Plagiarism, according to Oxford Languages, is “the practice of taking someone else’s work or ideas and passing them off as one’s own”. We might all remember the grueling days of high school English class, learning about in-text citations and the importance of citing and referencing factual information correctly, but how does that translate for authors in academic publishing?
Just as Artificial Intelligence (AI) is advancing, so are the tools used by editorial offices to screen for plagiarism. iThenticate/Crossref and Turnitin are just a few notable tools that are used.
iThenticate is probably the most common in the industry and the most used in scholarly publishing. The basis of these tools is that they compare each manuscript against a large database of previously published literature to detect plagiarism. iThenticate prepares a Similarity report, and once this report is run, an editor or editorial office can review the report to determine if a manuscript falls into a normal threshold.
The “normal” threshold depends on the publisher or journal’s standards. For most, the industry standard for iThenticate falls below the 20—30% range. Anything much over that standard can indicate a need to check where some of these plagiarism scores may be coming from. If a significant portion of overlap comes from a single source, it can also raise concerns. Editors may request an author revise the manuscript to reduce overlap from that source to below 2–3% to ensure originality and avoid plagiarism issues.
iThenticate has particular sections in a manuscript that it screens for plagiarism. A few of these include the bibliography, quotes, citations, abstract, methods and materials, small matches, and preprint sources. Adjustments can be made to the report to exclude certain elements, such as the bibliography, if an editor is not concerned about those findings.
This brings to mind the question of what defines plagiarism and what does not.
Here are a few items that, traditionally, do not count as plagiarism:
- Preprints: On a similarity report, preprints can often be one of the main reasons a report comes back high. In these cases, as long as an author has referenced or included their published preprint as a citation, most publishers and editors do not consider this a point of plagiarism.
- Overlap noted in the abstract, title page, or end matter: Often, plagiarism tools might flag a manuscript where overlap is noted within the abstract, title page, or end matter (i.e., disclosures and references). This type of overlap tends to be okay and not a red flag to an editor when reading the report.
- Quotations: Within scholarly manuscripts, references and quotations are used frequently. Incorporating quotations within a manuscript will cause a tool like iThenticate to flag for overlap; however, as long as the author cites the source, and the cited source is included in the reference section, this would not be of concern.
Here are a few elements that are considered plagiarism and should be noted by authors before submitting their manuscripts to a journal:
- Self-Plagiarism: Often, an author might submit a manuscript that builds off previously published work. This, in itself, is fine! However, if an author uses previously published work and passes it off as new material, this is considered plagiarism and is likely to be detected by a plagiarism tool. If any author discusses their own previously published work, it needs to be cited within the text and added as a source to the reference section of their submitted manuscript.
- Overlap with larger text portions in the methods or larger manuscript sections: Large overlap in the methods section can be a cause for concern, especially if what is being noted as overlap are full sentences without any citations.This is direct plagiarism and would be noted as a more concerning ethical issue.
Overall, plagiarism is a major topic, and one that needs to be avoided in scholarly work and elsewhere. If an author is curious about the best way to avoid plagiarism, COPE is a great resource to consult to ensure their manuscript follows ethical standards.
For more information, COPE posted a series of seminars and webinars titled “Plagiarism in scholarly publishing: Publication Integrity Week 2024” that may be helpful for authors.
The key takeaway is that if you are including someone else’s work within your own, or citing previous published work of your own, always be sure to cite it and ensure the original work is included within the references for readers to refer to if needed.
*While writing this article, I did reference a colleagues’ presentation titled “iThenticate Primer: Continue Education”. A link to this PowerPoint presentation can be found here!
By Emma Montes
