Skip to main content

Unfortunately we don't fully support your browser. If you have the option to, please upgrade to a newer version or use Mozilla Firefox, Microsoft Edge, Google Chrome, or Safari 14 or newer. If you are unable to, and need support, please send us your feedback.

We'd appreciate your feedback on this new experience.Tell us what you think(opens in new tab/window)

Elsevier
Publish with us
Connect

7 ways to avoid academic plagiarism

1 March 2023 | 9 min read

By Michael Seadle, PhD

Man holdiing mask of his face

Plagiarism can be unintentional; the co-founder of a center to educate authors on research integrity highlights potential pitfalls

Editors note: Plagiarism, the re-use of published work without appropriate credit, is surprisingly common and not always intentional. For academics, however, accusations of plagiarism can be devastating, marring their reputation and career. Prof Dr Michael Seadle is co-founder and director of the Humboldt-Elsevier Advanced Data and Text Centre (HEADT Centre)(opens in new tab/window) and heads up its research integrity efforts. Globally recognized for his work to combat false information, he runs through some of the nuances and danger points authors should be aware of and offers seven helpful tips.

The first and best way to avoid plagiarism is not to plagiarize intentionally. This sounds simplistic, but there are people who persuade themselves that copying just a few useful phrases will help them meet their pressing deadlines. While that strategy may work in some corporate settings, in academic publishing, it is a fool’s game. Most serious academic publishers use software tools, such as iThenticate(opens in new tab/window), that can scan an enormous database of academic publishing and detect copying at a highly granular level. Not all universities use such tools, but many do, and the consequences of being caught include failed classes, lost degrees or lost jobs. The risk is simply not worth the savings in time.

Unintentional plagiarism is surprisingly common, but the fact that it is unintentional does not excuse it. The rest of this article looks at how you can avoid some common practices that risk this kind of inadvertent plagiarism in writing. Image plagiarism represents a different kind of problem that is too complex for this short article to address fully.

Confidence in Research

Plagiarism is one of many factors that can undermine confidence in research. Elsevier has partnered with leading science organizations and Economist Impact for a global collaboration to understand the impact of the pandemic on confidence in research — and identify areas for action to support researchers.

The fine line between paraphrasing and plagiarism

English is overwhelmingly the most common language for modern academic publishing, but English is by no means the native language of all scholars. In the more mathematical natural sciences, certain phrases are so standardized that they no longer really count as plagiarism. One example is the language used to describe the results of certain statistical tests; widely used statistical textbooks often recommend using particular phrases.

In more verbally oriented fields such as humanities, law or economics, the quality of the writing matters, and it is natural to want to improve it by drawing on examples from published sources. While this is a sensible idea in itself, it carries risks if the author reproduces a phrase that is too long or too individual. This problem not only affects non-native speakers; it also happens to authors who “borrow” explanations when writing about a subject far outside their field of expertise. The risk is that they copy so much text that they cross the line into plagiarism. The best solution is to always use quotation marks and proper footnotes, even when that results in a page layout that seems unattractive.

The issue is, of course, how much must be in quotation marks, and how much can just be paraphrased?

Decades ago, paraphrasing rules were fairly relaxed, but today they are much stricter. There is no clear, simple agreement on the number of words that can overlap with the original before a paraphrase becomes plagiarism. Some institutions have published limits, then later removed them from their websites, presumably because they realized they were unrealistically restrictive.

The reason for paraphrasing can be to clarify a statement or to say it more compactly or correctly. At the very least, a paraphrase ought to be able to contain key words referring to the substance of the context and, ideally, function words (words that in themselves carry no meaning and merely structure the sentence). However, for some publishers and universities, every overlapping word counts, so authors need to think carefully about their justification for paraphrasing.

Professors routinely paraphrase in the classroom, where it is generally impossible to use direct quotes without a photographic memory. Even though written texts are more formal, people imitate what they hear rather than looking for rules that are themselves ambiguous.

Paraphrasing is especially common in literature reviews for the simple reason that authors must talk about the specifics of what other writers have said, and they must be able to reuse some of their words to convey the same meaning. This makes literature reviews especially prone to plagiarism accusations. This should serve as a warning to students and scholars to footnote the literature review sections carefully, especially in cases where a direct quotation would not work because of the grammatical structure or length of the phrases, or unclear references.

Separating fact from plagiarism

What may seem like simple facts can also become targets of plagiarism accusations. The six-word statement “Berlin is the capital of Germany” represents a fact and gets over 84,000 hits on Google. Wikipedia and numerous documents confirm that the statement is purely factual; nonetheless, the multi-word overlap means it can be flagged by unsophisticated plagiarism checkers and plagiarism hunters. The real problem is the lack of a clear agreement on what kinds of facts need to be sourced. Standard facts like the boiling point of water (100°C) should need no reference, but facts outside the scope of the reading audience may need one (example: the birth date of the artist Michelangelo(opens in new tab/window) is 6 March 1475).

The importance of acknowledging intellectual ownership

The University of Oxford includes “ideas” in its definition of plagiarism(opens in new tab/window): “Plagiarism is presenting someone else’s work or ideas as your own, with or without their consent, by incorporating it into your work without full acknowledgement.”

Idea plagiarism is a particularly problematic concept because it can cover almost everything and anything. In very specific cases, it can include things like the ideas behind a software program, (if they are legally protected by a patent or have been published and are subject to copyright law). It can also include technical concepts like the structure of a type of DNA, such as a double helix. Without that level of specificity, the claim that someone has plagiarized an idea becomes so vague as to be meaningless.

The Oxford definition(opens in new tab/window) goes on to say: “All published and unpublished material, whether in manuscript, printed or electronic form, is covered under this definition.” This could be seen to include anything on anyone’s computer or in online storage. This serves as a warning about how totalitarian plagiarism policing could become if allowed unfettered access to personal resources.

Recently, artificial intelligence writing tools such as ChatGPT from OpenAI(opens in new tab/window) have been much in the news. Even when an author tells such tools to use references, AI can generate authoritative-sounding output that can be incorrect, incomplete or biased. Applying the technology should be done with human oversight and control and authors should carefully review and edit the result. The more important question is to what degree the use of an artificial intelligence tool counts as the human author’s own work? The Oxford definition implies that presenting something generated by an AI tool could be considered plagiarism since it is not the author’s own work in any conventional sense.

7 ways to avoid plagiarism

While a few bullet points cannot capture all the nuances, these seven steps serve as a summary.

  1. Never intentionally plagiarize. It is not worth the risk because the likelihood of being caught is high at high-ranked journals, and it grows higher every year as more institutions acquire plagiarism identification software.

  2. Always use quotation marks and references. It takes a bit of extra effort and time to find exact quotations and create a well-formed reference, but it is essential. There are many freely available programs that make the preparation and tracking of references easier.

  3. Paraphrase only when necessary and include references. Use quoted passages whenever possible and ensure that any word overlap with the original is minimal and justifiable.

  4. Cite references when using facts that may be outside of the knowledgebase of the intended audience.

  5. Copy phrases with special caution if you are non-native speaker. Do not take whole sentences from published works, or even phrases of more than three words, unless those phrases are purely technical, as in a basic description of statistical results.

  6. Always credit people for their ideas to avoid the risk of “idea plagiarism.”

  7. Generative AI and AI-assisted technologies should only be used to improve readability and language of your work. Authors are ultimately responsible and accountable for the contents of their work.

About the HEADT Centre

The Humboldt-Elsevier Advanced Data and Text Centre (HEADT Centre)(opens in new tab/window) is part of Humboldt-Universität zu Berlin(opens in new tab/window) (HU Berlin) in Germany. It was founded to educate researchers on plagiarism and research image and data fraud, along with other elements of research integrity.  It offers workshops, online modules and certification on these topics via its Information Integrity Academy.

Acknowledgements

Let me thank Dr Thorsten Beck and Ms Melanie Sterzer for their excellent suggestions.

Contributor