Many authors picture plagiarism as a single act, namely copying another person's sentences. Yet similarity software and editorial review pick up several separate patterns, and authors sometimes fall into these by accident while sincerely trying to behave properly. Knowing each pattern makes problems in your own draft easier to spot.
Forms of Plagiarism That Authors Often Fail to Recognise
- Direct copying — lifting text exactly as written with no quotation marks or citation
- Poor paraphrasing — exchanging some synonyms yet leaving the source's sentence structure untouched
- Self-plagiarism — recycling a large part of your earlier published text without telling readers
- Mosaic plagiarism — stitching together phrases from several sources with insufficient credit, even when no one sentence is copied entire
- Uncited data or figures — taking a table, chart or dataset from elsewhere and not acknowledging the source
What Really Happens in Similarity Checking
Software such as Turnitin and iThenticate checks your manuscript against a huge store of published papers, web content and work submitted earlier, and then reports a similarity percentage that shows how much of the text matches known sources. That figure alone is not a finding of plagiarism; it gives a human reviewer a place to start when examining the flagged sections.
Why 15% Is a Typical Limit and Not a Magic Figure
Like many journals, JMVRA asks for a similarity score below 15% before a manuscript goes on to peer review. The figure is not a harsh arbitrary rule; it reflects the fact that some matching text is unavoidable and legitimate. Standard methodology wording, correctly quoted material, common technical terms and reference list entries all show up as "matches" even in wholly original work. Going through each highlighted match shows which ones need action and which can safely stay.
Cases Where a High Score Is Not Plagiarism
| Flagged As | Often Actually Is |
|---|---|
| Strong match in the references section | Conventional citation formatting that mirrors the reference lists of other papers |
| Match within the methodology description | Common stock phrasing for established procedures (e.g. familiar statistical test descriptions) |
| Match in quoted passages | Material that is quoted and cited correctly; the tool rightly marks it as a match, yet it raises no ethics concern |
| Overlap with your own earlier paper | It should be disclosed as related or earlier work, and is not automatically self-plagiarism when cited correctly |
That is precisely the reason a person examines the report instead of letting an automatic cutoff alone decide on rejection.
Honest Ways to Lower Your Similarity Score
- Paraphrase properly by rebuilding the sentence with your own framing instead of only swapping synonyms, and then cite the source
- Put quotation marks and a citation around every passage you retain word for word
- When you build on your earlier work, declare and cite it instead of quietly reusing the text
- Let the literature review reflect your own analysis: summarise and compare the sources, and avoid linking quotations one after another
Consequences of Confirmed Plagiarism
If plagiarism is confirmed in a manuscript before publication, it is sent back for revision or rejected, depending on how serious it is. When it is confirmed after publication, it is handled as research misconduct under JMVRA's COPE Ethics Policy, which may lead to retraction; the full flowchart is documented there.
At-a-Glance Checklist
- Sources creditedAll of them
- Quotes flaggedVerbatim text
- Previous work citedYour past papers and thesis
- Report reviewedEvery match
- Below 15%Prior to submission
Avoiding Plagiarism as You Write
- Record notes in wording of your own — and enclose anything copied within quotation marks, citing its source at once
- Log sources while reading — in a reference manager, so no citation is forgotten afterwards
- Draft from your notes rather than from open papers — this preserves your voice and structure
- Reference your previous work — reusing published text of your own with no citation amounts to self-plagiarism
- Run a similarity check prior to submitting — and go through each highlighted match instead of looking only at the percentage
Paraphrasing Software and AI
| Use | Acceptable? |
|---|---|
| Correcting the grammar of your own text | Yes, and declare it if the journal asks for that |
| Writing notes that summarise papers you have read | Yes, as an aid to your own understanding; cite the sources |
| Rewording another person's text to escape detection | No — that is plagiarism |
| Producing references without verifying them | No — citations invented by AI are a grave mistake |
Refer to the JMVRA AI content policy and the plagiarism guidelines.
Interpreting Similarity Results During Revision of Your Paper
Receiving a report with highlighted text can be stressful. These suggestions help researchers respond methodically, correct genuine issues and explain legitimate overlap when needed.
- Start with the largest matches. Review the biggest blocks of overlapping text first, since they usually carry the most risk. Smaller scattered matches often come from common phrases and names.
- Exclude what is properly excluded. Methods that rely on standard procedures, titles and reference lists commonly produce matches. Understand which settings the checker applied before judging the result.
- Rewrite methods with care. Describe standard procedures in your own words while keeping the technical meaning intact, and cite the original protocol. Never alter terms in a way that changes the science.
- Quote sparingly and precisely. Use direct quotation only when the exact wording matters, and mark it clearly. Overusing quotes raises similarity and weakens your own voice.
- Keep a note for the editor. If a high match is justified, briefly explain it in your cover letter. A calm, honest explanation demonstrates good faith.
Address large overlaps first, rewrite through real understanding and be open with editors about anything that cannot be avoided.
Questions Researchers Often Ask
Understanding Similarity Scores
No percentage counts as “acceptable” everywhere. The software finds text that matches, and an editor then judges whether it is quoted and cited correctly. Before review, JMVRA screens with Turnitin / iThenticate and accepts similarity below 15%.
Ready to Publish? Submit to JMVRA
JMVRA is aligned with UGC's Suggested Parameters for Choosing Peer-Reviewed Journals.
JMVRA is a Transparent, Peer-Reviewed, Open Access Journal (Refereed Journal). Following the discontinuation of the UGC and UGC-CARE journal listing, and in alignment with the latest UGC-CARE recommendations and guidelines, JMVRA is aligned with UGC’s Suggested Parameters for Choosing Peer-Reviewed Journals.
Pages worth visiting on jmvra.org
Key Terms
Subjects and search phrases this article addresses.
Related Reading
Further reading on the same theme from the JMVRA blog.
Ethics Policy
Read the guide RelatedRecognising a Predatory Journal
Read the guide RelatedPredatory Journals and Their Red Flags: A Complete Checklist
Read the guide RelatedSimilarity Check Service for Papers & Theses
Read the guide RelatedGetting Your Research Cited the Ethical Way
Read the guide RelatedCloned & Fake Journal Alert: Protect Your Paper
Read the guideKeep Reading on the JMVRA Blog
As part of the JMVRA journal blog, this guide shares practical help on publishing, indexing, metrics and research integrity, and is written to be verifiable rather than promotional.