Why it is a mistake to use percentage to measure plagiarism

percent-700x716

I’ve been working with plagiarism detection for some time, and I’m often asked what percentage of plagiarism is acceptable.

My answer is quite straightforward: There is no acceptable plagiarism percentage!

I understand that a simply copied sentence, without properly mentioning the source or author, already can configure plagiarism.

Suppose a University decides that 1% suspected plagiarism, calculated by a plagiarism detector of your choice, is an acceptable index. Does it look low, right? Now imagine a 100-page work, where a whole page has been copied from the internet without the proper references. The software would probably detect 1% suspected plagiarism, but this is a full page copied – I can not imagine how this could be ignored. Obviously, this is an approximation, but it makes clear how fragile a percentage-based analysis is.

There are some programs or services of plagiarism detection slightly more sophisticated than others, but at the moment none can identify precise quotations, much less indirect quotations. Such feature would require a very advanced algorithm, certainly based on artificial intelligence.

Since no program can identify the quotations, this information will be summed to the percentage of suspected plagiarism, another factor that compromises the percentage computation.

So, what is the solution? In fact, it is still not possible to automate the whole process; a mechanical analysis, based only on percentages will be imprecise. Therefore, a manual review of the reports of these tools is imperative.

Plagiarism detection programs are still great tools and, if well used, will save hours of work. That’s where I strive for the Plagius to stand out: the Plagius will not bring a magic and ready number, but it’s the tool that supports the reviewer’s work, makes it easy to check the sources, and allows the user to control what they consider relevant.

In summary: Leave the hard work for the Plagius, just take care of the review and stay in control!

Would you like to know more? See this guide about interpreting the Plagius results.

Advertisements

First blog post

Written by the creator of the program Plagius this blog was created to publish tips, tutorials and relevant information about the Plagius program and detection of plagiarism in general.

The main purpose is to help those who need to detect plagiarism in their documents, always looking to assist with explanations focused on specific points, running away from the traditional (and boring) format of the user’s manual. 😉

Eventually posting with tips and tutorials can be updated, looking to follow the evolution of the Plagius program, thus preventing outdated information.

The first postings will be focused on answering common questions sent to the program support, in the future, posts with General issues (in the area of plagiarism) may arise..