes the fact that, before we did the experiment,
we were very uncertain about which of the two hypotheses was true. Now present a
Byes’ Box, which lists all the hypotheses that might be true and the prior probabilities.
There are some extra columns which we haven’t discussed yet, and will be needed in
order to figure out the posterior probabilities in the final column. The first column of a
m
Totals: 1
this case there are just two. If need to construct a Bayes’ box for a new problem
then just think about what the possible answers to the problem are, and list them in
the first column. The second column lists the prior probabilities for each of the
hypotheses. Above, before we did the experiment, we decided to say that there was
a 50% probability that BB is true and a 50% probability that BW is true, hence the 0.5
(c
values in this column. The prior column should always sum to 1. Remember, the prior
probabilities only describe our initial uncertainty, before taking the data into account.
Likelihood
Notes
ity
The third column is called likelihood by which we cancalculating the posterior
probabilities. It is synonymous with probability. In statistics, likelihood is a very specific
kind of probability. To fill in the third column of the Bayes’ Box, we need to calculate two
likelihoods, so you can tell from this that the likelihood is something different for each
hypothesis. But what is it exactly?
Here is the Bayes’ Box with the likelihood column filled in.
rs
Hypotheses prior likelihood h = prior × likelihood posterior
BB 0.5 1
BW 0.5 0.5
ve
Totals: 1
First calculate the value of the likelihood for the BB hypothesis. Remember, the
data we are analysing here is that we chose one of the balls in the bag “at random”, and
it was black. The likelihood for the BB hypothesis is therefore the probability that we
would get a black ball if BB is true.
ni
Imagine that BB is true. That means both balls are black. What is the probability
that the experiment would result in a black ball? That’s easy – it’s 100%! So we put the
number 1 in the Bayes Box as the likelihood for the BB hypothesis.
U
Now imagine instead that BW is true. That would mean one ball is black and
the other is white. If this were the case and we did the experiment, what would be
the probability of getting the black ball in the experiment? Since one of the two balls
is black, the chance of choosing this one is 50%. Therefore, the likelihood for the BW
hypothesis is 0.5, and that’s why we put 0.5 in the Bayes’ Box for the likelihood for BW.
ity
In general, the likelihood is the probability of the data that you actually got,
assuming a particular hypothesis is true. In this example, it was fairly easy to get the
likelihoods directly by asking, “if this hypothesis is true, what is the probability of getting
)A
the black ball when we do the experiment?”. Sometimes this is not so easy, and it can
be helpful to think about ALL possible experimental outcomes/data you might have seen
– even though ultimately, we just need to select the one that actually occurred.
ity
The third column of the Bayes’ Box is the product of the prior probabilities and the
likelihoods, calculated by simple multiplication. The result will be called “prior times
likelihood”, but occasionally we will use the letter h for these quantities. This is the un
normalized posterior. It does not sum to 1 as the posterior probabilities should, but it is
at least proportional to the actual posterior probabilities.
To find the posterior probabilities, we take the prior likelihood column and divide
rs
it by its sum, producing numbers that do sum to 1. This gives us the final posterior
probabilities, which were the goal all along. The completed Bayes’ Box is shown below:
ve
BB 0.5 1 0.5 0.667
We can see that the posterior probabilities are not the same as the prior
ni
probabilities because we have more information now! The experimental result made BB
a little bit more plausible than it was before. Its probability has increased from 1/2 to 2/3.
Interpretation
U
The posterior probabilities of the hypotheses are proportional to the prior
probabilistic and the likelihoods. A high prior probability will help a hypothesis have a
high posterior probability. To understand what this means about reasoning, consider the
meanings of the prior and the likelihood. There are two things that can contribute to a
hypothesis being plausible:
ity
Suppose the prior probability is high. That is, the hypothesis was already plausible
before we got the data.
Suppose the hypothesis predicted the data well. That is, the data was what we
would have expected to occur if the hypothesis had been true.
m
Bayes Rule:
Bayes’ rule is an equation from probability theory and conditional probabilities.
For example, the left-hand side of the equation is P (A B), and that means the
probability of A given B. That is, it’s the probability of A after taking into account the
)A
The reason for these letters is that we should interpret H as hypothesis and D as data.
Then we can interpret Bayes’ rule as telling you the probability of a hypothesis given
some data, another words, a posterior probability.
ity
In Bayesian statistics, most of the terms in Bayes’ rule have special names. Some
of them even have more than one name, with different scientific communities preferring
different terminology. Here is a list of the various terms and the names we will use for
them:
rs
probabilities is the main goal of Bayesian statistics!
P (H) is the prior probability, which describes how sure we were that H was true
before we observed the data D.
ve
P (D\ H) is the likelihood. If you were to assume that H is true, this is the probability
that we would have observed data D.
P (D) is the marginal likelihood. This is the probability that we would have
observed data D, whether H is true or not
Phone Example
ni
This example is based on the question from an example in David MacKay’s
wonderful book “Information Theory, Inference and Learning Algorithms” (available
online as a free PDF download. You’re welcome to check it out, but it is a large book
and only about 20% of the content is relevant to this course!).
U
You move into a new house which has a phone installed. You can’t remember
the phone number, but you suspect it might be 555-3226 (some of you may recognise
this as being the phone number for Homer Simpson’s “Mr Plow” business). To test this
hypothesis, you carry out an experiment by picking up the phone and dialling 555-3226.
ity
If you are correct about the phone number, you will definitely hear a busy signal
because you are calling yourself. If you are incorrect, the probability of hearing a busy
signal is 1/100. However, all of that is only true if you assume the phone is working, and
it might be broken! If the phone is broken, it will always give a busy signal.
When you do the experiment, the outcome (the data) is that you do actually get
m
the busy signal. The question asked us to consider the following four hypotheses and
to calculate their posterior probabilities: Note that the four hypotheses are mutually
exclusive and
The above table has four hypotheses about the state of the phone and the phone
Notes number. The prior probabilities are also given.
ity
exhaustive. If you were to come up with hypotheses yourself, “phone is working”
and “555-3226 is correct” might spring to mind. They wouldn’t be mutually exclusive, so
you couldn’t do a Bayes’ Box with just those two, but it is possible to put these together
(using “and”) to make the four mutually exclusive options in the table.( Do Yourself)
rs
Summary :
From above, we can be knowledgeable with the following
ve
●● Testing Hypothesis and significance levels
●● Degree of Freedom
●● One-tailed and Two-tailed test
ni
Questions:
●● What are the type-1 and type-2 errors discuss with example
●● What is the significance level of hypothesis.
●● Discus about the degree of freedom with the following
U
●● Example: a + b/2 = 5. fix a = 3, b has to be 7.
●● Which are the confidence interval.
ity
m
)A
(c
ity
Unit Outcomes:
●● Discuss Total, Treatment & error of Squares
●● ANOVA for random effect model
●● Estimation of Variance Components
rs
●● Methods of Variance Component Estimation
●● Model Adequacy Checking
Introduction:
ve
Definition: Variance analysis is that the study of deviations of actual behaviour
versus forecasted or planned behavior. This can be essentially concerned with how
the difference of actual and planned behaviors indicates how business performance is
being impacted.
3. Service delivery might not be up to the mark, e.g. planning may have taken into
consideration an eight hour every day, however actual ground conditions may only
allow six hours a day
4. In certain cases, there can be no basis for planning, e.g. output of creative activities
cannot be benchmarked to a high level of accuracy.
m
and the quantity initially allocated for production. This can also happen due to the
difference in price paid and the price budgeted for materials used.
2. Labour variances:- This denotes the particular wage paid to workers versus the
quality wage prevalent for the output specified. When the actual labour costs are
more than budgeted ones, the variance is unfavourable.
(c
3. Overhead variances:- It may be defined as the sum total of indirect material, labour
and expense costs. Overhead variances may arise due to the difference between
standard overhead costs budgeted and the actual overheads incurred.
ity
This is easy to understand and few words with practical examples will make a
whole lot of sense while design and analysis of an actual experiment (in the next
section of this module). Curious and interested readers can consult the Oscar
Kempthorne textbooks for more real sugar (Kempthorne 1952; Hinkelmann and
Kempthorne 2005; Hinkelmann and Kempthorne 2008)!
Although identical twins are used for the experimental units with the exact same
rs
experimental conditions, there will be natural variations in almost anything in this world.
Therefore, the response is the function of the factors or the explanatory variables with
some experimental error as in Equation 1.
ve
Means Model
In the fuel economy study, the research question is “whether there is a difference
in fuel types with respect to the mean miles per gallon, mpg.” Therefore, Equation 1 can
be written for testing the mean difference in response (mpg) from treatment to treatment
ni
(different fuel types in this case) as in Equation 2. As the response is written with
respect to the mean treatment effect, this model is known as the means model. More
precisely, the population means model as the model is intended for the generalization
of the sample results for the population. It can also be noted that the model is linear as
the power of the mean term (μ) is one
U
yij = µi +εij
The error term εij=is assumed to be identical, independent and normally distributed.
ity
Effects Model
Assume that the three fuel types are made by mixing three different ingredients
that improve fuel economy when added with a common base fuel. The base fuel without
the ingredients will run the vehicle just fine with some mileage. If the ingredient effects
are separated from the base fuel economy, the model Equation 2 can be written as
m
Equation 3.
yij = µ +εij+ Ƭi
The error term εij =is assumed to be identical, independent and normally distributed
Error
The error could occur from many different sources, including the following
(c
Experimental Error
●● treatment error - error due to our inability to replicate a treatment from one
application to the next
Amity Directorate of Distance & Online Education
Research Methodology 95
ity
Observational Error
●● measurement error - error due to imprecision in our measurement or scoring
procedure
●● sampling error - error due to the random selection of observational units
rs
(OUs) for the investigation.
ve
the SSTR(treatment sum of squares) and SST( total sum of squares). When we
compute SSE, SSTR, and SST, you then find the error mean square (MSE) and
treatment mean square (MSTR), from which we can then compute the test statistic.
After you find the SSE, our next step is to compute the SSTR. This is a measure of
ni
how much variation there is among the mean lifetimes of the battery types. With a low
SSTR, the mean lifetimes of the different battery types are similar to each other.
Firstly we need to calculate the overall average for the sample, known as
the overall mean or grand mean.
U
For example, say a manufacturer randomly chooses a sample of four Electrical
batteries, four Ready forever batteries, and four Voltages now batteries and then tests
their lifetimes. This table lists the results (in hundreds of hours).
ity
If we have 12 total observations (four batteries chosen from each of three battery
types, as shown in the table), then we may obtain the overall mean by adding up the
12 sample values and dividing by 12:
)A
we then compute the SSTR with the following steps for each column:
1. Compute the squared difference between the column mean and the overall
(c
mean.
2. Multiply the result by the number of elements in the column.
So in this example, SSTR equals
Amity Directorate of Distance & Online Education
96 Research Methodology
Notes
ity
The calculations are based on the following results:
rs
After you compute SSE and SSTR, the sum of these terms is calculated, giving the
SST.
ve
The total sum of squares (SST) equals the sum of the SSTR and the SSE. So
using the battery example, we get
ni
Two way (factor) ANOVAThis is an extension of the one-factor situation to take
account of a second factor. The levels of this second factor are often determined by
groupings of subjects or units used in the investigation. As such, it is often called a
U
blocking factor because it places subjects or units into homogeneous groups called
blocks. The design itself is then called a randomised block design.
Solution
In (1), although the 20 programs are similar, any differencesbetween them may
m
affect the compilation times and hence perhaps any conclusions. Thus within the ‘worst
scenario’, the 5 programsallocated to what’s really the fastest compiler could be the
5requiring the longest compilation times, leading to the compiler appearing to be the
slowest! If used, the results would require a one factor analysis of variance; the factor
)A
In (2), since all 5 programs are run on each compiler, differences between
programs must not affect the results. Indeed it’s going to be advantageous to use 5
programs that differ markedly so thatcomparisons of compilation times are more
general. For thisdesign, there are two factors; compiler (4 levels) and program (5
levels). The factor of principal interest is compiler whereas theopposite factor, program,
(c
The actual compilation times, in milliseconds, for this two factor (randomised block)
design are shown in the following table. Notes
ity
1 Compiler
2 3 4
Program A 29.21 28.25 28.20 28.62
rs
Program C 30.91 30.18 30.52 30.09
ve
Assumptions and interaction
The 3 assumptions for a 2 correlational analysis of variance when there’s only one
observed measurement at every combination oflevels of the two factors are as follows.
●●
distributed.
ni
The population at every issue-level combination is (approximately) usually
(
N mij,s2 . )
ity
sugar (levels none and 2 teaspoons), and stirring (levels none and 1 minute), on
the sweetness of a cup of tea.
m
Stirring has no effect on sweetness if sugar is not added butcertainly does have an
effect if sugar is added. Similarly, adding sugar has little effect on sweetness unless the
tea is stirred.
)A
Interaction can only be assessed if more than one measurement is taken at each
combination of the factor levels. Since suchsituations are beyond the scope of this text,
it will always beassumed that interaction between the two factors does not exist.
Thus, for example, since it would be most unusual to find onecompiler particularly
suited to one program, the assumption of no interaction between compilers and
(c
Is it likely that the assumption of no interaction is valid for the data you collected in
ity
Notation and computational formulae
As illustrated earlier, the data for a two-factor ANOVA can bedisplayed in a two-way table.
It is thus convenient, in general, to label the factors asa row factor and a column factor.
rs
Number of levels of column factor =c
ve
(ithlevel of row factor and i=1,2,K,r
ni
Sum of r observations in jth column =TCj=åx
These lead to the following computational formulae which again are similar to
U
those for one factor anova except that there is anadditional sum of squares, etc for the
second factor.
ity
m
What are the degrees of freedom for SST , SSR and SSC whenthere are 20
)A
For a two factor analysis of variance this takes the following form
(c
ity
squares freedom square
Between rows SSR r-1 MSR MSR MS
E
Between columns SSC c-1 MSC MSC MS
rs
Error (residual) SSE (r - 1)(c - 1) MSE
ve
4.2.4: Estimation of Variance Components
Abstract
Estimating variance components is a method often used in population genetics and
its application in animal breeding. Even experienced population geneticists nowadays
feel lost if confronted with the huge set of different methods in variance component
ni
estimation,especially due to the fact that there exists no uniformly best method; a
decision on which method should be used is often difficult to fell.However,the one-way
lay-out is of limited practical interest but can best be used to explain to animal scientists
the basic principles(valid for higher classifications) of the methods. The symbols used
U
are the standard biometric symbols as given in Raschet. al. (1994). We can say that all
the methods offered by SPSS can be recommended.
Introduction
We characterize methods of variance component estimation for the simplest case,
ity
the one-way ANOVA and demonstrate most of them by some data set. For this, we
assume that a sample of levels of a random factor A1 has been drawn from the universe
of factor levels which is assumed to be large. Not to be too abstract, let us assume that
the levels are sires. From the i-th sire a random sample of ni daughters is drawn and
a
their milk yield yij recorded. The scheme of the N =
∑n i
observations is given in Table
m
i =1
1. This case is called balanced if for each of the sires the same number of daughters
has been selected. Balancedness in higher nested classification means equal
subclass numbers as well as equal numbers of levels of nested factors within each
)A
of the levels of factors of a higher order in the hierarchy. Some of the methods
described below differ from others only in the case of unbalanced data what means
in the one-way ANOVA that not for all the sires the number of daughters are the
same. Therefore we simulated an unbalanced data set.
Not for all methods formulae will be given and only the principle is explained. The
reason is that we wrote this paper mainly for non-mathematicians who often use several
methods and need some basic understanding of what they are doing in applying special
(c
ity
level A1 level A2 … level Aa
y11 y21 … ya1
y12 y22 . ya2
.
. . .
.
. . .
.
. .
rs
y1n1 y 2n2 … y an a
(1)
ve
The ai are the main effects of the levels Ai. They are random variables. The e.ijare
the errors and also random. The constant µis the overall mean. Model (1) is completed by
the assumptions (E() = expectation of=mean of;:V() = variance of).
E(a.i) = 0, V(a.i) = σ 2a , E(eij)=0, V(eij)=σ2 ; all components on the right hand side of (1)
are independent (2)
σ a2
ni
and s2are called variance components. The total number of observations is
always denoted by N , in the balanced case we have N = an . In the sequel we give the
formulae for the balanced case, generalization can be found in the references.
U
Let us assume that all the random variables in (1) are normally distributed even if
this is not needed for all the methods.
The normal distribution (Gauss-distribution) has the density (or likelihood) function
ity
1
1 - ( y − µ )2
f ( y )= e 2σ 2
(σ > 0 ) (3)
σ 2π
From (1), (2) and the normality assumption it follows that the ai and the eijare
m
independently from each other N(0;σa2) - and N(0;σ2) – distributed respectively. The
yijare not independent from each other N(µ; σ2+ σa2) – distributed.
The dependence exists between variables within the same factor level because
( ) ( )
cov y ij , y il = cov µ + a i + e ij , µ + a i + e il = cov(a i , a i ) = var(a i ) = σ a , if j ≠ l .
2
)A
σ a2
ρ IC = 2
σa +σ 2
(c
Table 2: ANOVA Table of the one-way ANOVA model II in the balanced case
Notes
ity
Source of SS df MS E(MS)
variation
Factor A a n a-1 s2+nsa2
= ∑∑ ( y i⋅ − y )
2
S S A
A
M
S = n
i =1 j =1 A
a −1 σ2+ ∑ ai2
a −1
rs
Residual a n a(n-1) s2
= ∑∑ ( y ij − y i⋅ )
2
S e s
R S Res
M
S =
a(n − 1)
i =1 j =1 Res
Total an-1
ve
a n
= ∑∑ ( y − y)
2
S T ij
i =1 j =1
ni
1
y i. =
ni
∑y ij Some of the methods are developed for unbalanced data. We
ni
j =1
assume therefore also a generalization of (1) by a general linear model for one random
factor in (5).
Y = Xa + Uâ + e
U
(5)
is carried out in order to measure and diagnose the true value of a function or a
particular of information. It is done on the basis of observations on the samples, which
are a combined piece of the target population or function. Several statistics are used to
perform the task of estimation.
There are two very important properties of estimators that are used: the estimator
and the estimate. To understand the concept of the estimator and estimate in detail,
(c
we will use an example. Let’s say that a1, a2, a3 and so on is a collection of samples
from some group of a certain population with ‘x’ as its parameter. Here, if T= T (a) is
a statistic, then E (T(a)) = x. From these equations we can realize that an estimation
of the statistic has been carried out, where the statistic T is an estimator and the
ity
Estimation has many important properties for the ideal estimator. These properties
include unbiased nature, efficiency, consistency and sufficiency.
The estimators that are unbiased while performing estimation are those that have
0 bias results for the entire values of the parameter. Hence, Unbiased estimators are
those where the mathematical expectation or the mean proves to be the parameter of
the target research. In the above mentioned example for estimation, T is going to be
rs
the unbiased estimator only if its estimate comes out to be equal to ‘x.’
In estimation, the estimators that give consistent estimates are said to be the
consistent estimators. As the number of random variables increase, the degree of
concentration should be higher and higher around the estimate in order to make the
ve
estimator of estimation the consistent estimator. If the estimator gives an unbiased
estimate and the variance of the estimator comes out to be zero, then the estimator of
estimation is called the consistent estimator. These two conditions need to be fulfilled
only if the numbers of random variables reach infinity.
In estimation there are many estimators that have ample incidences of consistent
ni
estimators, and according to the property of efficiency in estimation, the consistent
estimators should be normally distributed. This kind of property was taken into account
in the theory of estimation, because there were incidents of the estimators having
ample consistent estimations but were not the efficient estimators.
U
An estimator is known as the sufficient estimator only when the joint conditional
distribution function of the sample/observation has the condition of T1 T2 T3 T4 (and
so on and so forth), and are the values under the given estimator function ‘T.’ Thus, the
resultant joint conditional estimation has to be absolutely sovereign of the parameter
‘x.’While carrying out the task of estimation, a researcher should always know that the
ity
best estimator is the one that is the minimum variance unbiased estimator (MVUE).
This means that the estimator has the minimum variability when it is compared to other
estimators.
general considerations.
an estimate θˆ( y ) .
Definition 2: An estimator () ()
èˆ = èˆ ( y ) is unbiased if E èˆ = θ . The difference E èˆ
- θ = vn(θ ) is the bias of an estimator. The bias of an unbiased estimator is zero. The
(c
{
expression is
} ()
E èˆ − θ = V èˆ + v 2 n (θ )
2
èˆ = èˆ ( y ) (MSE). The mean squared error of an unbiased estimator equals its Notes
ity
variance. In Figure 1 èˆ1 èˆ 2 n
∑ yi
Example 1The estimator ìˆ = = y of the mean µ of the components of a
i =1
n
random sample yT = ( y ,..., y ) is unbiased.
1 n
rs
linear (MVUL) orminimum variance unbiased quadratic (MVUQ) estimator, if its variance
ve
Definition 4 : An estimator
The estimator
∑y i
ìˆ = i =1
= y of the mean µ in example 1 is a minimum variance
m
unbiased linear estimator.
ni
Definition 5: Let f(y,θ) be the density function of θ , the parameter of the distribution
of a random variable y . This is a function of two variables, θ and y. If we call itdensity
function we consider it as a function of y for a fixed θ. But the same function f(y,θ) as a
function of θ for fixed y is called likelihood function. If we estimate θ sothat it maximizes
U
the likelihood function in the parameter space of θ , we call the corresponding estimator
a Maximum- Likelihood Estimator (MLE).
Note :For more details see Rao (1971), the first paper on MINQUE.
remark. In previous topics we discus about the model equation of the random
variables y. Mathematical operations as minimizing (least squares) or maximizing
(likelihood) can be performed for the realizations only. The result of such an
m
s2+nsa2 = MSA
(c
s2= MSres
s2= MSres
Notes 1
ity
sa2 = (MSA – MSres)
n
Properties of the estimators:
rs
The probability of a negative estimator is given by
.
(
P sa
2
)
< 0 = P F (a − 1, a(n − 1)) <
1
ve
σ 2
1 + n a2
σ
( ( ))
Where F a − 1, a n − 1 is a random variable with an F –distribution with a-1 and
a(n-1) d.f.Tables for these probabilities are given by Verdooren (1982).
(
)
P s a < 0 = P F (9 ,2497 ) < q =
2
1
1 + 26
ni=
1 5,3
1
= 0,1875 = 0.
U
6
2
For some other values of σ a and σ 2 0 1 the corresponding probabilities can be
found in bellow table
Table for2 Probabilities of negative estimates for f1=99, f2=2497, n = 26 and several
σ
ity
values of a
h2 q probability
value of σ a 2 /σ 2
6 0,4 0,19 ,000000000
20 0,17 0,43 ,000000233
m
ity
quasi-maximum-likelihood method. If we replace in this minimum the
realizations of the random variables by the random variables we obtain
the quasi-maximum likelihood estimator. Because the family of normal
distributions is a two-parametric exponential family with the set of
complete minimal sufficient statistics (see Rasch (1995))
rs
1 a n
y. =
an
∑∑ y
i =1 j =1
ij , S res ,S A ,
ve
1 1 1 an
− 2 S res + S A+ ( y. − µ )2
2 σ σ 2 + nσ 2 a σ 2 + nσ 2 a
e
f ( y1 ,..., y ab ) = a ( n −1) a
(2π ) 2 (σ 2 ) (σ )
an
2 2
+ σ 2a 2
With respect to µderiving the logarithmof this likelihood function and the two
variance components without side conditions to restrict the solutions to the parameter
space leads to the following estimators:
µˆ = y. ni
U
s2= MSres
1 a −1
sa2 = ( MSA – MSres)
n a
ity
The estimator of µ is needed to replace the µ in the equations stemming from the
derivations with respect to the variance components only.
S res +S A
S2 =min[MSres, ]
an
a − 1
)A
s 2 a = max M
S A −M
S res ;0 .
a
Both estimators are biased.
the restriction that the solutions are non-negative. The solutions are:
Notes
ity
S res+S A
s2 =min[MSres, ],
an − 1
1
s2a = max{[M
S A −M
S res ];0}
n
rs
Modified Maximum-Likelihood method (MML)
ML-estimators which have uniformly smaller mean squared error compared with
the ML estimators of Herbach , derived by Stein (1964) and Kotz et al. (1969) .
ve
S res+S A S res + S A + an y 2 .
s2 =min[MSres, , ],
an + 1 an + 2
ni
S A + an y .
2
1 a − 1
s2a = min max M
S A −M
S res ;0 ; max −M
S res ,0
n a + 1
a+2
Federer´s Estimator:
U
Some of the truncated estimators above are inadmissible from Bayesian point of
view. The following non-truncated and nonnegative estimators could not be proved to
be inadmissible but also not be proved to be admissible (there is no uniformly better
estimator). The advantage of the estimators proposed by Federer (1968) is that they
ity
S2= MSres
1
sa2 [ MSA – MSres (1 − e −δM
S A
) ]
n
1
For δFederer proposed to choose a value in the interval δ ∈ 0, .
m
1 S res
In our numerical example we will useδ = 0 and δ= .
S res
4.2.7 :Model Adequacy Checking
)A
ity
violations are small, the ultimate result might not be changed. But if the deviations
are large, the model obtained may become unstable within the sense that a different
sample could lead to an entirely different model with opposite conclusions. One main
thingconfine mind is that these assumptions are for the population, and that we work
only with a sample. Therefore the main issue is to create a choice about the population
on the basis of a sample of knowledge. Several diagnostic methods to check the
rs
violation of regression assumption are based on the study of model residuals with the
help of varities of graphics.
ve
1. Case of one explanatory variable If there’s only one explanatory variable in a model,
then it’s easy to test the existence of the linear relationship between y and X by
scatter diagram with the assistance of accessible data. If the scatter diagram shows
a linear trend, it indicates that the relationship between y and X is linear. If the pattern
is not linear, then it suggests that the relationship between y and X is nonlinear. For
instance, the subsequent figure indicates a linear trend
ni
U
ity
Case of more than one explanatory variables : To test the idea of linearity
among the study variable and also the explanatory variables, the scatter plot matrix of
the information is used. A scatter plot matrix could be a two-dimensional array of two-
dimension plots where each form contains a scatter diagram aside fromthe diagonal.
Each scenario established the link between a pair of variables. It gives more information
(c
than the coefficient of correlation between each pair of variables because it provides
a way of linearity or nonlinearity of the relationship and some awareness of how the
individual data points are arranged over the region. It is a scatter diagram of (Y versus
X1 ), (Y versus X 2), …, (Y versus X3).
Amity Directorate of Distance & Online Education
108 Research Methodology
ity
●● display the scatterplots within the upper triangular part of the plotmatrix.
●● Mention the corresponding correlation coefficients within the lower triangular
part of thematrix.
Suppose there are only two explanatory variables and the model is y =X1b1 +X 2
b2 +ε,matrix looks like as follows.
rs
ve
ni
U
ity
Residual analysis
The residual is outlined because as the distinction between the discovered and
m
ei=yi~ŷi=yi-ŷi,i=1,2,...,nith
whereyi is an observation and ŷiisthecorrespondingfittedvalue
)A
Residual can be viewed as the deviation between the data and the fit. So it is
also a measure of the variability in the response variable that is not explained by the
regressionmodel. Residuals can be thought of as the observed values of the model
errors. So it can be expected that if there is any departure from the assumptions on
random errors, then it should be shown up by the residual. The analysis of residuals
facilitate to find the modelinadequacies.
(c
Assuming that the OLSE estimates the regression coefficients in the model y =X
β+ε, we find that:
●● Residuals have zero meanas
E(ei)=E(yi-ŷi)
Notes
ity
=E(X i β+ei -X i b)
=Xi β+ 0 -Xi b
=0
rs
1. Analysis of variance is a statistical method of comparing the of several populations.
a. Means
b. Variances
ve
c. Standard Deviations
d. None Of The Above
2. Sum of squares calculate the variability of the observed values around their respective
treatment means.
a. Error
b. Total
c. Treatment
d. Interaction
ni
U
3. When conducting an ANOVA, FDATA will always fall within ....... and.........range?
a. between 0 and infinity
b. between 0 and 1
c. between negative infinity and infinity
ity
5. In ANOVA with 4 groups and a total sample size of 44, the computed F statistic is
2.33 In this case, the p-value is:
A. < than 0.05
B. exactly 0.05
C. cannot tell - it depends on what the SSE is
(c
ity
the critical value?
A. MSTR, MSE
B. mean, sample standard deviation
C. expected frequency, obtained frequency
rs
D. sample size, number of groups
7. Residual can be viewed as the deviation between the data and the fit(T/F)?
8. A scatter plot matrix could be a two-dimensional array of 3-dimension plots(T/F)?
9. Maximum likelihood estimate is obtained by maximizing the likelihood function of the
ve
sample(T/F)
10. Estimating variance components is a method often used in population genetics and
its application in animal breeding(T/F)
11. Which of the following is FALSE about hypothesis?
ni
a. Hypothesis is a tentative statement, which is subject to verification
b. Hypothesis is a tested, well-substantiated, complete explanation for a set of
proven factors
c. Hypothesis is conceptually different from theory
U
d. Hypothesis is a testable relationship between at least two variables
12. A research firm conducts a study to establish the minimum purchasing power
required for the medium and large retail stores as Rs. 150 million and Rs. 300 million,
respectively. Identify the INCORRECT statement.
ity
a. The null hypothesis is - total purchasing power is less than Rs. 150 million
b. One of the alternative hypotheses is - total purchasing power is between Rs.
150 million and Rs. 300 million
c. The null hypothesis is - total purchasing power is more than Rs. 300 million
d. One of the alternative hypotheses is - total purchasing power is more than Rs.
m
300 million
13. A renowned automobile company is under the process of launching a new luxury
car. The car is aimed at catering to the HNI (High Net Worth Individual) population.
The company wants to conduct a detailed market research on the popular choices
)A
of luxury cars in the country. It recruits a research team, which comes up with
the research problem - “Is luxury car a popular among HNI clients?” Which of the
following is FALSE about the case?
a. “The HNI population does not prefer luxury cars” can possibly be the null
hypothesis
(c
b. “At least 50% of the HNI population prefers luxury cars” can possibly be one of
the alternative hypotheses
c. Only one alternative hypothesis exists in this case
d. “Less than 50% of the HNI population prefers luxury cars” can possibly be one
of the alternative hypotheses Notes
ity
14. Read the below statements and identify the wrong one(s). 1 - Complex hypothesis
establishes a causal relationship between two variables 2 - Null hypothesis establishes
a relationship among more than two variables 3 - An alternative hypothesis is used
for a reverse strategy 4 - “Performance at work is not related to salary alone” is an
example of alternative hypothesis
rs
a. Only 1 and 3 are wrong
b. Only 4 is wrong
c. Only 2 and 4 are wrong
d. All 4 statements are wrong
ve
15. A renowned Fashion magazine conducts a market research to find out the need for
advertisement. The research team constructs the below hypotheses: H0: At least 30
% of the readers consists of women H1: Less than 30 % of the readers consists of
women What decision would the management take if the research team commits a
type II error?
Ans:
ity
1. a
2. a
3. a
4. a
5. d
m
6. d
7. T
8. F
)A
9. T
10. T
11. b
12. c
13. c
(c
14. d
15. b
ity
Structure:
rs
5.1.2 Structure of Scientific Report
5.1.3 Components of a Scientific Report
5.1.4 Types of Reports
5.1.5 Steps in Report Writing
ve
5.1.6 Format & Writing Style
5.1.7 Illustrations & Tables
5.1.8 Citing & Referencing Sources
5.1.9 Footnotes
Unit-5.2: Presentation
5.2.1 Oral Presentation
5.2.2 Making Presentation
ni
U
5.2.3 Use of Visual Aids
5.2.4 Importance of Effective Communication
5.2.5 Convention & Strategies of Authentication
ity
ity
Unit Objectives:
●● Research Report Writing
●● Presentation
●● Writing Research Paper
rs
●● Planning a Project Proposal
ve
sciences, social sciences, engineering and business disciplines. Basically
searching pertaining toa particular task are written up into a report. It may be noted that
reports are consideredas a legal documents in every workplace.
●●
●●
a predefinedstructure,
independent sections ni
U
●● reaching unbiased conclusions.
Predefined structure: InBigger sense, these headings may indicate sections within
a report, suchas an introduction, discussion, and conclusion.
The preliminaries
●● Title page.
●● Preface or foreword, acknowledgements.
)A
1. Introduction
●● Objectives of the study, statement of the matter, hypotheses and definition of
(c
ideas.
●● Review of literature and analysis studies.
●● Time, place and materials of the survey.
ity
●● Methods, tools and techniques used for knowledge gathering.
rs
●● Summary of findings and recommendations.
ve
●● Appendices— questionnaires/statistical tables etc.
●● Glossary of terms
●● Index
ni
A technical report is mainly written for fellow researchers and therefore might be
organized on a different footing altogether. In such kind of report, the researcher is
expected to give a full detail of the technical aspects, both in the sampling methods and
the subject matter. Fellow professionals are more concerned about the methods used.
U
In fact, the amount of the searching depends on the techniques adopted. Conceptual
and analytical framework sample design must be clearly explained. A technical report
supports the bellow characteristics.
●● Main searching and contents: It will contain the main searching within two or three
pages.
ity
●● Nature of the research work:Which describes the general objectives of the study,
formulation of the problem in operational items, the working hypothesis, the type of
analysis, data required, etc.
●● Research methodology: Explains the various methods used and their limitations.
For instance, sample size, sample selection, etc.
m
●● Data analysis: The report must analyze the data and their sources, characteristics
and limitations. If secondary data are used, then these data should be suitable to
the problem. In case of a survey, the manner of data collecting method should be
fully described.
)A
●● Presentation of findings: The researcher presents his main outcomes of the study
with supporting data in the form of tables and charts. This part is the main body of
the report, usually extending over several chapters.
●● Bibliography: This contains the main sources of secondary data.
●● Technical appendices: These contain all technical matters relating to
(c
ity
There are two types of reports (1) Oral report (2) Written report
1. Oral Report When the researchers are asked to make an oral presentation then
they prepare this type of oral report.It is more difficult to prepare as compared to
the written report. Because here the reporter interact directly with the audience. Any
delay during an oral presentation can leave a negative impression on the audience.
This may also lower the self-confidence of the presenter. Communication plays a
rs
big role In an oral presentation. A lot of planning and thinking is required to decide
‘What to say’, ‘How to say’, ‘How much to say’. The presenter should have to face
a bunch of questions from the audience. A lot of preparation is required; the broad
classification of an oral presentation is as follows. Nature of an Oral
ve
Presentation Opening: A descriptive statement may be made on the basis of nature
of discussion that will follow. The opening statement might explain the characteristics of
the project, how it came about and what was attempted.
ni
the end of the presentation, question-answer session may be follow from the audience.
Procedure of presentation: Visuals, if need to be exhibited then can be use. The use of
tabular form for statistical information must help the audience.
●● Vital data such as figures may be printed and circulated to the audience so
that their ability to comprehend increases, since they can refer to it when the
presentation is going on.
●● The presenter should know his target audience well in advance to prepare
)A
(2) Weekly
(3) Monthly
(4) Quarterly
Notes
ity
(5) Yearly
(B) Type of reports:
(1) Short report
(2) Long report
(3) Formal report
rs
(4) Informal report
(5) Government report
Short report: Short reports are written when the problem is very well defined and if
ve
the scope is limited.
For example, Monthly sales report. It will run into about five pages. Itholds the
report about the advancement made with respect to a specific product in a particular
geographical locations.
ni
It canrepresent the outcome of the research in detail.
(a) Technical report: This will include the sources of data, research procedure,
and sample Design, tools employed for collecting info , information analysis
technique must be used, appendix,Conclusion and detailed recommendations
U
with respect to specific findings. If anyjournal, paper or periodical is referred,
these references should be given for the benefitof reader.
(b) Non-technical report: This report is meant for those who are not technically
qualified.
ity
4. Informal report: The report published by the supervisor by way of filling the shift log
book, to be used by his colleagues.
5. Government report: These may be prepared by state governments or the central
governmenton a given issue.
)A
1. Title Page
2. Page Contents
3. Executive Summary
Notes
ity
4. Body
5. Conclusions and Recommendations
6. Bibliography
7. Appendix
1. Title Page: Title Page should indicate the topic on which the report is prepared. It
rs
should include the name of the person or agency who has prepared the report.
2. Table of Contents: The table of contents shows actually “what the report contains”.
The table of contents should indicate the various parts or sections of the report. It
might be indicate the chapter headings along with the page number.
ve
Chapter no. Title of the chapter Page no.
1 Declaration
2 Certificates
3 Acknowledgement
4
5
6
Executive summary
Introduction to the project
Research design and methodology ni
Theoretical perspective of the study
U
Company and industry profile
Data analysis and interpretation
Summary of findings, suggestions and
conclusions
ity
Bibliography
Appendix
3. Executive Summary: If the report is long and drawn out, the person to whom we
have prepared the report may not have the time to read it in detail. For that introducing
an executive summary will help in highlighting major points. It is a summarized
m
version of the whole report. It should be written in one or two pages. Since top
executives read only the executive summary, it should be accurate and well-written.
An executive summary should help in decision-making.
An executive summary should have,
)A
●● Objectives
●● Brief methodology
●● Important findings
●● Key results
●● Conclusion
(c
●● Methodology
Notes ●● Limitations
ity
●● Analysis and interpretations
Introduction: Explain clearly the decision problem and research objective. The
background knowledge should be assign on the product and services provided by the
organization which is under study.
rs
As an instance, was primary data collected or secondary data used? Was a
questionnaire used? What was the sample size and sampling plan and method of
analysis? Was the design exploratory or conclusive?
ve
Limitations: Every report will have some limitation such as time, geographical
area, the methodology adopted, correctness of the responses, etc.
Analysis and interpretations: collected data will be tabulated. Statistical tools if any
will Notes be applied to make analysis and to take decisions.
ni
(a) What was the outcomes drawn from the study?
(b) Based on the study, what recommendation do you make?
6. Bibliography: If the report are based on secondary data, use a bibliography section
U
to list the publications or sources that you have consulted. The bibliography might
include the title of the book, name of the journal in case of article, volume number,
page number, edition, etc.
7. Appendix: In an appendix used to provide a place for material which is not absolutely
essential to the body of the report. The appendix may store the copies of data
ity
collection forms called questionnaires, details of the annual report of the company,
details of graphs/charts, photographs, CDs, interviewers’ instructions. Following are
the items to be placed in this section.
o Data collection forms
o Project related paper cuttings
m
A well formed physical layout is most important, as it may help the report these are:
●● An effective layout for the title page and a clear table of contents.
ity
bold, for headings of chapters; size 14 bold for headings of major sections;
size 12 bold, for headings of subsections, etc.
●● Good quality printing and photocopying. Correct drafts with carefully by spell
check and critical reading for clarity by other team-members.
●● Assign a particular number for each and every figures and tables, provision of
rs
clear titles for tables, and clear headings forcolumns and rows, etc.
●● Accuracy and consistency in quotations and references
ve
Remember that the reader:
●●
●●
●●
Simplify. Keep to the essentials.
ni
Justify. Make no statement that is not based on facts and data.
Quantify when you have the data to do so. Avoid large, small, instead, say
U
50%, one inthree.
●● Be precise and specific in your phrasing of findings.
●● Inform, not impress. Avoid exaggeration.
●● Use short sentences.Use adverbs and adjectives sparingly. Notes
ity
Tables, figures, illustrations, and other such items can be identified with the word
“Table”, “Figure”, or other appropriate descriptor, and include a title and/or caption. The
title or caption must be included within the List of Tables, List of Figures, or other list.
We must use a homogeneous format for titles and captions of tables, figures,
)A
●● lettering in tables and figures should be minimum 2 mm high to make sure that the
information is easy to read
●● tables, figures, illustrations, and other such items must have titles or captions, and
must be numbered
(c
●● titles or captions can go either above or below the table, figure, or illustration
●● headings must be repeated on the second and subsequent pages of tables that
split over two pages or more
ity
●● the format for titles and captions of tables, figures, illustrations, and other such
items must be consistent throughout the thesis.
Numbering
Tables, figures, illustrations and other such items must be numbered consecutively
rs
so as of appearance within the thesis.
There are two methods for numbering Tables, Figures and other items:
ve
Tables in Chapter 1: Table 1.1, 1.2, 1.3…
Whichever method we decide, the numbering style must be the identical for all
numbered items; for example: Table 1.1 and Figure 1.3, or Table 1 and Figure 3, not
Table 1 and Figure 1.3.
Location
ni
There are three acceptable locations for tables and figures:
U
●● within the chapter immediately following first relevancy them
●● grouped at the bottom of the relevant chapter
●● grouped at the bottom of the thesis before the bibliography
Whichever method weselect, we want to be consistent.
ity
If tables and figures are grouped at the bottom of the thesis, then there must
include an entry within the table of contents that directs the reader to their location.
If the caption for a figure, table, etc., won’t fit on the identical page as its
accompanying illustration, place the illustration on a separate page.
m
and readable.
Copyrighted Material
If we remove copyrighted tables, figures, or illustrations from our thesis we need
to insert the subsequent at the spot where the table, figure, or illustration previously
appeared:
(c
ity
lists of figures and tables. Place them along on similar page if they work, as shown in
the illustration below. we can combine the two lists under the heading, “List of Figures
and Tables,” and establish the things as figure or table as is done in the illustration
below.
rs
ve
ni
U
ity
m
Ref: https://openoregon.pressbooks.pub/technicalwriting/chapter/10-5-list-of-
figures-and-tables
)A
These list the documented publications consecutive within the order they’re
referred to in our research report. This will be helpful as it enables us to incorporate
comments and footnotes as well as references.
●● The layout of individual references within the bibliography is the same as that for
Notes the Harvard system.
ity
●● If we discover that our refer to the similar item over once we will use standard
bibliographic abbreviations to save repeating the references in full.
●● The publications said solely embody those we have cited in our report. They must
so be headed ‘References’ rather than ‘Bibliography’ as shown below
rs
Abbreviation Explanation
Op. cit. (opereciato Meaning, in the work cited. This refers to a work previously
referenced and so you must give the author and date and
if necessary the page number, like: Robson (1993) op. cit.
pp. 23-4.
ve
Loc. Cit. (loco ciato Meaning, in the place cited. This refers to the same page of
a work previously referenced. So you must give the author
and date, like: Robson (1993) loc. Cit.
Ibid. (ibidem) Meaning, the same work given immediately before. This
refers to the work referenced immediately before and
5.1.9: Footnotes ni
replaces all details of the previous reference other than a
page number if necessary.
U
Researchers must insert footnotes within the appropriate places. These fulfil two
purposes:
Based on the footnotes’ description, one can easily refer the cross references,
citation of authorities and sources, acknowledgement and elucidation or
explanation of a degree of view. The recent trend is to avoid footnotes. Some
people feel that they enhance display of the scholarship of the researchers. But it’s
neither an end nor a means of displaying scholarship.
When using footnotes, a varieties of references are shown within the research
m
report.
For example: ‘Recent researchindicates that…’ This number refers directly to the
references
)A
(c
Unit-5.2: Presentation
Notes
ity
5.2.1: Oral Presentation
Oral Report
When the researchers are asked to make an oral presentation then they prepare
this type of oral report.It’s more difficult to prepare as compared to the written report.
rs
Because here the presenterdirectlyinteract with the audience. Any delay during an
oralpresentation can leave a negative impression on the audience and also violates the
selfconfidenceof the presenter. Communication plays a big roleIn an oral presentation.
A lot ofplanning and thinking is required to decide ‘What to say’, ‘How to say’, ‘How
ve
much to say’. Also,the presenter should have to face a bunch of questions from the
audience. A lot of preparation isrequired; the broad classification of an oral presentation
is as follows.
(b) What form of presentation could be a root question? Is it brows out from
a manuscript or memorized ordelivered ex-tempo. Memorization is not
recommended, since there could be a slip during presentation.
(c) Secondly, it produces speaker-centric approach. Even reading from the
manuscriptis not recommended, because it becomes monotonous, dull and
m
that theirability to comprehend increases, since they can refer to it when the
presentation isgoing on.
o The presenter should know his target audience well in advance to prepare
Notes tailor-madepresentation.
ity
o The presenter should know the purpose of report such as “Is it for making a
decision”,”Is it for the sake of information”, etc.
Golden rule
rs
Human attention is extremely limited. Don’t cram an excessive amount of
information, either in each slide, or within the whole talk. Avoid details: they will not be
remembered anyway.
ve
Organization
●● Have a really clear introduction, to motivate what we are doing and to present the
matter we want to resolve. The introduction isn’t technical in nature, but strategic.
●● If we have got a companion paper, mention it during the talk and recommend
it for more details. Don’t put all the main points within the talk. Present only the
●●
●●
important ones.
Use only 1 idea per slide.
ni
Have a good conclusions slide: put there the most ideas, those we actually want
U
people to recollect. Use only 1 conclusions/slide.
●● The conclusion slide should be the last one. Don’t put other slides after
conclusions, as this may weaken their impact.
●● Having periodic “talk outline” slides (to show where we are within the talk) helps,
especially for extend talks. A minimum ofone “talk outline” slide is extreamly useful,
ity
thread. The talk should be sequential in nature (i.e. no big conceptual leaps from
one slide to the next).
●● Try to chop out the maximum amount as possible; less is healthier.
)A
●● Help the audience understand where we’re going. Often it’s best to allow them
a high-level overview first, and then plunge into the details; then, while paying
attention to the details they can relate to the high-level picture. This also helps
them save important brain power for later parts of the talk which may be more
important.
Mechanics
(c
●● Humour is extremely useful; prepare one or two couple of puns and jokes
beforehand. However, if we’re not familiar with jokes then better way to avoid them Notes
ity
altogether.
●● The more we rehearse the talk, the better it will be. A rehearsal is most helpful
once carried out loud. 5 rehearsals is a minimum for an important talk.
●● The more people criticize our talk (during practice), the better it will be; pay
attention to criticism, not necessarily to all suggestions, but try to see what and
rs
why people misunderstood your ideas.
●● Everything hasn’t to be written down.
●● Be enthusiastic.
●● Present to explain, ask instant questions, act surprisingly, etc.
ve
●● Give individuals some time to think about the vital facts by slowing down, or by
putting on a mute for a moment.
●● Do not go overtime under any circumstance.
●● Listen to the querries terribly carefully; several speakers answer totally different
●●
questions than the ones asked.
ni
Don’t treat audience as mentally-impaired: don’t explain the completely obvious
things.
U
Text
●● Slides ought to have short titles. A extended title shows one thing is wrong.
●● Use uniform capitalization rules.
●● All the text on one slide ought to have the identicalstructure .
ity
●● Put little or no text on a slide. Don’t put more than one plan in one slide. If we
have lots of text, people will read it faster than we talk, and will not pay attention to
what we say.
●● Don’t use small fonts.
●● Use very few formulas (one per presentation). The identical goes for program code
m
Illustrations
●● Use suggestive graphical illustrations to the maximum amount as attainable. Don’t
shun graphical metaphors. Prefer an image to text.
●● Do not put in the figures details which will not mention explicitly. The figures ought
to be as schematic as attainable.
(c
●● Don’t “waste” info by victimising spare color. Each different color should signify
something different, and something important but don’t use too many different
colors. Have high-contrast colors.
●● For some strange reason, rectangles with shadows appear to seem look far better
Notes than without .
ity
●● Sometimes a matte pastel background looks far better than a white one.
●● Exploit animation with restraint. Don’t use fancy animation effects unnecessarily.
●● However, there’re places where animation is much more effectible, e.g., to depict
the evolution of a complex system, or to introduce related ideas one by one.
rs
●● Use strong colors for important stuff, pastel colors for the unimportant.
●● Employ information cleverly: e.g. make arrow widths showing flows proportional to
the flow capacity.
●● Use thick lines in drawings (e.g. 1 1/2 points or more).
ve
5.2.3: Use of Visual Aids
Visual aids can enhance our presentations - they can increase the audience’s
understanding about topic, explain points, make an impact and create enthusiasm. It’s
become more useful to make data visual
ni
Visual aids are peripherals of a visual manner, like as graphs, photographs, video
clips etcare rather than spoken information. Visual aids shall be selected depending on
their purpose, as an example:
U
Summarise information.
Reduce the amount of spoken words, for example, show a graph of results rather
than reading them out.
ity
Create more of an impact, for example, if our presentation is on the health risks of
smoking,then show images of the effects of smoking on the body rather than describing
this.
Preparation
Think about how can a visual aid support our message and what do you want the
audience to do?
Amity Directorate of Distance & Online Education
Research Methodology 127
●● Ensure that our visual aid follows what we’re saying or this will confuse the
audience. Notes
ity
●● Avoid cluttering of the picture.
●● Visual aids must be clear, concise and of a high quality.
●● Keep the style consistent, such as, the same font, colours, positions etc
●● Use graphs and charts to present data.
rs
●● The audience should not be trying to read and listen at the same time - use visual
aids to highlight points.
●● One message per visual aid, for example, on a slide there should only be one key
point.
ve
●● Employ visual aids in moderation to emphasise and support main points.
●● Ensure that presentation still works without visual aids in case of technical
problems.
As soon as show the visual aid the audience’s attention will be drawn to it so must
U
immediately explain it.
●● Make it clear to the audience why using it.
●● When no need the visual aid ensure that the audience can’t see it.
There are a variety of different types of visual aids like as:
ity
PowerPoint
Microsoft PowerPoint is vastly employed for presentations because it’s too easy
to create. We can insert a range of visual items into the slides which will improve the
audience’s focus. Also, the audience can generally see slideshows better than other
visual aids.
m
Tips:
1. Have a clear and simple background.
)A
6. Consider providing handouts of diagrams because the audience will find the
diagrams easier to read.
7. Avoid putting too much text on a slide.
ity
9. There should only be one key point for each slide.
Whiteboards
Whiteboards are great for providing further explanations, such as, showing the
order of a process, creating diagrams or explaining complex words or phrases. They’re
often used to display headings and write down audience suggestions. Whiteboards are
rs
also ideal for displaying vital data for the whole period of the presentation, such as, key
definitions, as a result of the audience will simply look at the whiteboard for a reminder.
Tips:
ve
●● Ensure that enough time has passed for the audience to take notes before
rubbing something off of the whiteboard.
●● Write concisely to avoid facing away from the audience for too long.
●● Handwriting must be large and legible.
Handouts
ni
Handouts are papers that contain key information from presentation or they may
provide further information that prevent us from overwhelming the audience as there’ll
be less data on the slides and soless data they have to put in while writing down.
U
Tips:
●● If given at the beginning and middle of presentation the audience may be
reading rather than listening.
●● If given at the end of presentation the audience may be trying to take lots
ity
of notes which may reduce the amount of information they are actually
understanding.
To manage this, provide the audience with partially completed handouts so they will
have to listen to what we’re saying to be able to fill in the gaps. Providing the audience
with graphs and charts beforehand is also beneficial because the audience will find
them easier.
m
Video clips
Using videos are a good way to engage the audience and increase their attention.
Use video to bring motion, pictures and audio into presentation.
)A
Tips:
●● Ensure that any videos used are relevant to the presentation’s content.
●● Only show as much of the video as necessary.
●● Never show a really long clip.
(c
Flip chart
Flip charts offer a low cost and low tech solution to record and convey information.
Amity Directorate of Distance & Online Education
Research Methodology 129
They’re more beneficial for smaller audiences and they are favoured for brainstorming
sessions for gathering ideas easily. Flip charts are also widely used for summarising Notes
ity
information and, like with a whiteboard.
Posters
Poster boards can be created using a variety of visual devices, such as graphs and
images. They’re generally quite portable and we can make them as elaborate as we
rs
want. However, they can be expensive to produce if the poster is quite complex.
ve
personal life also. Leaders who know how to communicate effectively with those around
them and will see better productivity and improved relationships in every aspect of their
lives.
Effective communication involves knowing how to listen attentively. It’s the ability
to offer empathy, open-mindedness, and helpful feedback. Also, a friendly demeanour,
confidence, and quality nonverbal communication will also help us, as a manager,
develop good relationships with the members of our team.
ni
Let’s explore some benefits of effective communication you will see in- and outside
your office when you take the time to nurture these skill.
U
1. Building trust
Effective communication fosters trust with others. Our ability to listen attentively
and embrace different points of view helps others keep a trust that we are making
optimal decisions for everyone in the group.
ity
track. They will understand their specific tasks and responsibilities, as well as those of
their teammates, which will help eliminate conflicts and confusion.
5. Increases engagement
With people feeling more confident in their work and in their understanding of what
they need to do, they become more engaged with their work as a whole. According to
Amity Directorate of Distance & Online Education
130 Research Methodology
a recent study exploring the psychology of employee engagement, only 15% of adult
Notes employees are engaged with their employers. By prioritizing effective communication,
ity
we can increase engagement, and thus boost satisfaction, among our team members.
6. Improves productivity
When team members understand their roles, the roles of others and our
expectations, they can focus more on their work and less on workplace issues. With
effective communication, conflicts are resolved quickly, employees can better manage
rs
their workload and distractions are minimized. These benefits contribute to greater
productivity .
ve
With improved communication, team members will be better able to rely on each
other. Build positive feelings and relationships between the team members, which leads
to improved morale and work experiences.
Good communication skills can play an important role in nurturing positive work
experiences for your entire team.
ni
5.2.5 Convention & Strategies of Authentication
Endless description without interpretation is another pitfall. Tables need
conclusions, not detailedpresentation of all numbers or percentages within the cells
U
which readers can see for themselves.
Sometimes qualitative data (e.g., open opinion questions) are just coded and
counted like
ity
5.3.1 :Preparing Research Papers for Journals
Here we described a series of steps to make research paper for journals.Such as
rs
be supportedbyestablished guidelines and should make explicit the estimated
contribution of every individual to the project. We recommend that every research
group establish and make known to its members the criteria for authorship on papers
resulting from the work to be conducted. A list of authors makesure that all individuals
to be involved within the project understand at the outset whether or not they will
ve
expect to be an author and, if so, what their contribution is to be.
2. Start writing before the experiments are complete. We have to Start writing while
we are still doing the experiments. Writing often evokes new ideas: we may realize
that there are additional experiments to run or additional controls that we need to
add. If we wait until completion in the lab, have dismantled the equipment, and
these ideas.
ni
possibly moved on to another position, then we will not have the opportunity to test
5. (Re)examine the list of authors. When we are determined which experiments will
be included in this paper then we must select the authors and the order in which
they will appear. Re-evaluate it based on the contributions that were made to
those experiments and the additional contributions that will be made through the
preparation of the manuscript. If an inventory already exists, make adjustments to
make sure about the compliance with guidelines.
m
6. Determine the basic format. There are three basic formats for peer-reviewed
research articles:
●● Full-length research articles: These articles contain a comprehensive
investigation of the subject matter and are viewed as the standard format. It
)A
“hot” findings, usually during a brief communication format. Articles that have
immediate implications for public health would be appropriate for such a
format, as it might findduring a highly competitive and quickly moving field.
7. Select the journal. There are several factors to think about when choosing a journal.
Notes It’s unlikely that one journal will have all the features we are looking for, so we may
ity
have to compromise. However, there’s one essential feature we should always not
compromise on – manuscripts must be peer reviewed for publication if they’re to be
considered research articles.
Language: If we are interested in communicating our results widely to the
international scientific community, then it is essential to publish in English. If, on the
other hand, we wish to communicate to a more localized community then we can chose
rs
a journal that permits another language.
Focus: What type of research does the journal publish? Is its focus broad or
narrow? Which disciplines are represented? What is the journal’s orientation – for
instance, is it clinical or basic, theoretical or applied?
ve
Indexing: Is the journal indexed in the major electronic databases .
Reputation: Ask colleagues which journals they respect. Look at recent articles
ni
and judge their importance. Check the members of the editorial board and determine
if they are leaders in their fields. Determine the journal’s impact factor. How selective is
that the journal in accepting papers for publication? Try to find out the acceptance rate
of the journal.
U
Format: Do we like the appearance of published articles – the format, typeface,
and style used in citing references? If relevant, does the journal publish short and/or
rapid communications?
Figures: Do figures published within the journal have the resolution that we
simply need? Time to Print: Using the “date submitted” and a “date accepted” that are
ity
published on the article, along with the date of the issue, we can estimate the length of
the review process also because the time from acceptance to publication in print.
Charges: Some journals bill the author for page charges, a price per final printed
page. Most journals have a separate charge for color plates. This may be as much as
$1000 per color plate. Many journals will waive page charges if this presents a financial
hardship for the author; color plate charges are less-readily waived and would at least
m
require evidence that the color is imortant to the presentation of the information.
8. Stock the sections of your paper. As we think about your paper, store relevant
material in folders marked with Introduction, Methods, Results, and Discussion. This
may save time and avoid frustration once the writing begins. Stored items might
)A
10. Outline the paper. An outline is like a road map which define how we will get from
here to there, and helps ensure that we take the most direct and logical route. If
Amity Directorate of Distance & Online Education
Research Methodology 133
there have co-authors, then we may wish to get feedback from them before proceed
to the actual writing phase. Notes
ity
11. Write the first draft. Write the primary draft of the complete manuscript. If we are
writing with co-authors, then may wish to assign different aspects of the manuscript
to different authors. This will save time, enable a lot of people to feel that are making
substantive contributions to the writing method, and make sure that the best use
of expertise. However, it also can result to a mix of design. Thus, if you take this
rs
approach, be certain that the final product is carefully edited to provide a single
voice. “Components of a Research Article” discusses what goes into each section of
the manuscript.
12. Revise the manuscript. This step involves three major tasks, each to be carried out
in the order given:
ve
Make major alterations: Fill in gaps, correct flaws in logic, restructure the
document to present the material in the most logical order.
Polish the style: Refine the text, then correct grammar and spelling.
ni
important to do the tasks in the stated order. Otherwise, we may be spending a lot of
13. Check the references. Make sure that the citations are correct and complete. Do
one last literature search to make certain that we’re up to date.
U
14. Write the final title and abstract. Several changes are created throughout the
editing method. Make certain that our title and abstract match the final version.
15. Reread the journal’s Instructions to Authors. Review the main points of however
the manuscript is to be formatted and submitted.
ity
16. Prepare the final illustrations. Ensure that tables, figures, and figure legends are
complete, clear, self-contained, and in the format required by the journal.
17. Get feedback on manuscript and then revise manuscript again. Getting feedback
is one of the most important things by which we can improve our article. First, be
sure our co-authors have had a chance to read and comment on the draft. Then,
m
once it’s prepared, provide the manuscript to some colleagues. After you get their
comments (on the science, logic, language, and/or style), revise our manuscript to
address their concerns.Do not submit manuscript until it is ready for publication.
Once it is accepted.
)A
18. Submit the manuscript to the editor. Follow the Instructions to Authors to determine
what items we need to submit, how to submit them, and to whom we should send
them. Note that some journals permit (or even require) a “pre-review,”i.e., a letter
indicating the content of the article so that the editors can determine whether they
will accept the manuscript for a full review. At this point we may wish to list possible
reviewers (or individuals to be avoided). If required then contact with the editor to
make sure that the manuscript was received.
(c
19. Deal with reviewers’ comments. Most manuscripts are not accepted on the
first submission and ask forresubmitting a revised manuscript. If we agree withit
then we need to respond to the reviewer comments. Answer every comments of
the reviewers, and highlight where the corresponding changes were made in the
Notes manuscript if happened, indeed, made. We do not need to make all of the changes
ity
that the reviewer recommended, but we do need to provide a convincing rationale
for any changes that we did not make. When we resubmit the manuscript, indicate in
our cover letter that this is a revised version.
20. Check the proofs. Once the manuscript is accepted and ready for print, the
publisher may send the corresponding author page proofs of the article. It should be
rs
accompanied by a list of queries, such as missing information regarding a reference.
The proofs either be sent through e-mail or as hard copy. If there is a chance that
we will be away when the proofs arrive, have a plan for making certain that they are
received and we are notified.
20+. Celebrate! Robert Day says in How to Write and Publish a Scientific Paper (1998),
ve
“The goal of scientific research is publication…. A particular experiment, no matter
how spectacular the results, is not complete until the results are published.” Your
experiment – at least one phase of it – is now complete.
ni
Notice that no correct number of letters or numbers exists; the sole determining
factor is that the number of points we need to create for the specified length of our
paper.
I. Introduction
U
A. Background information connecting the reader to the topic
B. Thesis statement
II. First Main Point
ity
a) First comment
b) Second comment
B. Second sub point
1. Supporting example or detail
2. Supporting example or detail
(c
ity
A. First sub point—Comment on sub point (only one inquiry into this detail)
B. Second sub point
1. Supporting example—only one comment on this example:
2. Supporting example
a) Comment
rs
b) Comment
Continue with the identical sequence alternating numerals and letters
ve
A. Incorporate reactions to the source
B. Include insights about the research topic
C. Synthesize critical thinking threads
V. Conclusion
Affirm that the thesis has been proven
Example:
ni
U
ity
m
)A
(c
ity
The impact factor (IF) may be a calculation of the frequency with that the typical
article in a very journal has been cited in a explicit year. It’s employed to measure the
importance or rank of a journal by calculative the daysits articles are cited.
rs
times articles were cited by the quantity of articles that are citable.
ve
journals throughout 2010.
ni
The impact factor of ten or higher is taken into account as a wonderful score where
3 is flagged nearly as good and also the average score is less than 1. This is often
a rule of thumb. However, the wild card to concentrate to is that impact factor and
examination journals are most effective in the same category.
U
Example:
ity
m
reference may be a citation. A citation count is that the frequency of a writing article
cited by other articles. A citation index could be a kind of bibliographic database, an
index of citations between publications, allowing the user to simply establish which later
documents cite which earlier documents
a. Institute for Scientific Information (ISI): Citation indexing has long been
dominated by the ISI, which is now a part of Thomson Reuters. It publishes the
citation indexes in print and CDforms, which are generally accessed through
the online under the name ‘Web of Science’ (WOS). WOS provides access Notes
ity
to 7 databases: Science Citation Index (SCI), Social Sciences Citation Index
(SSCI), Arts & Humanities Citation Index (A&HCI), Index Chemicus, Current
Chemical Reactions, Conference Proceedings Citation Index: Science and
Conference Proceedings Citation Index: Science and Humanities.
b. Elsevier Which publishes Scopus, is offer online only. Scopus is one among
rs
the biggest abstract and citation database of peer-reviewed literature and web
sources with smart tools to trace.
There are also varity of other readily available citation databases. Few of the
notable ones among these are:
ve
●● CiteSeer is both a citation engine and a digital library. CiteSeer is predicted on
the SmealSearch engine and provides citation, citation graph analysis, related
document retrieval and other searching of scientific literature.
Research Papers in Economics (RePec) provides databases in economics and
otherdisciplinespecific indexes.
●●
ni
Google Scholar (GS) provides citation and search of scholarly literature
across many disciplines and sources. Although limited to the recent articles,
it’s a freely-accessible web programe. Google Scholar is mostly praised for its
speed.
U
●● EBSCOhost is one among of the world’s foremost, mostused, premium,
reference database service.
●● Compendex (Engineering Index, COMPuterizedENgineeringinDEX) is thatthe
computerized version of the Engineering Index and is one in all the foremost
comprehensive engineering literature databases.
ity
Indian Citation Index Indian Citation Index (ICI) database was developed with
specific motives to market knowledge contents, published in Indian journals and bridge
the gap between the content sources and content users via World Wide Web. ICI
database covers about 1000 Indian scholarly journals, encompassing all disciplines of
information including the sciences, social sciences, arts, and humanities’
m
The ICI database also produces other useful by-products like Indian Science
Citation Index (ISCI), Indian Social Science and Humanities Citation Index (ISSHCI),
Indian Journals Citation Reports (IJCR), Indian Science and Technology Abstracts
(ISTA) and Directory of Indian Journals (DOIJ).
)A
International Standard Book Number is assign for monographs or books but the
(c
ISBN is assigned for one or separate book, and ISSN is assigned for a series of
books.
ISBN identifies the particular volume or issue, the ISSN only identifies the series of
Notes the amount or issue.
ity
IISSN is simply optional, which implies that the publisher isn’t legally sure to use it.
Butthe ISBN is obligatory if the book falls under the ISBN application.
Unlike the ISBN, the ISSN are the identical altogether in all the volumes or the
problems of a single series. But the ISBN is different for each volume and issue.
rs
The ISBN could be a 13-digit standard code. It absolutely was Gordon Foster
who created for the beginning a nine-digit ISBN code. Later the world Organization for
Standardization developed the ten-digit standard number code. After 2007, the ISBN
incorporate a 13-digit number standard code.
ve
The International Standard Serial Number is an eight-digit standard number. It’s
the amount given to single set of the series, and once the series changes, another
ISSN code is allotted. It was in 1971 that the ISSN system was first drafted. The
ISSN standard number codes are assigned by the ISSN National Centres and are
coordinated by the ISSN International Centre that’s based in Paris.
Summary:
ni
1. “ISBN” is “International Standard Book Number” and “ISSN” is “International Standard
Serial Number.”
2. One of the person differences between the ISBN and ISSN is that the previous one
U
identifies the publisher whereas the latter one doesn’t identify the publisher.
3. International Standard Book Number is given for monographs or books whereas
International Standard Serial number is given to a series of monographs or books.
4. Inthe case of the ISSN, it’s only optional, which suggest that the publisher isn’t
ity
legally bound to use it. But the ISBN is obligatory if the book falls under the ISBN
application.
Bellow are the example
ity
ISBN 978-951-45-9694-0 (paperback)
rs
ve
ni
U
Planning a Project Proposal
Research Question:
Review of Literature:
m
Colors may just seem simple and unimportant, but they affect our daily lives more
than we may know. If someone is feeling angry, it could just be because they’re angry,
or it could be perhaps that they are surrounded by or looking at the color red. That’s
right! People’s moods can change just because they are looking at different colors!
)A
There are many theories on how just a simple color can change one’s whole mood.
and can make one feel weak. In conclusion, Johnson says that depending on the color,
one’s body can do things (like producing chemicals) that cause a certain emotional
reaction (mad, sad, etc.).
Another idea, by Smith (2007), is that the effect color produces is based on what
Notes one’s body does in response. For example, yellow is mentally stimulating, and activates
ity
memory, whereas red increases confidence. Also, brown can make a person feel
orderly and stable, while a dark blue can make one feel sad. Therefore, Smith says that
different colors do in fact change one’s mood and the consequences can be negative or
positive.
A third writer, Wollard, (2000) seems to think that color can affect one’s mood, but
rs
the effect also can depend on one’s culture and what one’s personal reflection may be.
For example, someone from Japan may not associate red with anger, as people from
the U.S. tend to do. Also, a person who likes the color brown may associate brown with
happiness. However, Wollard does think that colors can make everyone feel the same,
or close to the same, mood. According to Wollard, pink reduces aggression, which
ve
is why the walls of the jail cells in the Seattle prison are pink! Also, brown can make
one feel comforted. Wollard feels that colors do affect one’s mood, but there are other
factors that can alter what one is supposed to feel.
Eric, John, and Paraag’s (2007) main point about color psychology is that color has
both a physiological and psychological effect. For example, green makes people feel
ni
relaxed because it relaxes their muscles and makes them breathe deeper and more
slowly. Furthermore, blue lowers blood pressure, which makes one feel calm. Eric,
John, and Paraag conclude that color affects one’s mood because of what it does to the
body.
U
Yet another idea, by Airey (2006), is that color is energy, and it can have a physical,
mental, spiritual, and/or emotional affect on people. He states that black can make one
feel sophisticated and secure, but it can also make one feel depressed. Also brown can
make a person feel reliable and serious, while yellow lifts self-esteem. Therefore, Airey
concludes that different colors can have different kinds of affects on people.
ity
Analysis
There are three ideas about color psychology in these sources, and they all say
that color affects one’s mood. They differ based on what factors influence the effects of
color, such as culture, opinion, and what goes on inside one’s body. One of the three
ideas is that color affects mood based on one’s personal opinions. For example, if a
m
person dislikes the color pink, he may associate pink with hate. Another idea states that
color affects mood based on one’s culture. For example, someone from the U.S. may
think of the color green when referring to envy, while people in Japan think of yellow in
connection with wanting what someone else has. However, the majority of the sources
)A
consulted say that color affects mood by influencing what goes on inside of people. For
example, seeing the color blue releases calming chemicals, which in turn makes one
calm. Also, because yellow is the hardest color for the eye to focus on, people may
become irritated when looking at yellow, and it is a proven fact that babies cry most in
yellow nurseries. These theories do not seem to have much in common.
Methodology
(c
In order to test ideas about how color affects one’s mood (color psychology) this
researcher will test sixty middle school students (thirty girls and thirty boys) on how the
colors green, blue, black, yellow, red, pink, and brown make them feel. The students will
mark the feeling that each color makes them feel, according to the choices shown on
the example survey below. Notes
ity
Example Survey
(The only thing that changes in each different survey is the color being tested and
the gender of the student responding):
Choose as many answers as fit your feeling for the color. Does the color make you
rs
feel:
Mad/Angry?
Sad/Depressed/Cold?
ve
Secure/Safe?
Irritated/Annoyed?
Calm?
Relaxed?
Fun/Excited?
Happy/Cheery?
Tired/Sleepy/Drowsy?
ni
U
Bored?
Losing Energy/Weak?
Neutral/Balanced?
ity
Love/Affection?
Dislike/Hate?
Survey Results:
Most female middle school students associated these feelings with these colors…
m
B a l a n c e d Happy/Cheery H a p p y / B o r e d Mad/Angry C a l m
(27%) (53%) Cheery (27%) (40%) ( 3 0 % ) (23%)
and Love/ and Love/
Affection Affection
(27%) (30%)
So, the greatest number of female middle school students thought that black made
(c
them feel sad, cold, and depressed; and green made most of them feel neutral and
balanced, and so on. The numbers in parentheses by the feelings show the percentage
of girls that felt that way. The reason most of these percentages are so low is that there
were many different responses, and for most feeling categories there were only one or
Notes two girls who felt that way.
ity
Most male middle school students associated these feelings with these colors…
rs
(37%) (17%) (33%)
So, the greatest number of male middle school students thought that black made
them feel sad, cold, and depressed; and green made most of them feel secure and
safe, and so on. The numbers in parentheses by the feeling show the percentage of
ve
boys that felt that way. The reason these percentages are so low is that there were
many different responses, and for most feeling categories there were only one or two
boys who felt that way.
ni
U
ity
m
)A
(c
ity
Unit Outcomes:
At the end of this unit, participants will be able to:
rs
●● Identify the salient features of a research project
ve
and properly defining a research problem. A researcher must find the problem
and formulate it so that it becomes susceptible to research. Like a medical doctor, a
researcher must examine all the symptoms (presented to him or observed by him)
concerning a problem before he can diagnose correctly. To define a problem correctly, a
researcher must know: what a problem is?
ni
Identifying a research problem, in general, refers to some difficulty in the context of
either a theoretical or practical situation and wants to obtain a solution for the same.
Defining a research problem is the fuel that drives the scientific process and is the
foundation of any research method and experimental design.
●● There must be an individual or group which has some difficulty or the problem.
●● There must be some objective(s) to be attained at.
●● There must be an alternative means for obtaining the objectives one wishes to
)A
attain.
●● There must remain some doubt in the mind of a researcher with regard to the
selection of alternatives. This means that research must answer the question
concerning the relative efficiency of the possible alternatives.
●● There must be some environment to which the difficulty pertains.
●● The research problem undertaken for study must be carefully selected.
(c
●● This difficulty in selecting a problem or the gap arise not due to limited range
Notes of reachable problems but he inability to locate the socially relevant topic
ity
without duplicating earlier studies.
Problem Aspects:
Let us start with the question: What does one mean when he/she wants to define a
rs
research problem? The answer may be that one wants to state the problem along with
the bounds within which it is to be studied. In other words, defining a problem involves
the task of laying down boundaries within which a researcher shall study the problem
with a pre-determined objective in view.
ve
How to define a research problem is undoubtedly a herculean task. However, it is a
task that must be tackled intelligently to avoid the perplexity encountered in a research
operation. The usual approach is that the researcher should himself pose a question
(or in case someone else wants the researcher to carry on research, the concerned
individual, organization or an authority should pose the question to the researcher)
ni
and set-up techniques and procedures for throwing light on the question concerned for
formulating or defining the research problem. But such an approach generally does not
produce definitive results because the question phrased in such a fashion is usually in
broad general terms and as such may not be in a form suitable for testing.
U
Defining a research problem properly and clearly is a crucial part of a research
study and must in no case be accomplished hurriedly. However, in practice this
frequently overlooked which causes a lot of problems later on. Hence, the research
problem should be defined in a systematic manner, giving due weightage to all relating
points. There are a lot of aspects which needs to kept in mind before we deal with it.
ity
(i) Statementof the problem: Problems should be stated in either practical or scientific
or some intellectual interest. It is can be stated through a preliminary field survey or
observation. This helps in stating the problem clearly; it also helps in narrowing the
concern in a phrase the problem in operational terms. In case the problem is stated
in a broad and general way it will involves a lot of ambiguities which can be resolved
m
by a looking at the detailed observations and surveys available. Thus the feasibility
of a particular solution can be considered.
(ii) Understanding the nature of the problem: Another aspect to be kept in mind is to
understand the origin of the problem. The best way to understand is to find out the
)A
what materials are available at the moment and what other materials are required.
“Knowing what data are available often serves to narrow the problem itself as well
as the technique that might be used.”This would also help a researcher to know if
there are certain gaps in the theories, or whether the existing theories applicable to
Amity Directorate of Distance & Online Education
Research Methodology 145
the problem under study are inconsistent with each other, or whether the findings
of the different studies do not follow a pattern consistent with the theoretical Notes
ity
expectations and so on.All this will enable a researcher to take new strides in the
field for furtherance of knowledge i.e., moving up starting from the existing premise.
Studies on related problems are useful for indicating the type of difficulties that may
be encountered in the present study as also the possible analytical shortcomings. At
times such studies may also suggest useful and even new lines of approach to the
present problem.
rs
(iv) Deduction from the theory: This related to deductions made from the available
source. Once the irrelevant details of the problems are kept aside and the core of the
problem is identified then we can carry out a systematic investigation to the research
to assess the data.
ve
(v) Rephrasing the research problem into a working proposition: Rephrasing is important
because the nature of the problem has been clearly understood, defined. Through
rephrasing, the researcher puts the research problem in as specific terms as possible
so that it may become operationally viable and may help in the development of the
research.
Considerations
ni
The research problem undertaken for study must be carefully selected. The task is
a difficult one, although it may not appear to be so. Help may be taken from a research
guide in this connection. A research guide can thus help in considering a problem.
U
Considering a problem can be defined in three categories:
1. Identify concepts and terms that make up the topic statement: The research
problem can have many aspects which needs to be kept in mind but identifying
the concept and terms which comprises of the topic makes its easier for the
ity
3. Additional sources: since the research work can be extensive it ban be helped
with additional data which can broaden the knowledge and strengthen the initial
thoughts and arguments related to the research.
)A
problems.
ity
This step focuses on developing a research strategy or overall approach to
resolving the identified issue or problem. A research plan or approach is a blueprint
or framework for carrying out a marketing research project. Its purpose is to design
a study that will test the hypotheses of interest, determine possible answers to the
research questions, and provide the information required for decision making.
rs
●● Analyzing secondary data
●● Qualitative analysis
●● Quantitative data collection methods (survey, observation, and
experimentation)
ve
●● Definition of the data required
●● Procedures for measurement and scaling
●● Designing a questionnaire
●● Process of sampling and sample size
●● Data analysis strategy
Collecting Data
ni
This step was all about gathering the data you’ll need to solve the issue or problem
U
you’ve identified. A field force or staff collects data in the field (in-home, mall intercept,
or computer-assisted personal interviewing), over the phone (telephone or computer-
assisted telephone interviewing), or through the mail (traditional mail and mail panel
surveys with recruited households).
Begin by putting your findings and the information you gathered in Step 3 into
order. After that, write a rough draught of your conclusion, recommendations, and
findings. The rough draught will assist you in organising your thoughts. The last step
is to refine the draught into your final research conclusion. Before the final product is
ready for Step 5, you will most likely revise the draught as many times as necessary.
(c
understandable format so that they can be used in the decision-making process quickly.
In addition, to improve clarity and impact, an oral presentation to management should Notes
ity
be made using tables, figures, and graphs.
●● Formal Paper
●● Published Article
rs
●● PowerPoint Presentation
●● Audio or Video
●● Spreadsheet
ve
1. The research report will differ based on the …………of the particular managers using
the
report.
a) need
b) position
c) designation
d) none of the above
ni
U
2. Accuracy refers to the degree to which information reflects……………..
a) reality
b) light
c) unreality
ity
b) trainees
c) other researchers
d) none of the above
4. …………….refers to the time span between completion of the research project and
)A
b) final report
Notes
ity
c) formal report
d) none of the above
6. Writing of report is the ………..step in a research study and requires a set of skills
somewhat different from those called for in respect of the former stages of research.
a) final
rs
b) semifinal
c) primary
d) none of the above
ve
7. ………………means bringing out the meaning of data.
a) Interpretation
b) translation
c) transformation
ni
d) none of the above
8. Successful interpretation depends on how well the data is……………...
a) analysed
U
b) collected
c) interpreted
d) none of the above
9. In the ………………method, one starts from observed data and then generalisation
ity
is done
a) induction
b) conduction
c) coronation
d) invention
m
c) visual effects
d) none of the above
11. ………….report presents the outcome of the research in detail.
a) Long
b) short
(c
c) medium
d) none of the above
12. The …………….statement should explain the nature of the project, how it came
about and Notes
ity
what was attempted.
a) opening
b) Closing
c) Starting
rs
d) ending
13. The ………………..should indicate the various parts or sections of the report.
a) table of contents
ve
b) chair of contents
c) stool of contents
d) none of the above
14. …………..Page should indicate the topic on which the report is prepared.
ni
a) Title
b) introduction
c) conclusion
U
d) none of the above
15. A selected bibliography lists the items which the author thinks are of ………….
interest to
the reader.
ity
a) primary
b) secondary
c) no
d) none of the above
16. In a report there must be …………….in margins and spacing.
m
a) Consistency
b) inconsistent
c) aligned
)A
Answers:
Notes
ity
1. need
2. reality
3. decision maker
4. Currency
5. Research report
rs
6. final
7. Interpretation
8. analysed
ve
9. induction
10. communication
11. Long
12. opening
13. table of contents
14. Title
15. primary
ni
U
16. Consistency
17. systematic
ity
m
)A
(c