Academic Plagiarism in Research: What Indian PhD Students Must Know

Your supervisor just returned your thesis draft with a red flag. The similarity report shows 38%. For thousands of Indian research scholars, academic plagiarism in research stops being a theoretical concern the moment that number appears. Under the UGC Regulations 2018, anything above 10% triggers mandatory review. Above 40%, you’re debarred from resubmitting for a full year. What counts as plagiarism, how Indian universities detect it, what the penalty levels mean in practice, and what to do if you’re already flagged — that’s what this covers.

Key Takeaways

• Academic plagiarism covers six types, from verbatim copying to authorship fraud. Similarity scores, not intent, determine the penalty under UGC 2018.
• The UGC 2018 Regulations set four penalty levels. Level C, above 60% similarity, means cancellation of PhD registration.
• All Indian PhD theses must be deposited in Shodhganga before the degree is awarded, making every submission permanently searchable against 550,000+ existing theses.
• A 2024 study found that 31.1% of Indian research students don’t recognise content similarity as plagiarism (BioMed Central, 2024).
• Since 2024, AI-generated content submitted without disclosure is explicitly treated as plagiarism under UGC and AICTE guidance — and Turnitin’s AI detector is active on most institutional accounts.

What Is Academic Plagiarism in Research?

Academic plagiarism in research means presenting someone else’s work, ideas, data, or expressions as your own without proper attribution. The source doesn’t matter — a published journal article, an unpublished thesis, a conference paper, a website, all of it qualifies. Intent doesn’t change the outcome under Indian academic regulations. Accidental plagiarism carries exactly the same penalty as deliberate copying once the similarity percentage crosses the threshold.

The UGC (Promotion of Academic Integrity and Prevention of Plagiarism in Higher Educational Institutions) Regulations, 2018 provide the legal definition and require every UGC-recognised university to maintain an institutional anti-plagiarism policy. These regulations were notified in the Gazette of India on 31 July 2018 and remain in full force in 2026.

The scope is broader than most students expect. Works covered include:

• PhD dissertations and M.Phil. theses
• Master’s dissertations
• Research articles submitted by faculty at UGC-recognised institutions
• Any work submitted for academic evaluation within the university system

Field doesn’t matter. A chemistry PhD, a social science dissertation, and a law research paper all fall under the same framework. Similarity percentage, not subject matter or career stage, determines the penalty level.

How Plagiarism Detection Works in Indian Universities

India’s detection system is built around Shodhganga, the national thesis repository maintained by INFLIBNET (Inter-University Centre for Information Technology). Every PhD thesis submitted to a UGC-recognised university must be deposited in Shodhganga before the degree is formally awarded. The repository holds over 550,000 PhD theses, one of the largest searchable academic databases in the world, and growing every year.

Detection software compares your thesis against this repository and against commercial databases covering journal articles, conference proceedings, and web content. Turnitin, iThenticate, and Drillbit — the tools most common in Indian universities — each produce a similarity report showing what percentage of your text matches existing indexed content, and which specific sources are responsible.

There’s a key distinction between your raw score and your UGC-assessed score. The excluded similarity score strips out:

• Properly attributed direct quotations
• Bibliography and reference lists
• Matches under 14 consecutive words
• Standard academic declarations and formulaic phrases

Your institution’s plagiarism officer applies these exclusions before classifying your thesis under the penalty framework. The adjusted score, not the headline percentage, determines your level. In our experience at Research Experts, students consistently panic at the raw number, only to find their adjusted score is a full penalty level lower once standard exclusions are applied.

The UGC 2018 Penalty Framework for Academic Plagiarism

Penalties are applied to the excluded similarity score and fall across four levels under the UGC Regulations, 2018:

• Level 0 (Below 10%): Accepted without penalty. Most properly cited theses land here once exclusions are applied.
• Level A (10% to 40%): The thesis is returned for revision and resubmission within six months. No debarment, but the clock starts from the date the formal complaint is filed, not from when you receive your score.
• Level B (40% to 60%): The student is debarred from resubmitting for one full year. For a programme that may have already taken four to six years, this is a serious disruption — stipends, scholarships, and supervisory continuity all affected.
• Level C (Above 60%): Programme registration is cancelled. This is not a suspension — it is a permanent revocation at the same institution under the same registration number.

Supervisors aren’t off the hook either. A guide whose student submits a Level B thesis is barred from supervising master’s students for two years and loses one annual salary increment. Level C brings a three-year supervision ban and two years of increment freeze. Your similarity score affects your supervisor’s career too. This is why most guides who have been through this once now insist on running a preliminary check before any official submission.

Penalties go through institutional panels. The DAIP (Departmental Academic Integrity Panel) handles initial inquiries. If it escalates, the IAIP (Institutional Academic Integrity Panel) steps in, with authority to impose Level B and Level C sanctions. You receive written notice, a right to respond, and a formal hearing before any penalty is finalised.

Six Types of Academic Plagiarism Explained

The UGC Regulations 2018 list eight categories of academic misconduct, though in practice they collapse into six distinct types, each with a different detection profile and risk level:

• Verbatim copying: Lifting text word-for-word without quotation marks or citation. Highest detection rate of any type. String-matching algorithms catch it immediately.
• Paraphrase plagiarism: Rewriting another’s ideas in your own words without crediting the source — the most common mistake among Indian research students who genuinely don’t realise they need to cite a paraphrase. Modern semantic analysis flags this even when vocabulary changes significantly, and a 2024 update to Turnitin’s algorithm made synonym-swapping considerably harder to hide.
• Idea or concept plagiarism: Taking someone’s original theory, framework, or argument and presenting it as your own contribution. Software rarely catches this — which is exactly what makes it dangerous. Expert reviewers and viva examiners almost always do, typically at the viva stage when it’s far harder to address.
• Self-plagiarism: Reusing your own previously published or submitted work without disclosure. This catches a lot of Indian PhD students off guard — many publish a chapter as a conference paper before the final thesis is submitted, assuming their own prior work is exempt. It isn’t. Once indexed, those papers show up in your own Shodhganga screening.
• Mosaic or patchwork plagiarism: Stitching fragments from multiple sources to produce text that looks original but is almost entirely borrowed. Scores low on automated string matching because no single source provides a large block. It surfaces during expert review when a reviewer recognises a specific source.
• Authorship fraud: Ghost authorship (leaving out real contributors), gift authorship (adding names without contribution), or purchasing authorship from paper mills. Subject to both UGC sanctions and COPE (Committee on Publication Ethics) retraction procedures.

Of these, idea plagiarism and mosaic plagiarism are the two most likely to slip through automated screening undetected. Both tend to surface at viva stage or post-publication. By that point, they are considerably harder to address than a high similarity score caught during routine pre-submission checks.

What to Do If Your Thesis Is Flagged

A high similarity score is alarming. But a formal DAIP notice isn’t the end of your programme — most Level A and Level B findings are recoverable, provided you respond methodically rather than rushing a revision under pressure. The sequence that actually works:

1. Get the full similarity report, not just the percentage. The section-by-section breakdown tells you what’s actually driving the score. A 35% score concentrated in bibliography and properly quoted passages is a very different problem from 35% spread across your original analysis chapters.
2. Confirm your adjusted score with the institution’s plagiarism officer. Before editing anything based on the raw number, find out what the excluded score is after standard exemptions. Many students find their actual penalty level is one or two tiers below what the headline number suggests.
3. Categorise each flagged passage by type. Verbatim matches need citation or deletion. Paraphrase matches need genuine rewriting and proper attribution at the point of use. Mosaic sections are different — they need to be rebuilt from scratch, not reworked through synonym substitution.
4. Prepare a written response for the DAIP hearing. Address each flagged section: what was the source, was it cited, was the overlap accidental or structural? Panels assess intent. A specific, cooperative written response shapes that assessment more than most students realise.
5. Rewrite flagged chapters methodically before resubmission. If flagged content spans multiple chapters, reducing the score under time pressure is genuinely difficult. Genuine paraphrasing — reading the source, closing it, and reconstructing the idea in your own sentence structure — is the only approach that works. Do not use synonym-swapping tools. They don’t reduce the conceptual similarity that semantic analysis detects, and they produce grammatically awkward text that raises a separate red flag with examiners.

The Level A revision window is six months from the date the complaint is filed, not from when you first see your score. Students who wait until the last few weeks consistently find the revision scope unmanageable. Start the moment you know there’s a problem.

Conclusion

Academic plagiarism in research in India isn’t a matter of scholarly etiquette anymore — it’s a legally enforced framework with real consequences that follow you for years. Shodhganga makes every thesis permanently searchable, and as AI-generated content gets easier to produce, detection tools and institutional enforcement are both tightening. The students who come through this process intact are almost always the ones who understood the rules before sitting across from a DAIP panel.

The number to hold onto is 10%: excluded similarity below that is accepted without penalty; above it, the revision clock starts and the stakes are real. Always get the adjusted score, not the raw percentage, before assuming the worst. And if you’re already dealing with a flagged submission, start with the full report — knowing which type of plagiarism is driving each passage is the difference between a targeted fix and weeks of unfocused editing.