Skip to content

Research Ethics, Publication and Scientific Communication

CSIR NET Part A research aptitude: misconduct and plagiarism, consent and animal ethics, authorship, peer review, citation metrics and patents.

  • 20 min read
  • Updated
On this page · 16 sections
  1. Prerequisites
  2. The Hook: The Same Panel, Twice
  3. Research Integrity
  4. Ethics of Research with Humans and Animals
  5. Data, Records and Openness
  6. Authorship and Credit
  7. Publishing a Paper
  8. Measuring Research: Citation Metrics
  9. Intellectual Property
  10. Memory Tricks
  11. Don’t Confuse
  12. Common Mistakes
  13. Practice
  14. Quick Revision Table
  15. Key Takeaways
  16. What’s Next

Prerequisites

The Hook: The Same Panel, Twice

Research Integrity

Research integrity means doing, recording and reporting research honestly. Its core values, as most codes of conduct phrase them: honesty, accuracy, objectivity, openness (transparency), accountability, fairness to colleagues and respect for research subjects.

Research misconduct: FFP

The widely used definition of research misconduct has three parts, known as FFP:

MisconductDefinitionExample
FabricationMaking up data or results and recording or reporting themReporting a third replicate that was never run
FalsificationManipulating materials, equipment or processes, or changing or omitting data, so that the record does not accurately represent the researchDeleting outlying points without a stated, pre-planned rule; splicing gel lanes without saying so; selectively enhancing part of an image
PlagiarismUsing another person’s ideas, processes, results or words without giving appropriate creditCopying a paragraph from a paper into your thesis without quotation and citation

Honest error and honest differences of scientific opinion are not misconduct. Intent, or reckless disregard, is what separates falsification from a mistake.

Questionable research practices

These fall short of FFP but distort the record, and Part A options often describe them without naming them.

PracticeWhat it is
p-hacking (data dredging)Trying many analyses, subsets or outcome measures and reporting only those with p < 0.05
HARKingHypothesising After the Results are Known: presenting a post-hoc hypothesis as if it had been stated in advance
Selective (outcome) reportingReporting only the favourable outcomes of a study
Salami slicingSplitting one study into several minimal papers to increase the publication count
Duplicate (redundant) publicationPublishing substantially the same results twice without disclosure
Inappropriate image manipulationAdjusting part of an image rather than the whole, or combining images without marking it
Gift, guest or honorary authorshipListing someone who did not contribute enough to be an author
Ghost authorshipLeaving out someone (often a paid writer) who made a substantial contribution
Citation manipulationCiting to inflate a journal’s or a colleague’s metrics rather than for relevance

Plagiarism in detail

TypeWhat happens
Verbatim (direct)Text copied word for word without quotation marks and citation
Mosaic (patchwriting)Phrases from a source stitched together or lightly reworded, without citation
Paraphrase without creditIdeas restated in your own words but not cited
Idea plagiarismPresenting another person’s hypothesis, method or interpretation as your own
Self-plagiarism (text recycling)Reusing your own published text or data without disclosure
Source-citation plagiarismCiting a primary source you never read, copying the citation from a secondary source

In India, the UGC’s regulations on academic integrity and the prevention of plagiarism in higher education institutions (2018) grade similarity in a thesis or paper, after the permitted exclusions, into levels:

LevelSimilarityCategory
Level 0Up to 10%Minor; no penalty
Level 1Above 10% to 40%Penalties increase with level
Level 2Above 40% to 60%
Level 3Above 60%Most severe

The regulations also list what may be excluded from the check (for example, properly acknowledged quotations, the references and bibliography, and generic terms, standard symbols and equations). Penalties differ for students and for faculty; check the current text of the regulations before relying on the details.

Ethics of Research with Humans and Animals

Human participants

The modern framework grew out of the abuses of human experimentation in the mid-twentieth century. Three documents are commonly cited:

DocumentMain contribution
Nuremberg CodeVoluntary consent of the human subject is absolutely essential
Declaration of Helsinki (World Medical Association)Ethical principles for medical research involving people, including independent ethics-committee review
Belmont ReportThree principles: respect for persons, beneficence, justice
PrincipleIn practice
Respect for persons (autonomy)Informed, voluntary consent; extra protection for those with reduced autonomy (children, people with cognitive impairment, prisoners)
BeneficenceMaximise possible benefits and minimise harm; favourable risk–benefit balance
Non-maleficence (often listed separately)Do no harm
JusticeFair selection of participants; burdens and benefits shared fairly

Informed consent needs: disclosure of the purpose, procedures, duration, risks and benefits; that participation is voluntary; the right to withdraw at any time without penalty; how confidentiality will be protected; and whom to contact. The participant must understand the information and agree without coercion or undue inducement. For children, parental permission plus the child’s own assent is usual.

Research on people is reviewed before it starts by an independent Institutional Ethics Committee (IEC; called an Institutional Review Board, IRB, in some countries). Data should be anonymised (identifiers removed so a person cannot be re-identified) or at least pseudonymised (identifiers replaced by codes, with the key kept separately and securely).

Animals: the 3Rs

The 3Rs were set out by William Russell and Rex Burch in The Principles of Humane Experimental Technique:

RMeaningExample
ReplacementUse methods that avoid animals where possibleCell culture, organoids, computer models, lower organisms
ReductionUse the minimum number of animals that gives a reliable answerGood design and a power calculation instead of “as many as we have”
RefinementMinimise pain and distress and improve welfareAnaesthesia and analgesia, humane end-points, better housing

Animal protocols in Indian institutions are reviewed by an Institutional Animal Ethics Committee (IAEC). Work with recombinant DNA and genetically modified organisms is reviewed by an Institutional Biosafety Committee (IBSC). Laboratories are classed by biosafety level (BSL-1 to BSL-4) according to the risk of the organisms handled, BSL-4 being the most stringent.

Data, Records and Openness

  • Laboratory notebook: a bound or electronic, dated, contemporaneous record of what was done and observed, including failed experiments. Entries are not erased; corrections are struck through and dated.
  • Raw data (original images, instrument files) are kept unaltered; processed versions are derived from copies.
  • Data retention: institutions and funders set minimum periods for keeping raw data after publication.
  • FAIR principles for shared data: Findable, Accessible, Interoperable, Reusable.
  • Conflict of interest: any financial or personal interest that could, or could appear to, influence the work (consultancy, shares, patents, family ties) must be declared, in papers and in peer review.

Authorship and Credit

The criteria of the International Committee of Medical Journal Editors (ICMJE), widely adopted beyond medicine, require all four:

  1. Substantial contributions to the conception or design of the work, or to the acquisition, analysis or interpretation of data; and
  2. Drafting the work or revising it critically for important intellectual content; and
  3. Final approval of the version to be published; and
  4. Agreement to be accountable for all aspects of the work.

People who contributed but do not meet all four (for example, someone who only acquired funding, provided general supervision, gave technical help or supplied a reagent) are thanked in the Acknowledgements.

RoleResponsibility
First authorUsually did most of the work; conventions vary by field
Corresponding authorHandles communication with the journal and readers; takes responsibility for the submission
Senior (often last) authorIn many life-science fields, the head of the group; conventions vary
Contributorship statementMany journals ask what each author did (e.g. CRediT roles)

Publishing a Paper

Kinds of publication

TypeContentPeer reviewed?
Original research articleNew results, full methodsYes
Short communication / letterBrief, important new resultsYes
Review articleSummary and synthesis of published workUsually
Systematic review / meta-analysisReview with a stated, reproducible search and (for meta-analysis) pooled statisticsYes
Case reportOne unusual case or observationYes
Methods / protocol paperA new or improved techniqueYes
PreprintA manuscript posted publicly before (or without) peer reviewNo
Conference abstract / proceedingsShort report of work presented at a meetingVariable
Thesis / dissertationFull account submitted for a degreeExamined, not journal-reviewed

The structure of a research paper: IMRaD

SectionAnswersUsual tense
Title, abstract, keywordsWhat was found, in briefAbstract: mixed
IntroductionWhy was it done? What was known, what gap, what aimPresent (known facts), past (aim)
MethodsHow was it done, in enough detail to repeat itPast
ResultsWhat was found, without interpretationPast
(and) DiscussionWhat does it mean, how does it fit, limitationsPresent and past
References, acknowledgements, conflict-of-interest and data-availability statementsCredit and transparency

A thesis typically adds a full literature review chapter, more detailed methods, all the data (including negative results), and a summary and future-directions chapter.

Sources and citation

SourceWhat it isExamples
PrimaryFirst report of original data or workResearch articles, theses, patents, lab notebooks, datasets
SecondaryAnalyses or summaries of primary sourcesReview articles, meta-analyses, monographs
TertiaryCompilations that point to or condense secondary and primary sourcesEncyclopaedias, textbooks, handbooks, indexes and databases
Grey literatureMaterial not published commerciallyReports, theses, conference posters, government documents

Citation styles fall into two families: author–date (for example Harvard and APA, which cite by author and year in the text) and numbered (for example Vancouver: “[12]”, listed in order of first citation). Whatever the style, cite what you actually read, and cite the primary source for a specific finding.

Peer review

graph TD
    A["Submission"] --> B["Editor's initial check
scope, quality, ethics"] B -->|"desk rejection"| X["Rejected"] B --> C["Sent to independent
expert reviewers"] C --> D["Reviewers' reports
to the editor"] D --> E{"Editor's decision"} E -->|"accept"| F["Accepted, proofs, published"] E -->|"minor or major revision"| G["Authors revise and respond
point by point"] G --> D E -->|"reject"| X F -.->|"error found later"| H["Correction (erratum, corrigendum),
expression of concern or retraction"]
ModelWho knows whom
Single-blindReviewers know the authors; authors do not know the reviewers
Double-blindNeither knows the other
OpenIdentities known; reports sometimes published with the paper

Reviewers must keep manuscripts confidential, declare conflicts of interest, and must not use unpublished ideas from a manuscript they review.

After publication, a correction (erratum or corrigendum) fixes an honest error that does not invalidate the conclusions; an expression of concern flags a possible problem under investigation; a retraction withdraws a paper whose findings are unreliable (from misconduct or from serious honest error). A retracted paper stays visible, clearly marked as retracted.

Open access and predatory journals

ModelHow the reader gets free access
GoldThe journal publishes the article openly, often for an article-processing charge (APC)
GreenThe author deposits a version in a repository (institutional or subject)
HybridA subscription journal makes individual articles open for a fee
Diamond (platinum)Open to read and free to publish; costs met by an institution or society

Predatory journals charge fees while providing little or no genuine peer review or editorial service. Warning signs: unsolicited emails promising very fast publication, a very broad scope, fake or unverifiable metrics, an editorial board that cannot be confirmed, and hidden fees.

Measuring Research: Citation Metrics

Journal impact factor

The two-year impact factor of a journal for year \( Y \):

$$\boxed{\text{IF}_Y = \frac{\text{Citations in year } Y \text{ to items published in } Y-1 \text{ and } Y-2}{\text{Number of citable items published in } Y-1 \text{ and } Y-2}}$$

It is a property of the journal, not of any single paper or author. Citation distributions are highly skewed (a few papers collect most citations), so a journal’s IF says little about a given article.

Author metrics

  • h-index (proposed by J. E. Hirsch in 2005): the largest number \( h \) such that the author has \( h \) papers each cited at least \( h \) times.
  • i10-index (used by Google Scholar): the number of papers with at least 10 citations.
  • Total citations and citations per paper.

Intellectual Property

RightProtectsKey point
PatentAn invention (product or process)Must be novel, involve an inventive step (non-obvious) and be capable of industrial application; usual term 20 years from filing; the details are disclosed in return for the monopoly
CopyrightThe expression of an idea (text, figures, software, images)Not the idea itself; arises automatically on creation
TrademarkA sign that distinguishes goods or servicesCan be renewed
Geographical indication (GI)A product whose qualities come from its place of originDarjeeling tea is a well-known Indian example

Memory Tricks

Don’t Confuse

ThisvsThat
Fabrication: data inventedFalsification: real data altered or selectively omitted
Honest error: corrected by an erratumMisconduct: may lead to retraction
Similarity index: text overlap figurePlagiarism: a judgement about improper credit
Gift authorship: listed but did not contributeGhost authorship: contributed but not listed
Primary source: original dataSecondary source: review or synthesis of it
Preprint: not peer reviewedPublished article: peer reviewed
Impact factor: about a journalh-index: about an author
Anonymised: cannot be linked backPseudonymised: coded, key kept separately
Patent: protects an inventionCopyright: protects expression, not the idea

Common Mistakes

Practice

All questions below are original practice questions written for this page. They are not from any CSIR NET paper.

Level 1: Foundation

  1. Expand FFP.
  2. Which of the 3Rs is achieved by switching from an animal model to cell culture?
  3. Is a review article a primary or a secondary source?
Answers
  1. Fabrication, falsification, plagiarism.
  2. Replacement.
  3. Secondary. It summarises and synthesises primary research articles.

Level 2: Part A Style

Practice Q1. A student repeats an experiment five times. Three runs support her hypothesis and two do not. She reports only the three supporting runs, without mentioning the others. This is best described as:

(a) fabrication (b) falsification (c) plagiarism (d) acceptable, since the excluded runs were negative

Answer: (b) falsification

No data were invented (so not fabrication), and no one else’s work was taken (so not plagiarism). But omitting data so that the record does not accurately represent the research is falsification. If there was a genuine, independent reason to exclude the two runs, it must be stated.

Practice Q2. A journal published 150 citable items in 2024 and 90 in 2025. In 2026, items from those two years were cited 960 times. Its 2026 two-year impact factor is:

(a) 3.2 (b) 4.0 (c) 6.4 (d) 10.7

Answer: (b) 4.0

\( \text{IF}_{2026} = \frac{960}{150 + 90} = \frac{960}{240} = 4.0 \).

(c) divides by 150 alone, and (d) by 90 alone; both forget one of the two years. All numbers here are invented for practice.

Practice Q3. Under the ICMJE criteria, which of the following people should be listed as an author?

(a) The head of department who secured the grant but did not take part in the study or the writing (b) A collaborator who designed the experiments, analysed the data, revised the manuscript critically, approved the final version and agreed to be accountable for it (c) A technician who ran the sequencing as a routine service (d) A colleague who gifted the antibody

Answer: (b)

Only (b) meets all four criteria. (a), (c) and (d) made real contributions that belong in the Acknowledgements, but none of them drafted or revised the paper, approved it and accepted accountability.

Practice Q4. Which statement about a preprint is correct?

(a) It has been peer reviewed but not yet typeset. (b) It is publicly posted before formal peer review, so its claims have not been independently vetted. (c) It cannot later be published in a journal. (d) It is the same as a retracted article.

Answer: (b)

A preprint is shared before (or without) peer review. Many journals accept submissions that have been posted as preprints, so (c) is false as a general rule; check each journal’s policy. (a) describes an accepted manuscript awaiting production.

Level 3: Multi-Step (Part C depth)

Practice Q5. A researcher’s 12 papers have the following citation counts: 48, 33, 20, 14, 11, 10, 9, 9, 6, 3, 1, 0.

(i) Find the h-index. (ii) Find the i10-index. (iii) If the paper with 9 citations (the higher-ranked of the two) gains 3 more citations, does either index change?

Answer

(i) Ranks and citations: 1–48, 2–33, 3–20, 4–14, 5–11, 6–10, 7–9, 8–9, 9–6. Rank 7 has 9 citations (\( \geq 7 \)); rank 8 has 9 (\( \geq 8 \)); rank 9 has 6 (\( < 9 \)). So h = 8.

(ii) Papers with at least 10 citations: 48, 33, 20, 14, 11, 10. i10 = 6.

(iii) That paper now has 12 citations. Re-sorted: 48, 33, 20, 14, 12, 11, 10, 9, 6, … Rank 8 has 9 (\( \geq 8 \)), rank 9 has 6. h stays 8. Papers with \( \geq 10 \): now seven. i10 becomes 7.

The h-index rises only when a paper outside the top \( h \) gains enough citations (here, a ninth paper reaching 9); extra citations to papers already counted do not move it.

Practice Q6. A clinical study on a new oral rehydration formula in children is being planned. Consider the statements:

  1. Parental permission is sufficient; the children’s own assent is never relevant.
  2. The protocol should be reviewed by an Institutional Ethics Committee before recruitment starts.
  3. Participants may withdraw at any time without losing the standard care they would otherwise receive.
  4. Recruiting only from one community because it is convenient, while the benefits are meant for all children, raises a question of justice.

Which statements are correct?

(a) 2 and 3 only (b) 2, 3 and 4 only (c) 1, 2 and 3 only (d) 1, 2, 3 and 4

Answer: (b) 2, 3 and 4 only
  • 1 is wrong. For children able to understand, their assent is sought in addition to parental permission.
  • 2 is correct. Ethics review comes before any participant is enrolled.
  • 3 is correct. Voluntariness includes the right to withdraw without penalty.
  • 4 is correct. The Belmont principle of justice concerns a fair distribution of the burdens and benefits of research; placing the burden on one convenient group is the classic example.

Practice Q7. Match each scenario to the best description.

Scenario
P. A thesis chapter reuses three paragraphs from the author’s own published paper without citing it1. Ghost authorship
Q. A pharmaceutical company’s medical writer drafts a paper but is not named anywhere2. Salami slicing
R. One dataset is split into four short papers, each reporting one outcome3. Text recycling (self-plagiarism)
S. A lab adds hypotheses to the introduction after seeing which results were significant4. HARKing

(a) P-3, Q-1, R-2, S-4 (b) P-3, Q-2, R-1, S-4 (c) P-1, Q-3, R-2, S-4 (d) P-4, Q-1, R-2, S-3

Answer: (a) P-3, Q-1, R-2, S-4
  • P: reusing your own published text without acknowledgement is text recycling (self-plagiarism). Reuse with clear citation and permission, as many universities allow for thesis chapters built on published papers, is acceptable.
  • Q: a substantial contributor who is not named is a ghost author.
  • R: splitting one study into minimal publishable units is salami slicing.
  • S: presenting post-hoc hypotheses as if planned is HARKing.

Quick Revision Table

IdeaKey point
MisconductFFP: fabrication, falsification, plagiarism; honest error is not misconduct
QRPsp-hacking, HARKing, selective reporting, salami slicing, duplicate publication
Similarity indexScreening figure; plagiarism is a judgement
UGC 2018 levels0: up to 10%; 1: 10–40%; 2: 40–60%; 3: above 60%
Human researchRespect for persons, beneficence, justice; informed, voluntary consent; IEC review
Animal research3Rs: replacement, reduction, refinement; IAEC review
Authorship (ICMJE)All four: contribution, drafting/revising, approval, accountability
IMRaDIntroduction, Methods, Results and Discussion
SourcesPrimary (original), secondary (reviews), tertiary (textbooks, encyclopaedias)
Peer reviewSingle-blind, double-blind, open; correction vs expression of concern vs retraction
Impact factorCitations in Y to items of Y−1, Y−2, divided by citable items of Y−1, Y−2
h-indexLargest h with h papers cited at least h times
PatentNovelty, inventive step, industrial applicability; file before you publish

Key Takeaways

What’s Next

Advertisement
Advertisement