We recently became aware of some private and public universities using plagiarism software (probably Turnitin) for appointments, service confirmations, and/or evaluating promotion applications of faculty members. We are clearly very alarmed to observe the application of the software. This is an egregious abuse of a useful tool of faculty members and to some extent deemed irresponsible use by the authority. The software available known as plagiarism software was never intended to be used this way, and to our knowledge, it is not used this way anywhere in the world. The evaluation of a promotion application, for example, should clearly be based on the merit of the applicant, the documents supplied with the application, the teaching and research skills of the applicant, and his/her contribution to the university and to the discipline. Whether a candidate has misappropriated someone else’s work as his/her own is an issue that may appropriately be addressed in the subsequent analysis of the merit of the application. Whether or not an author is entitled to credit for a particular publication is a reasonable evaluation issue, and it can be addressed more appropriately through a different process. If an author’s credit is to be contested, there are better ways. Research questions, objectives, rationale of the study, and more specifically, the uniqueness or contribution of the study should get priority for evaluating research work instead of ward, phrase, or sentence similarities.Plagiarism software: What does it do?
Different plagiarism software uses a different algorithm to check the inappropriate use of someone else’s work without proper attribution. Use of exact phrases used in a different work is discouraged. A typical way is to check if a long phrase matches exactly or closely with some other article, writing, or publication in the data bank of the software. If the author had used the same phrase in another place in one of his/her own articles, the software will flag the phrase indicating the existence of a match. It cannot tell that it has not been inappropriately borrowed from somebody else. This is one way the software helps an aspiring author by indicating that it will be better for the author to rephrase the same message.
Another way of checking a match is to compare, let's say every third word, in a paragraph. This will detect extensive borrowings of somebody else's work. This is effective when a student had borrowed a paragraph from someone else and used “find & replace” function to substitute keywords with other words. In the end, the software reproduces the similarity index. It does not use the term “plagiarism”, and the similarity does not necessarily imply inappropriate borrowing of somebody else's work. When there is a match between article A and article B; there is no way to say that A borrowed from B or B borrowed from A.
Intended use of the software
With the advent and increased popularity of the Internet, a great thing that happened is that the students got access to many sources that we could not dream of in our time as a student. Unfortunately, the technology also ushered in things like “cut and paste” which students deployed very effectively in quickly putting together an article with extensive inappropriate use of others’ proprietary work. Instructors could easily identify that the submitted works were not truly or materially the particular student's work either because the maturity of the phrases clearly indicates a mismatch with the student's intellect, or callously or accidentally leaving certain words or phrases in the article that left the gun smoking. The smoking gun cases were easy cases to implicate the students, but many other times, as an instructor, you knew that a student plagiarised, but you could not locate the source right away.
The plagiarism software came to the rescue. The instructor could now locate the source of the original article, provided the original article (or the source article) is in the data bank of the software. Use of the software resulted in a change in the behavior of the students. It also turned out to be a great tool to teach young students how to make proper citations and attributions. This actually led to better quality students' papers. This software is mostly marketed to check student papers and is a widely used tool, and it is not intended to be used in checking the authenticity of the papers written by a professor.Can you, should you use plagiarism software to check authenticity of an author’s work?
Plagiarism is a very serious form of academic dishonesty. In recent years, academic dishonesty has been in frequent discussion and it is true also in Bangladesh. However, there is simply no justification for accusing someone of dishonesty based on a similarity index produced by software intended for use in the classroom. High similarity index does not necessarily imply plagiarism. In fact, certain type of works is expected to have high similarity index values. The more technical is the nature of the work, the higher is the likely value of the index.
If an author’s published work is to be challenged, the investigation should be thorough since it will be highly improper to jeopardise somebody’s career with untrue accusations. The journal editor should be contacted to do an investigation if an accusation I serious enough and credible. The editor will take it very seriously and will withdraw a paper if a thorough investigation reveals academic dishonesty which will be an ignominious event in the career of an academic. What the editor is likely to do, unless the validity of the accusation is obvious, is run IThenticate software (not Turnitin or other software) and generate a report of all the matches which will be sent to a number of experts on the subject. Matches do not imply plagiarism and do not indicate who borrowed from who. If misuse or abuse is identified, all flagged articles from that journal (and possibly other journals) will be withdrawn.
We talked about matching phrases. What does the software compare your article to? Your article will be treated as a brand new article and the software will compare this article to all the articles published and student articles in the data bank of the software. So, if you have assigned a project on a similar project to your student, and the student or you have run this software to detect matches, the similarity index will be high. If you run the same article twice, you are going to have near 100 per cent match the second time because,in the latter run,the comparison will also include the article you ran the first time. That is why it is very important that when you are writing an article or thinking of submitting it somewhere, make sure that you choose “no repository option” before your run the software. The articles that have lots of technical words or phrases are likely to have higher similarity index. It’s not possible to have a zero similarity index. If you have already published an article and this is your second article you are running through the software, you will find that you "plagiarised" your own name.
Some advice for prospective authors
A match of a sentence or sentences will not lead to a charge of plagiarism. Intentionally misuse and misappropriation of somebody else’s intellectual and/or creative work or finding will be treated as academic dishonesty. The following are some of the things you should do to avoid any suggestion that you have been academically dishonest.
1. Be sure to make proper attribution and reference when you use some previous work. This is true even for your own work. You must make a proper attribution to your previous work if you have used any part of it or any idea stated in the previous work of yours. "Plagiarising" your own previous work without attribution is regarded as academic dishonesty.
2. Publishing or trying to publish basically the same paper in different journals (with some rephrasing/paraphrasing) is treated as academic dishonesty. Your new article should make a new contribution. Publishing an English article in an English journal and publishing the same article in Bangla in a Bangla journal without making it clear that it is a translation of the original English article is regarded as academic dishonesty. They cannot, should not, and will not be counted as two separate publications.
3. Including someone’s name as an author when he/she makes no contribution in the paper is inappropriate.
4. If you submit a paper to a journal, it is likely to be checked via a plagiarism software, and if your article is original, you may still receive a "request" to paraphrase parts of your work to bring down the similarity index. For that reason, it is not a bad idea to run the paper through plagiarism software before you submit a paper. But if you do so, be sure to choose the "no repository" option. There is no benchmark as to what is an acceptable similarity index. Some technical papers and papers containing reviews and quotes from other papers are likely to produce high similarity index. As for ourselves, we try to keep the similarity index below 30 per cent. A paper that is a review of all the developments on a particular subject (synopsis) may produce a higher similarity index.
5. For academic administrators. Do not use plagiarism software to evaluate the merit of a paper. Journal rankings, impact factors, and citation index are widely available these days. If you still want to produce a “similarity index”, be sure to also produce the complete “matches” and “flags” and give the faculty member the opportunity explain the matches. Again, we emphasise that similarity does not imply “plagiarism”; it simply means that identical phrases were found in other articles.
The writers are Director of Graduate Studies, North South University and Chairman, Department of Marketing, Jagannath University respectively.