Playing Devil’s Advocate for Online Plagiarism Checkers
December 4, 2011 6 Comments
Let me begin by stating upfront that I do not support plagiarism. This blog posting is to play devil’s advocate to the current systems which compare a document to online sources as well as sources within internal databases.
Here’s an example: four graduate students, at a major mid-western university, work for hours writing a final position paper for their capstone finance class. The position paper is discussing a recent HBS case the team has analyzed throughly. Locked away in a conference room in the business school, they create the document by collaborating, choosing each word carefully and discussing the meaning of each sentence because of the limited work count and the importance of the final paper. Their professor demands that each paper be uploaded to a website online for submission. The website will automatically check for plagiarism before the document is submitted. The website gives the students the option (for a small fee!) to see the results of the plagiarism check before the document is delivered to the professor.
One of the students decides to pay the fee to check the final paper. The professor has demanded a plagiarism score of less than 5% (not unreasonable). The other teammates scoff as the team has been working together this entire time, carefully crafting each sentence to maximize meaning and space, there is no way the score could be over 1%…
Result: 14% plagarized
Had the team submitted the paper, without paying the fee, they would have violated university policy and may not have received their diplomas.
So what is the issue? How did an original work result in a score of 14%?
The online system in question, Turnitin, scans the internet for sentence order and word frequency within a given sentence. Additionally the system stores all previous submissions to create a database of original works to compare the document against. The idea being that a student at University A might be able to copy another student’s work from University B. The system builds a database to combat this problem.
However this system creates a new problem, original work is becoming harder to produce because of the immense volume of information currently on the internet and the growing volume within internal databases.
My own theory on this problem is that because so many schools are using the same HBS, Darden and Ivey (just to name a few) business cases, students are creating effective solutions and frankly there are only so many ways to state the same information. How diluted does the meaning of a sentence become by modifying the structure just to make it original. Furthermore with each minor modification, the system now has a new document to compare against making it increasingly difficult for future students.
I don’t have a solution to this problem. Plagiarism in academia is pervasive, and reducing the number of comparison documents doesn’t seem like a viable solution as given in the University A & B example above but as the volume of information on the internet continues to grow it already appears that creating original work (especially based on a limited case scope) is becoming increasingly difficult.
Does anyone agree / disagree?
Source:
http://cyberdash.com/plagiarism-detection-software-issues-gvsu
http://www.plagiarism.org/plag_article_did_you_know.html
http://www.plagiarismtoday.com/2008/12/16/review-the-plagiarism-checker/
About Jason B
jasonbucholz 
I defintely agree with your analysis with regards to case studies. As you mentioned how many different ways can you really analyze a case that is meant to be interpreted in a certain way. How do you combat this problem? I am not really sure. I imagine at some point common sense paired with the program might have to take over. Also, many of the papers submitted to that checker might not actually be available anywhere else. Therefore, students are not finding out that someone wrote a similar sentence until they are finished. Does this now mean that every idea must be original? It definitely raises interesting questions.
A program like this might make more sense if it is used for more academic papers that are meant to be based on research, rather than case studies.
The problem isn’t the systems themselves but how they are used. Plagiarism detection systems can never actually find plagiarism, only copying. It’s up to a human to determine what is or is not plagiarism. Saying that you want a paper under 5% matching is ridiculous as it’s easy to have a paper that’s more than 20% matching but not plagiarized at all.
Basically, if you hit your thumb with a hammer, you can’t blame the hammer. That’s how I see it.
Apart from the difficulty of producing new original content indefinitely from a limited number of source materials (which I think is an interesting question in its own right):
Is having the students pay a small fee extortionary? I know thats a strong word, but I’m not sure why the students have to cough up cash to see if they’ve violated some arbitrary, magical, “matching boundary.”
Why isn’t this included in whatever the professor is paying to use the service? Or, if it is free for the professor, why are the students charged?
The whole notion of paying for a “plagiarism free certification” (or at least, paying to see if you’ve passed under an arbitrary level) strikes me as a dangerous road to go down.
Haha – I’m glad you picked up on those subtle jabs at the ridiculousness of paying a fee to see the perceived ‘plagiarism’ prior to submission.
I agree with your questions and your final statement. It is a ‘dangerous road to go down’. Fortunately I haven’t heard of any BC professors requiring submission to any paid plagiarism checker.
I agree with your post and some of the comments above. As you said, so many schools now use the same cases, and eventually with that much information out there you’re bound to find a few papers that use the same words, phrases, and even sentence structure. While there is obviously a need for some sort of plagiarism checker in today’s world, basing a judgement of plagiarism from the results of these websites can be tricky. I agree with Adam’s comment above that some sort of common sense aspect needs to be used in conjunction with these sites. It almost makes you want to go back to when the internet wasn’t even available to students (wait, actually, no it doesn’t…..)
I agree with your analysis and actually think that this holds true not just in academia. Even the most recent tech lawsuits between Apple and Samsung remind me of this. In that case, it’s a little more extreme and a clearer case of patent infringement. Yet at the same time, most smartphones look the same to me and I feel like there are only so many possible designs for one. What I’m inelegantly trying to say is that we see so many cases of idea infringement throughout different aspects of our lives. I question how well an plagarism detection system works if ideas tend to criss cross organically.
I just finished a meeting with a group member where we each prepped separately and met to discuss our ideas. Turns out we had each written very similar wording and ideas. Yet if we had both turned our ideas in, I guess it would have been considered plagarism by a detection system. And yes, I agree that this may be due to the ubiquitous use of HBS case studies.