Research Ethics for Students: Plagiarism, Authorship and Data Explained
This guide explains research ethics students India should know before they submit anything: what counts as plagiarism under the UGC rules, who deserves authorship, and how to handle data honestly. Ethics is the system that makes research believable — if readers cannot trust that the words are yours, the authors did the work, and the data are real, then the work is worthless, however clever it is. Learn these rules as a student: the penalties for breaking them as a researcher are severe, and “nobody told me” is not a defence.
On this page
- The answer in short
- Research ethics students India: the UGC plagiarism rules
- Authorship: who deserves to be named
- Data: fabrication, falsification and images
- Self-plagiarism, AI tools and predatory journals
- What to do if you see misconduct
- Common mistakes
- Your checklist
- What to do next
- Frequently asked questions
- Sources checked
The answer in short
Three rules cover almost everything. First, never present someone else’s words, ideas, figures or data as your own — cite them, quote them, or rewrite them with a citation. Second, name as authors only the people who actually did the work and take responsibility for it — not your supervisor for status, or a friend who did nothing. Third, never invent, alter or cherry-pick data — report what happened, including failures, and keep your raw data and notebooks. In India, the UGC’s 2018 regulations set the penalties for plagiarism, from a required revision to cancellation of registration. Know the levels before you submit, not after.
Research ethics students India: the UGC plagiarism rules
In 2018, the University Grants Commission notified the Promotion of Academic Integrity and Prevention of Plagiarism in Higher Educational Institutions Regulations, 2018. They apply to students, faculty and staff across India, and cover theses, dissertations and publications. Every institution must have its own plagiarism policy and panels — a Departmental Academic Integrity Panel and an Institutional Academic Integrity Panel.
The regulations define plagiarism as passing off someone else’s work or ideas as your own. Similarity is classified into four levels, after standard exclusions. Level 0 is up to 10%: minor similarities, no penalty. Level 1 is more than 10% up to 40%: for a thesis, submit a revised manuscript within six months; for a publication, withdraw it. Level 2 is more than 40% up to 60%: the student is debarred from resubmitting for one year. Level 3 is more than 60%: the student’s registration is cancelled, and an already-awarded degree or credit can be put in abeyance.
Two things matter. First, a similarity percentage alone does not prove plagiarism — the regulations require a formal process, including a chance to respond, before any penalty. Second, certain content is excluded from the calculation: properly quoted and cited text, references and bibliography, the table of contents, standard equations and notations, generic terms and scientific names, and up to 14 consecutive words matching by coincidence. So do not chase a percentage to stay safe. Cite everything you did not write or discover yourself: a similarity score is a signal for your institution to look closer, not a verdict.
Authorship: who deserves to be named
Authorship is credit plus responsibility, and most journals follow the standard of the International Committee of Medical Journal Editors (ICMJE). Under the ICMJE Recommendations, a person qualifies as an author only if they meet all four criteria together: substantial contributions to the conception or design of the work, or to acquiring, analysing or interpreting data; drafting the work or reviewing it critically for important intellectual content; final approval of the version for publication, and agreement to be accountable for all aspects of the work. The connector is “and”, not “or”. Someone who meets fewer than all four belongs in the acknowledgements, not the author list.
That rule settles most authorship arguments. Funding a project, general supervision or administrative support does not by itself make a person an author, and neither does writing assistance or language editing alone. Equally, everyone who meets all four criteria should be named — leaving a real contributor off is also misconduct.
For a student, the practical rules are simple. Decide the author list with your supervisor before writing, not after the paper is drafted, and keep a note of who did what. Never add a name to gain favour, and never accept being left off work you did. If a dispute arises, the ICMJE says the institution where the work was done — not the journal editor — should investigate.
Data: fabrication, falsification and images
Fabrication means inventing data that was never measured. Falsification means changing real data — dropping inconvenient points, altering values, or cropping an image to hide what contradicts your claim. Both are the most serious forms of misconduct, and both are usually discovered — by a colleague who cannot reproduce the result, a reviewer who notices an inconsistency, or an image check.
The protection is boring and it works — keep everything. Keep raw data in its original files, lab notebooks with dates, the code that produced each figure, and original unedited images alongside any processed version. If a result looks too good, check it: repeat the measurement, re-run the analysis, ask a colleague to look. Report negative results honestly: a careful experiment that found nothing is useful science, and a perfect result that was invented is fraud. Never copy a figure, table or photograph from a published paper without permission and a citation — under the Indian Copyright Act, 1957, reusing published figures without permission can be copyright infringement, separate from any plagiarism rule.
Self-plagiarism, AI tools and predatory journals
Reusing your own published words without citation — self-plagiarism — is also misconduct: journals require original text, and a thesis chapter copied from your own earlier paper still needs a citation. Submitting the same paper to two journals is also misconduct. The UGC’s 2018 regulations contain no explicit clause on AI-generated text — they were written before AI writing tools existed — so check your institution’s and your journal’s policy before using AI tools on submitted work, and disclose what you used when in doubt.
Predatory journals are the other trap for beginners. They promise fast publication for a fee, with little or no peer review, and a paper in one can damage your record. Before submitting anywhere, check whether the journal is on the UGC-CARE list, is indexed in a recognised database, and states its peer-review process and fees clearly. A journal that emails you out of the blue praising work it has never seen is not one to trust.
What to do if you see misconduct
If you suspect plagiarism, invented data or authorship abuse in your own lab, write down what you observed, with dates, and keep copies of the relevant documents. Then raise it privately with a trusted teacher, your head of department, or your institution’s integrity panel. Raising a concern through the proper channel protects you and the record; spreading it as gossip helps nobody. You can also consult the Committee on Publication Ethics (COPE) guidance, which most journals follow.
Common mistakes
- Copying a paragraph and changing a few words is still plagiarism — cite it, quote it or rewrite it properly.
- Forgetting to cite figures and data. A borrowed figure needs a citation and usually permission, like borrowed words.
- Adding a supervisor’s name automatically. Authorship needs all four ICMJE criteria, not a job title.
- Deleting inconvenient data. If you exclude data, say so in the methods and explain why.
- Trusting a similarity score. A low score does not prove originality and a high score does not prove guilt — the process decides.
- Submitting to a journal you have not checked. Verify it on the UGC-CARE list before you pay or submit.
Your checklist
- ☐ Every borrowed word, idea, figure and dataset cited; quotes marked
- ☐ Author list decided before writing, on the four ICMJE criteria
- ☐ Raw data, notebooks, code and original images saved and dated
- ☐ Negative or inconvenient results reported, not hidden
- ☐ Journal checked against the UGC-CARE list before submission
- ☐ Institution’s AI-use policy checked before using AI tools
What to do next
Read the guide on how to read a research paper as a beginner, and the guide on how to publish your first research paper as a student. Set up a free ORCID at orcid.org if you have not already. Check the Research Guides hub for the other guides.
Frequently asked questions
Is it plagiarism if I copy from my own earlier assignment?
Reusing your own submitted or published work without citation is self-plagiarism, and most universities and journals treat it as misconduct. Cite your earlier work as you would cite anyone else’s, and check your institution’s rules on reusing your own thesis material.
What similarity percentage is safe for a thesis?
There is no safe percentage. The UGC levels treat up to 10% similarity as Level 0 with no penalty, but that is not permission to copy 10%: originality is still required, and your institution applies its own policy. The right target is zero uncited borrowing, not a number under a threshold.
Sources checked
- University Grants Commission, Promotion of Academic Integrity and Prevention of Plagiarism in Higher Educational Institutions Regulations, 2018 (notified 23 July 2018, published in the Gazette of India 31 July 2018) — https://www.ugc.gov.in/ — levels, penalties and exclusions verified live in the official regulation text (Gazette notification PDF, ugc.gov.in/pdfnews/7771545_academic-integrity-Regulation2018.pdf) on 21 November 2026.
- International Committee of Medical Journal Editors, Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals — https://www.icmje.org/recommendations/browse/ — four authorship criteria, all four required together (checked 19 November 2026).
- Committee on Publication Ethics — https://publicationethics.org/ — guidance on handling misconduct, followed by most journals (checked 19 November 2026).
- ORCID — https://orcid.org/ — free persistent researcher identifier (checked 19 November 2026).
- The UGC-CARE list of journals is maintained by the University Grants Commission; the Indian Copyright Act, 1957 position on reusing published figures is as stated in secondary legal commentary checked 19 November 2026.
- The AI-tools note reflects the position reported in secondary sources on 19 November 2026: the 2018 regulations contain no explicit AI clause, and many institutions have adopted their own rules — check your institution’s current policy.
- The checklists, examples and advice in this guide are editorial guidance from IISER Connect, not rules stated by any institute.