How much plagiarism? (revisited)

Print Friendly, PDF & Email

The Bangalore Mirror today publishes a report:  Ctrl-C-Ctrl-V, but only up to 25%, VTU tells its PhD students, with the sub-heading

After installing new anti-plagiarism software to sniff out borrowed material, the technology varsity has realistically left some room for ‘permissible lifting’.”

It seems that students have been turning in their PhD theses with more than 50% “borrowed” material.

A VTU official said the new plagiarism software aims to inculcate in students respect for academic integrity and discipline, even as it identifies acts of dishonesty.

To restore credibility to the University’s degrees, and (as stated in the article) “to inculcate in students respect for academic integrity and discipline, even as it identifies acts of dishonesty,” the amount of allowable plagiarism is to be capped at 25%:

VTU has now fixed the acceptable level of plagiarism at 25 per cent. But this comes with a caveat — even a 0.5 per cent variation from the ‘permissible level’, will lead to rejection of the thesis.

Be warned, students.

Some may think that 25% is generous (or is “disingenuous’ the word I am looking for?), but the University Vice-Chancellor is reported as suggesting that this is in fact very strict:

(VTU vice-chancellor ) Maheshappa said, “It is a known fact that most new theses being submitted are coming up short on uniqueness. In fact, worldwide, universities are reconciled to accepting up to 30 per cent of similarity with published material. We, however, have taken a tough view by putting a cap of 25 per cent,” he said.

It’s a pity that Dr Maheshappa did not name his sources for this pronouncement. There may be a … misunderstanding somewhere?

There is a complicated procedure for submission and for resubmission of theses. In the first stage, students submit their theses to the software, and are are permitted to edit their work to get below 25%.  Once the student signs off on the work, theses are subject to two more checks, and decision made as to the work’s authenticity acceptability.

A little investigation at the university’s website soon found what is surely confirmation of the new procedures: a document with the heading “Objectives of Anti-Plagiarism” states on the very first page

Permitted Similarity index <= 25%

Can’t argue with that.

This document also suggests that the software being used is Turnitin; there is an illustration of a Turnitin originality report.

One wonders – or at least I wonder – what education and training students and faculty are being given in the use of the software and the reports.

 

3 thoughts on “How much plagiarism? (revisited)

  1. This plagiarism software finds normal English words used together in a similar way as a published material as plagiarism!
    Then how is it possible to form new sentences without using common words?
    Introducing anti-plagiarism to MTech is a tough decision on the students. I mean I appreciate the intent behind it but the system of checking should be liberal and should consider the appropriateness of the content to the work.
    If the sentence used is exactly the same then it should be considered as “copy” but finding similar words doesn’t make sense to me.
    On the other hand how can students check the plagiarism percentage of their work before submitting or else the work will be rejected after couple of chances.

  2. Good questions!

    You are right – it is not possible to form new sentences without using common words. But it is the way that those common words are used which could be problematic – if you use words in new ways, make new sentences, write original ideas using words that anyone else might have used, this is not plagiarism. It is using the same words in the same order to express the same ideas, that is unoriginal and could well be plagiarism.

    You could be given the benefit of the doubt if it is a one-off occurrence, that could be pure coincidence … but lots of strings of sentences used by other people, that is suspicious. Instructors and readers are not looking for completely original vocabulary, but they do expect original thinking. (There is, of course, the option of quoting and citing, or paraphrasing and citing, someone else’s words or ideas if you do need to use them.)

    In any case, text-matching software (often but mistakenly called plagiarism detection software) should not be picking up single words because they have been used by someone else; it is strings of words they look for, and the more strings there are, the more suspicious a reader might be.

    Even then, even with 25% or more text matched, this is not necessarily plagiarism – the human mind must still decide. It is not the percentage alone that matters … it is what those matches reveal, and software cannot make those decisions. In a comment in an earlier post to this, How much plagiarism http://www.read2live.com/2013/03/27/how-much-plagiarism/, I have advised another reader what to check and when to dismiss text-matches picked up by Turnitin. Take a look – that might help answer your last question.

    • Thanks for your reply.
      I agree with you that Turnitin is a “text-matching software” rather “plagiarism detection software”.
      I am not sure as for how much the human minds at VTU would worry about the “Similarity index” of student’s reports.
      Anyways, hope for the best.
      Thanks again.

Leave a Reply to PavanJarmale Cancel reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.