|Year : 2019 | Volume
| Issue : 6 | Page : 481-482
Plagiarism detection: Perils and pitfalls
Department of Community Medicine, Dr. DY Patil Medical College, Hospital and Research Centre, Dr. DY Patil Vidyapeeth, Pune, Maharashtra, India
|Date of Web Publication||17-Oct-2019|
Department of Community Medicine, Dr. DY Patil Medical College, Hospital and Research Centre, Dr. DY Patil Vidyapeeth, Pune, Maharashtra
Source of Support: None, Conflict of Interest: None
|How to cite this article:|
Banerjee A. Plagiarism detection: Perils and pitfalls. Med J DY Patil Vidyapeeth 2019;12:481-2
Few years ago, we got an alert from a whistleblower in South Africa as follows, “I read the article Surgical repair of giant inguinoscrotal hernia containing the urinary bladder, published in your journal with great interest. I came across another article, which discusses similar issues. Unfortunately, the authors of the first paper have plagiarized the article extensively (229 words, 25%) from the latter article. The source has not been acknowledged in the published paper. It is unfair to publish such an article in your esteemed journal. I wish that the editors will consider this as an offense as this will compromise the quality of the journal as well as put a question on its credibility.”
This whistleblowing evoked mixed feelings in us. First, we were happy to know that our journal, which was not very old, was being read in another part of the globe. We were also pleased with the word “esteemed” to describe our journal. On the downside, we were greatly perturbed at the allegation of plagiarism and really concerned about the future credibility of our journal.
We went about investigating the allegation. For ease of narration, let us call the alleged source, paper A, and the alleged plagiarism, paper B. Imagine our surprise when we found that paper A was authored by one of the reviewers of paper B. We replied to the whistleblower, “We have examined the matter. It would interest you to know that one of the referees of the paper was Dr. Panagiotakis himself, the lead author of the paper from which you say the paper published by us has been plagiarized. He approved the paper after minor corrections. We are attaching the screenshot from the reviewer's panel of the journal system to indicate the date of review by him. There are likely to be common technical terms in both the papers which will inflate the similarity percentage by the plagiarism software. In such cases, we should use our judgment and grant benefit of doubt to the authors.”
We expected the whistleblower to be satisfied with this reply. We were wrong. He stood his ground, “I believe the present explanation is not strong enough to support that the published article is not plagiarized. If a referee has approved the manuscript, it does not prove that the article was not plagiarized. The evidence in favor of plagiarism has been ignored. It questions the integrity of the peer review and the efficiency of the reviewer (strong words and loud whistle indeed!). It would be interesting to know the views of Prof. Panagiotakis, whether he was aware of the fact that the paper was plagiarized.”
We sought the opinion of the reviewer. He responded, “I read with great interest the exchanges. My co-authors and I do not think that there was any kind of plagiarism.” At last, we thought that the whistleblower would be convinced and absolve the paper of plagiarism charges. However, still, he was not convinced, “I discussed with my colleagues and they feel the Committee on Publication Ethics (COPE) guidelines need to be followed. If a reader has suspected plagiarism and had provided the proof (referring to similarity index of 25% from the plagiarism software), it needs to be taken seriously. My colleagues have further advised to discuss the issue in the International Editor's Forum (whatever that means). Please maintain integrity of review process. Please find attached COPE guidelines for your reference.”
We replied, “We are extremely grateful to you for forwarding the COPE guidelines. However, we have investigated the matter more thoroughly than any guideline can recommend. We obtained the opinion of the reviewer from whose paper, a coincidence, the article has been alleged to have been plagiarized. We had forwarded his inputs on the issue to you. We do not see what needs to be resolved now as his verdict in consultation with his co-authors absolved the paper of any plagiarism. There has been no conflict of interest in the review process either. The author and reviewer are from different countries and not known to each other in the best tradition of a double-blind peer review. We are willing to cooperate with any international forum which desires to investigate the issue further. All the evidence and correspondence which we sent to you will be sent to them also on request.”
What do the above exchanges illustrate? They stress that plagiarism detection software can mislead. This software can generate both false-negative and false-positive results.
They may fail to detect plagiarism, for instance, when the content has been translated from text written in another language or taken from multiple sources. Similarly, the software will also fail to detect plagiarism if the text has been copied from sources which are not on digital platform such as old textbooks.
On the other hand, outputs from plagiarism detection software if not interpreted judiciously will lead one to wrongly conclude that the content is plagiarized as happened with our whistleblower. Plagiarism detectors can become a crutch and a problem instead of resolving the issue of plagiarism.
Going strictly by the numbers generated by the software to identify plagiarism is laziness. Human judgment should be combined with numbers, such as similarity percentage and duplication percentage, which different plagiarism software generates. Observations such as different font sizes and different styles of writings in the same manuscript should alert one to the possibility of plagiarism, besides these numbers.
Going solely by the plagiarism detection software is like making a clinical diagnosis by looking at the results of laboratory tests without taking a history from the patient or undertaking a thorough clinical examination. Alas, this too is becoming all too common in the present-day clinical practice.
| References|| |
Tomar SS, Bedi SS. Surgical repair of giant inguinoscrotal hernia containing the urinary bladder. Med J DY Patil Univ 2013;6:324-6. [Full text]
Panagiotakis GI, Spyridakis KG, Chatziioannou MN, Kontopodis NG, Kandylakis SE. Repair of an inguinoscrotal hernia containing the urinary bladder: A case report. J Med Case Rep 2012;6:90.
Weber-Wulff D. Plagiarism detectors are a crutch, and a problem. Nature 2019;567:435.