Module 1 RMIPR

Introduction: What Is Research?

Research refers to a careful, well-defined (or redefined), objective, and systematic
method of search for knowledge, or formulation of a theory that is driven by inquis-
itiveness for that which is unknown and useful on a particular aspect so as to make
an original contribution to expand the existing knowledge base. Research involves
formulation of hypothesis or proposition of solutions, data analysis, and deductions;
and ascertaining whether the conclusions fit the hypothesis. Research is a process of
creating, or formulating knowledge that does not yet exist.
Booth et al. [1] explains that the research cycle starts with basically a practical
problem: one must be clear what the problem being attempted to solve is and why
it is important. This problem motivates a research question without which one can
tend to get lost in a giant swamp of information. The question helps one zero in onto
manageable volume of information, and in turn defines a research project which is
an activity or set of activities that ultimately leads to result or answer, which in turn
helps to solve the practical problem that one started with in the first place as shown
in Fig. 1.1.
The building up of background for doing research includes one to acquire the
ability to connect different areas. The purpose is to prepare the mind for active work
as opposed to becoming a repository or an encyclopedia. Research is not just about
reading a lot of books and finding a lot of, gathering a lot of existing information. It
is instead adding, maybe small and specific, yet original, contribution to that
existing body of knowledge. So, research is about how one poses a question which
has relevance to the world that we are living in, and while looking for that answer
one has to be as systematic as one can be. There must be a balance between what is
achievable in a research program with a finite endpoint and also, the contribution it
is going to make. The objective of a good research program is to try and gain insight
into something. Or indeed, to try and solve a problem. Good research questions
develop throughout the project actually and one can even keep modifying them.
Through research, one would like to make, or develop, new knowledge about the

1
Fig. 1.1 The research flow diagram [1]

World around us which can be written down or recorded in some way, and that
knowledge can be accessed through that writing or recording.
The ways of developing and accessing knowledge come in three, somewhat over-
lapping, broad categories:

(i) Observation is the most fundamental way of obtaining information from a

source, and it could be significant in itself if the thing that we are trying to
observe is really strange or exciting, or is difficult to observe. Observation
takes different forms from something like measurements in a laboratory to a
survey among a group of subjects to the time it takes for a firmware routine to
run. The observational data often needs to be processed in some form and this
leads to the second category of knowledge, the model.
(ii) Models are approximated, often simplified ways of describing sometimes very
complex interactions in the form of a statistical relationship, a figure, or a set
of mathematical equations. For instance, the modeling equation captures the
relationship between different attributes or the behavior of the device in an
abstract form and enables us to understand the observed phenomena.
(iii) The final category is a way of arranging or doing things through processes,
algorithms, procedures, arrangements, or reference designs, to get a certain
desired result.

2
The categories of knowledge as enumerated above are shown in Fig. 1.2. Good
research involves systematic collection and analysis of information and is followed
by an attempt to infer a little bit beyond the already known information in a way that
is a significant value addition. Usually, engineering research is a journey that traverses
from a research area (example: Control Systems), to the topic (example: Control of
Microbial Fuel Cells) and finally onto the problem (example: Adaptive Control of
Single Chamber Microbial Fuel Cells) (Area → Topic → Problem). Getting a good
problem to solve is more than half the work done. However, sometimes the journey can
be reverse, for example, the traversal from (Problem → Topic → Area). This can
happen when one is led to a problem through a connection to another problem whose
top structure is different.

Fig. 1.2 The categories of knowledge in research

Engineering research is the process of developing the perspectives and seeking
improvements in knowledge and skills to enable the recognition, planning, design,
and execution of research in a wide range of forms relevant for engineering and
technology investigations and developments. We can start off by describing some
problem in the world that exists that is bugging or worrying us and that we should
be addressing. It could be that there is something we would like to do or accomplish
but currently cannot because we lack the knowledge to do so. It could be that there
is something that already works, but we do not know why and we would like to
3
understand it better. It could be that we want to do something to see what will
happen.

1.1 Objectives of Engineering Research

The objective of engineering research is to solve new and important problems, and
since the conclusion at the end of one’s research outcome has to be new, but when
one starts, the conclusion is unknown. So, the start itself is tricky, one may say. The
answer is, based on “circumstantial evidence”, intuition, and imagination, one
guesses what may be a possible conclusion. A guess gives a target to work toward,
and after initial attempts, it may turn out that the guess is incorrect. But, the work may
suggest new worthy avenues or targets which may be based on some modifications
of the initial target, or may need new techniques, or one may obtain negative results
which may render the initial target or some other targets as not realizable, or may
lead to fortunate discoveries while looking for something else (serendipity). Research
objectives can sometimes be convoluted and difficult to follow.
Knowing where and how to find different types of information helps one solve
engineering problems, in both academic and professional career. Lack of investiga-
tion into engineering guidelines, standards, and best practices result in failures with
severe repercussions. As an engineer, the ability to conduct thorough and accurate
research while clearly communicating the results is extremely important in decision-
making.
The main aim of the research is to apply scientific approaches to seek answers to
open questions, and although each research study is particularly suited for a certain
approach, in general, the following are different types of research studies: exploratory
or formulative, descriptive, diagnostic, and hypothesis-testing.
The objectives of engineering research should be to develop new theoretical or
applied knowledge and not necessarily limited to obtaining abilities to obtain the
desired result. The objectives should be framed such that in the event of not being able
to achieve the desired result that is being sought, one can fall back to understanding
why it is not possible, because that is also a contribution toward ongoing research
in solving that problem. Of course, someone else might come along and actually
propose a different approach where the desired objective is indeed possible to be
achieved.

4
1.2 Motivation in Engineering Research

The possible motives may be the result of one or more of the following desires:

(i) Studies have shown that intrinsic motivations like interest, challenge, learning,
meaning, purpose, are linked to strong creative performance;
(ii) Extrinsic motivating factors like rewards for good work include money, fame,
awards, praise, and status are very strong motivators, but may block creativity.
For example: Research outcome may enable obtaining a patent which is a good
way to become rich and famous.
(iii) Influences from others like competition, collaboration, commitment, and
encouragement are also motivating factors in research. For example: my friends
are all doing research and so should I, or, a person that I dislike is doing well
and I want to do better.
(iv) Personal motivation in solving unsolved problems, intellectual joy, service to
community, and respectability are all driving factors.

The following factors would be a mix of extrinsic and intrinsic aspects: (i) Wanting
to do better than what has been achieved in the world, (ii) improve the state of the
art in technology, (iii) Contribute to the improvement of society, (iv) Fulfillment of
the historical legacy in the immediate sociocultural context.
Several other factors like government directives, funding opportunities in certain
areas, and terms of employment, can motivate people to get involved in engineering
research.
1.3 Types of Engineering Research

The different types of research are

(i) Descriptive versus Analytical: Descriptive research includes comparative

and correlational methods, and fact-finding inquiries, to effectively describe
the present state of art. The researcher holds no control over the variables;
rather only reports as it is. Descriptive research also includes attempts to
determine causes even though the variables cannot be controlled. On the
contrary, in analytical research, already available facts for analysis and critical

5
 evaluation are utilized. Some research studies can be both descriptive and
analytical [3].
(ii) Applied versus Fundamental: Research can either be applied research or
fundamental (basic or pure) research. Applied research seeks to solve an
immediate problem facing the organization, whereas fundamental research is
concerned with generalizations and formulation of a theory. Research
concerning natural phenomena or relating to pure mathematics are examples
of fundamental research. Research to identify social or economic trends, or
those that find out whether certain communications will be read and understood
are examples of applied research. The primary objective of applied research is
to determine a solution for compelling problems in actual practice, while basic
research is aimed at seeking information which could have a broad base of
applications in the medium to long term.
(iii) Quantitative versus Qualitative: Quantitative research uses statistical observa-
tions of a sufficiently large number of representative cases to draw any con-
clusions, while qualitative researchers rely on a few non representative cases or
verbal narrative in behavioral studies such as clustering effect in intersections
in Transportation engineering to make a proposition.

1.4 Finding and Solving a Worthwhile Problem

A researcher may start out with the research problems stated by the Supervisor or
posed by others that are yet to be solved. Alternately, it may involve rethinking of a
basic theory, or need to be formulated or put together from the information provided
in a group of papers suggested by the Supervisor. Research scholars are faced with the
task of finding an appropriate problem on which to begin their research. Skills needed
to accomplish such a task at the outset, while taking care of possible implications are
critically important but often not taught [4]. Once the problem is vaguely identified,
the process of literature survey and technical reading, as described in the next chapter,
would take place for more certainty of the worthiness of the intended problem.
However, an initial spark is ideally required before the process of literature survey
may duly begin. Sometimes, an oral presentation by somebody which is followed
by asking questions or introspection provides this perspective which reading papers
do not. At other times, a development in another subject may have produced a tool

6
or a result which has direct implications to the researcher’s subject and may lead to
problem identification.
A worthwhile research problem would have one or more attributes. It could be
no intuitive/counterintuitive even to someone who knows the area, something that
the research community had been expecting for some time, a major simplification of
a central part of the theory, a new result which would start off a new subject or an
area, provides a new method or improves upon known methods of doing something
which has practical applications, or a result which stops further work in an area. The
researcher has to be convinced that the problem is worthwhile before beginning to
tackle it because best efforts come when the work is worth doing, and the problem
and/or solution has a better chance of being accepted by the research community.
Not all problems that one solves will be great, and sometimes major advancements
are made through solutions to small problems dealt with effectively.
Some problems are universally considered hard and open, and have deep impli-
cations and connections to different concepts. The reality is that most researchers in
their lifetime do not get into such problems. However, hard problems get solved only
because people tackle them. The question a researcher has to grapple with whether
the time investment is worth it given that the likely outcome is negative, and so it is
a difficult personal decision to make. At the same time, even in the case of failure
to solve the intended hard problem, there may be partial/side results that serve the
immediate need of producing some results for the dissertation.
George Pólya (1887–1985) suggested a 4-step procedure for mathematical
problem-solving [5], which is relevant to engineering researchers as well. Recent
work such as [6, 7] suggest the relevance of these recommendations. The recom-
mended steps to solve a research problem are

(i) Understand the problem, restate it as if its your own, visualize the problem by
drawing figures, and determine if something more is needed.

(ii) One must start somewhere and systematically explore possible strategies to
solve the problem or a simpler version of it while looking for patterns.
(iii) Execute the plan to see if it works, and if it does not then start over with another
approach. Having delved into the problem and returned to it multiple times, one
might have a flash of insight or a new idea to solve the problem.
(iv) Looking back and reflecting helps in understanding and assimilating the strat-
egy, and is a sort of investment into the future.
7
 In the subsequent chapters of this book, we present other different aspects of
research which together form different parts of research methodologies and are
important for a successful engineering research career.

Ethics in Engineering Research

Ethics generally refers to a set of rules distinguishing acceptable and unacceptable
conduct, distinguishing right from wrong, or wise aphorisms like the sayings of
Chanakya.1 Most people learn such norms in their formative years [1], but moral
development continues through different stages of growth. Although everyone rec-
ognizes some common ethical norms, but there is difference in interpretation and
application. Ethical principles can be used for evaluation, proposition or interpreta-
tion of laws [2]. Although ethics are not laws, but laws often follow ethics because
ethics are our shared values.
International norms for the ethical conduct of research have been there since the
adoption of the Nuremberg Code in 1947 [3]. According to Whitbeck [4], the issues
related to research credit dates back to the establishment of the British Royal Society
(BRS) in the seventeenth century to refine the methods and practices of modern
science [4]. This event altered the timing and credit issues on the release of research
results since BRS gave priority to whoever first submitted findings for publication,
rather than trying to find out who had first discovered.
Whitbeck [4] raised two simple but significant questions to address the tricky
issue of authorship in research: (1) who should be included as an author and (2) the
appropriate order of listing of authors. In an increasingly interconnected world, the
issue of coauthorship is very relevant to all researchers. There are issues around
individuals who may be deeply involved during the conduct of the research work, but
may not contribute in the drafting phase. Additionally, certain universities now put
restrictions on coauthorship to prevent malpractices which will be described later in
this chapter.

Government bodies, and universities worldwide have adopted certain codes for
research ethics. Research ethics and the responsible conduct of research are often
erroneously used interchangeably. Research ethics examines the appropriate appli-
cation of research outcomes, while responsible conduct of research deals with the

8
way the work is undertaken. In this chapter, let us take a look at specific challenges
posed by the application of ethics in engineering research.

5.1 Ethics in Engineering Research Practice

Technological developments raise a whole range of ethical concerns such as privacy

issues and data related to surveillance systems, and so engineering researchers need
to make ethical decisions and are answerable for the repercussions borne out of their
research as outcomes. The reason that ethics matter in data used in engineering
research is usually because there is impact on humans. Certain practices may be
acceptable to certain people in certain situations, and the reasons for unacceptability
may be perfectly valid. We have unprecedented access to data today, and unprece-
dented options for analysis of these data and consequences in engineering research
related to such data. Are there things that are possible to do with this data, that we
agree we should not do? Engineering ethics gives us the rule book; tells us, how to
decide what is okay to do and what is not.
Engineering research is not work in isolation to the technological development
taking place. Researchers make many choices that matter from an ethical perspective
and influence the effects of technology in many different ways:

(i) By setting the ethically right requirements at the very outset, engineering
researchers can ultimately influence the effects of the developed technology.
(ii) Influence may also be applied by researchers through design (a process that
translates the requirements into a blueprint to fulfill those requirements). During
the design process, decision is to be made about the priority in importance of
the requirements taking ethical aspects into consideration.
(iii) Thirdly, engineering researchers have to choose between different alternatives
fulfilling similar functions.

Research outcomes often have unintended and undesirable side effects. It is a vital
ethical responsibility of researchers to ensure that hazards/risks associated with the
technologies that they develop, are minimized and alternative safer mechanisms are
considered. If possible, the designs should be made inherently safe such that they
avoid dangers, or come with safety factors, and multiple independent safety barriers,
or if possible a supervisory mechanism to take control if the primary process fails.

9
Engineering research should be conducted to improve the state-of-the-art of tech-
nologies. Research integrity encompasses dealing fairly with others, honesty about
the methods and results, replicating the results wherever possible so as to avoid
errors, protecting the welfare of research subjects, ensuring laboratory safety, and so
forth. In order to prevent mistakes, peer reviews should take place before the research
output is published.
There may be different types of research misconduct as described in research
articles like [5] and [6], which can be summarized as follows:

(i) Fabrication (Illegitimate creation of data): Fabrication is the act of conjuring

data or experiments with a belief of knowledge about what the conclusion of
the analysis or experiments would be, but cannot wait for the results possibly
due to timeline pressures from supervisor or customers.
(ii) Falsification (Inappropriate alteration of data): Falsification is the misrepre-
sentation or misinterpretation, or illegitimate alteration of data or experiments,
even if partly, to support a desired hypothesis even when the actual data received
from experiments suggest otherwise.
Falsification and fabrication of data and results, hamper engineering research,
cause false empirical data to percolate in the literature, wreck trustworthiness
of individuals involved, incur additional costs, impede research progress, and
cause actual and avoidable delays in technical advancement. Misleading data
can also crop up due to poor design of experiments or incorrect measurement
practices.
Fabrication and falsification of data in published content can hurt honest
researchers getting their work published because what they can churn out
may short fall of what is already published through misconduct till the
misconduct is established and subsequently retracted.

The image of engineering researchers as objective truth seekers is often

jeopardy- dized by the discovery of data related frauds. Such misconduct can be
thwarted by researchers by always trying to reproduce the results
independently when- ever they are interested to do further work in a published
material which is likely to be part of their literature survey.

(iii) Plagiarism (Taking other’s work sans attribution): Plagiarism takes place when
someone uses or reuses the work (including portions) of others (text, data,
tables, figures, illustrations or concepts) as if it were his/her own without explicit
10
 acknowledgement. Verbatim copying or reusing one’s own published work is
termed as self-plagiarism and is also an unacceptable practice in scientific lit-
erature. The increasing availability of scientific content on the internet seems
to encourage plagiarism in certain cases, but also enables detection of such
practices through automated software packages.

How are supervisors, reviewers or editors alerted to plagiarism?

(i) Original author comes to know and informs everyone concerned.

(ii) Sometimes a reviewer finds out about it during the review process.
(iii) Or, readers who come across the article or book, while doing research.

Although there are many free tools and also paid tools available that one can
procure institutional license of, one cannot conclusively identify plagiarism,
but can only get a similarity score which is a metric that provides a score of the
amount of similarity between already published content and the unpublished
content under scrutiny.
However, a low similarity score does not guarantee that the document is pla-
giarism free. It takes a human eye to ascertain whether the content has been
plagiarized or not. It is important to see the individual scores of the sources,
not just the overall similarity index. Setting a standard of a maximum allowable
similarity index is inadequate usage of the tool. Patchwork plagiarism is more
difficult to evaluate.
There are simple and ethical ways to avoid a high similarity count on an about
to be submitted manuscript. Sometimes, certain published content is perfect for
one’s research paper, perhaps in making a connection or fortifying the argu-
ment presented. The published material is available for the purpose of being
used fairly. One is not expected to churn out research outcomes in thin air.
However, whatever is relevant can be reported by paraphrasing in one’s own
words, that is, without verbatim copy. One can also summarize the relevant
content and naturally, the summary invariably would use one’s own words. In
all these cases, citing the original source is important. However, merely because
one has cited a source, it does not mean that one can copy sentences (or para-
graphs) of the original content verbatim. A researcher should practise writing
in such a way that the reader can recognize the difference between the ideas or
results of the authors and those that are from other sources. Such a prac- tice
enables one to judge whether one is disproportionately using or relying on

11
 content from existing literature.
(iv) Other Aspects of Research Misconduct: Serious deviations from accepted con-
duct could be construed as research misconduct. When there is both deception
and damage, a fraud is deemed to have taken place. Sooner or later ethical
violations get exposed. Simultaneous submission of the same article to two
different journals also violates publication policies. Another issue is that when
mistakes are found in an article or any published content, they are generally
not reported for public access unless a researcher is driven enough to build on
that mistake and provide a correct version of the same which is not always the
primary objective of the researcher.

5.2 Ethical Issues Related to Authorship

Academic authorship involves communicating scholarly work, establishing priority

for their discoveries, and building peer-reputation, and comes with intrinsic burden
of acceptance of the responsibility for the contents of the work. It is the primary basis
of evaluation for employment, promotion, and other honors.

There are several important research conduct and ethics related issues connected
to authorship of research papers as described by Newman and Jones [7], and are
summarized herewith in the context of engineering research.

Credit for research contributions is attributed in three major ways in research

publications: by authorship (of the intended publication), citation (of previously
published or formally presented work), and through a written acknowledgment (of
some inputs to the present research). Authorship establishes both accountability and
gives due credit. A person is expected to be listed as an author only when associated
as a significant contributor in research design, data interpretation, or writing of the
paper.
Including “guest” or “gift” (coauthorship bestowed on someone with little or no
contribution to the work) authors dilutes the contribution of those who actually did
the work, inappropriately inflates credentials of the listed authors [8], and is
ethically a red flag highlighting research misconduct [9]. Sometimes, the primary
author dubiously bestows coauthorship on a junior faculty or a student to boost their
chances of employment or promotion, which can be termed as Career-boost
authorship [10].

12
 There is also an unfortunate malpractice of coauthorship that can be described as
“Career-preservation authorship” wherein a head of the department, a dean, a
provost, or other administrators are added as Coauthors because of quid pro quo
arrangement wherein the principal author benefits from a “good relation” with the
superiors and the administrator benefits from authorship without doing the required
work for it [11].
Sometimes, an actual contributor abstains from the list of authors due to non-
disclosed conflict of interest within the organization [10]. Such coauthor ships can be
termed as ghost coauthor ship. Full disclosure of all those involved in the research is
important so that evaluation can happen both on the basis of findings, and also whether
there was influence from the conflicts. In another type of questionable authorship,
some researchers list one another as coauthors as a reciprocal gesture with no real
collaboration except minimal reading and editing, without truly reviewing the work
threadbare.

Some authors, in trying to acquire a sole-authored work, despite relying on signif-

icant contribution to the research work from others, recognize that effort only by an
acknowledgment, thereby misrepresenting the contributions of the listed authors. The
unrecognized “author” is as a consequence, unavailable to readers for elaboration.

All listed authors have the full obligation of all contents of a research article, and so
naturally, they should also be made aware of a journal submission by the
corresponding author. It is imperative that their consent is sought with respect to the
content and that they be agreeable to the submission. In case of misconduct like
inappropriate authorship, while the perpetrator is easier to find, the degree of
appropriate accountability of the coauthors is not always obvious. Being able to
quantify the contributions so as to appropriately recognize and ascertain the degree of
associated accountability of each coauthor, is appealing.

Double submission is an important ethical issue related to authorship, which

involves submission of a paper to two forums simultaneously. The motivation is to
increase publication possibility and possibly decrease time to publication. Reputed
journals want to publish original papers, i.e., papers which have not appeared else-
where, and strongly discourage double submission.

13
14