Free scoping read. A written assessment of your manuscript in two business days, at no cost.

Start one
Contents

Academy · Research Publication · Common Types of Plagiarism in Research: Definitions, Forms and Prevention

Common Types of Plagiarism in Research: Definitions, Forms and Prevention

Most plagiarism in research does not involve deliberate theft. It more often arises from a citation intended to be added later, a paraphrase that remained too close to the source, or the undeclared reuse of an author’s own words. Familiarity with the categories enables authors to avoid the unintentional forms.

Pubrica AcademyWritten by the medical writing and biostatistics teams Reviewed byPubrica editorial team Published2 October 2026 Last checked2 October 2026 Reading time4 minutes

Definition of plagiarism

Plagiarism is the presentation of another person’s work, words or ideas as one’s own without acknowledgement. It encompasses direct copying, close paraphrase, reuse of an author’s own published work, and failures of attribution that were never intended. Intent affects how an editor responds, but it does not determine whether plagiarism occurred.

The forms plagiarism takes in academic writing. Most cases that reach editors are mosaic or paraphrasing plagiarism rather than verbatim copying.
The forms plagiarism takes in academic writing. Most cases that reach editors are mosaic or paraphrasing plagiarism rather than verbatim copying.

Forms of plagiarism

Direct or verbatim plagiarism. Another author’s text reproduced verbatim without quotation marks or citation. This is the least common form in research writing and the most readily detected by similarity software 3.

Self-plagiarism, or text recycling 1. The reuse of an author’s own previously published work without disclosure, for example submitting the same study to two journals, reusing passages from an earlier paper or presenting prior work as new. COPE treats some text recycling as acceptable where it is declared and where the reuse concerns methods rather than findings; undeclared duplicate submission is not acceptable 2.

Mosaic plagiarism, or patchwriting. Phrases and sentence structures taken from a source and incorporated into the author’s own text with a few words changed. It is the most common form in practice, because it usually begins as an unfinished paraphrase.

Paraphrasing plagiarism. Another person’s idea or argument restated in the author’s own words without citation. The wording is the author’s, but the contribution is not. A paraphrase still requires a citation: rewording removes the need for quotation marks but not the need for attribution.

Accidental plagiarism. The same acts, committed through inadequate note-taking, a mislaid source, a quotation that lost its quotation marks during editing, or a citation added to the reference list but not to the text. Most cases that editors encounter are of this kind.

Incorrect or incomplete attribution. Citing the wrong source, citing a source that has not been read because another paper cited it, or attributing a conclusion to an author who did not reach it. This form is harder to detect and more damaging to the scholarly record, because it propagates.

Contract cheating. Submitting work produced by another person as one’s own. It differs from the forms above in that it is never accidental, and it is treated as misconduct rather than as a citation failure.

The position of generative artificial intelligence tools

Text produced by a generative tool is not plagiarism of another author, but publishers treat undisclosed use as a failure of transparency of a related kind. The position on which most publishers have converged is that an artificial intelligence (AI) tool cannot be an author, because it cannot take responsibility for the work; that its use in preparing the manuscript must be disclosed; and that the authors remain accountable for all content, including any fabricated citation the tool produced.

Policies differ between journals rather than between publishers, so the instructions for authors of the specific target journal are the only reliable source.

Prevention of unintentional plagiarism

Four practices that prevent most cases

The source should be recorded when the note is taken, not afterwards. Direct quotations should be marked as quotations in the draft, with the page number. Paraphrasing should be done from memory with the source closed, followed by a check that its structure has not been reproduced. A similarity check should be run before submission rather than relying on the journal to detect overlap: a flagged passage that can be explained is routine, whereas one that cannot is not.

Where overlap with the author’s own earlier work is unavoidable, as in a methods section describing the same protocol, it should be declared in the cover letter and the earlier paper cited. Declared reuse leads to a discussion with the editor; undeclared reuse discovered later leads to an investigation.

Key points for authors and researchers

  • Most cases involve mosaic or paraphrasing plagiarism rather than verbatim copying, and most begin as a citation intended to be added later.
  • A paraphrase still requires a citation: rewording removes the quotation marks but not the need for attribution.
  • Reuse of an author’s own published work should be declared. Declared reuse prompts a discussion with the editor, whereas undeclared reuse prompts an investigation.
  • The source should be recorded at the time a note is taken, and a similarity check should be conducted before submission.
  • The use of generative tools should be disclosed; policies are set by individual journals rather than by publishers.

Frequently asked questions

Journals treat undisclosed reuse of an author’s own published work as a breach of publishing ethics rather than as theft, but the consequences (correction, retraction or rejection) are similar. COPE’s guidance on text recycling distinguishes acceptable reuse, such as a methods description, from duplicate publication of findings.

There is no threshold. A score of 30% composed of reference lists, standard methodological phrases and correctly quoted material may be acceptable, whereas a score of 6% that consists of one uncited paragraph is not. Editors read the report, not the percentage.

No, but the boundary varies by field and audience. Information that had to be looked up should be cited. Information that every textbook in the discipline states without a source need not be.

The journal follows COPE’s flowcharts: an explanation is sought from the authors and their institution, and the outcome ranges from a correction to a retraction depending on extent and intent.

No, and it often worsens the problem. Automatically reworded text is mosaic plagiarism produced by additional means, and it reads poorly enough that editors notice it. It is the citation that makes a paraphrase legitimate, not the degree of rewording.

References

  1. COPE guidance on text recycling, plagiarism and the associated flowcharts. read 2 October 2026. publicationethics.org/guidance
  2. ICMJE Recommendations, sections on overlapping publications and on authorship responsibility. read 2 October 2026. www.icmje.org/recommendations
  3. Council of Science Editors, white paper on promoting integrity in scientific journal publications. read 2 October 2026. www.councilscienceeditors.org

Each source carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.

How to cite this article

Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.

Pubrica Academy. (2026, October 2). Common Types of Plagiarism in Research: Definitions, Forms and Prevention. Pubrica. https://pubrica.com/academy/plagiarism-service/common-types-of-plagiarism/

This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.

Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.

How Pubrica can help

Plagiarism services conducts the similarity check and, more importantly, reviews the report with the author, distinguishing flags arising from reference lists and standard phrasing from passages that require rewriting or citation before submission.

Find something else

Search the Pubrica Academy

Fifty-eight free articles, templates, guidelines, checklists and worked examples are available without an account. Users may enter the subject of their work or open the full library to filter by category, sub-topic or resource type.

Similarity assessment before journal screening

A scoping assessment is free and takes two business days. It provides a written assessment of what the work requires and which specialist would undertake it.

Get a free quote

Regulations

Check your project against this rule

  1. 1Your work
  2. 2Your details
  3. 3Send
What do you need?

Material you send is held under signed confidentiality, with access limited to the assigned team and encrypted transfer. It is not used to train anything and is not passed to third-party tools. Ask for the agreement before you send anything you would rather cover first.