Ever since there have been plagiarists, there have been people trying to catch and stop them. But while the technology used by both plagiarists and plagiarism-fighters have improved drastically over the years, the cat and mouse game between the two hasn’t changed drastically.
For every new tool that the plagiarists start to use, a new tool is developed to counter it and every tool anti-plagiarists use to track content, there are countless people poking and prodding it to find weaknesses they can use against it.
But while there are weaknesses in nearly every plagiarism detection tool and method, it seems a lot of would-be mastermind plagiarists are a bit behind the times. Many put their faith in alleged weaknesses that, simply put, don’t exist or have long since been closed off.
Here are just five examples of sneaky ways plagiarists try to escape detection and why they don’t work.
1. The Macro Trick
This one was common in academia for a time and relied on a quirk with modern word preocessors, in particular Microsoft Word.
The idea was that you would write a non-plagiarized (often irrelevant) work in your document and then use macros to swap it out with the plagiarized one that you wanted traded. The idea is that most plagiarism checking applications can’t read macros so they would be scanning the non-plagiarized version and the instructor, who opened the paper in his or her word processor, would see the plagiarized copy.
Why it Doesn’t Work: As clever as it is, this one flat out doesn’t work.
There are several problems with it. First, some plagiarism checking software will reject Word files with Macros, refusing to even scan them on security grounds, the same as many teachers disabling macros on their machine for the same reason.
Second, even if it does go through, an attentive teacher is likely to notice that the paper run through the plagiarism checker is different from the one they’re grading. Finally, most grading systems used by schools involve putting the content online in a fashion where macros won’t work.
In short, this sytem can only work with very dated plagiarism detection tools, grading tools and inattentive teachers. Even worse, it’s likely to draw more attention to your paper if there’s some kind of technical problem preventing it from opening.
If anything, this kind of trick hurts your chances of getting away.
2. The Letter Swap
Similar to the macro swap, the letter swap uses a slightly different trick. The idea is to take a common letter in the language, usually the lowercase “E” and swap it for a similar-looking letter in another language.
To the instructor, it would look just like an “E” and be read as such but the plagiarism checker would be unable to interpret it and match it.
Why it Doesn’t Work: This trick was discovered over five years ago and was actually effective, for a very limited time, against at least some plagiarism checkers. However, developers quickly closed the loophole. Now, similar letters in the same language are matched.
However, even more simple plagiarism checkers are not fooled by this one as they can’t process non-standard characters and simply return an error. This strategy requires a plagiarism checker that can process the non-standard characters but hasn’t been patched to fix this trick, an unlikely combination.
But even if it does work, the paper would likely generate a 0% score for copying, which, as most instructors and editors can tell you, is even more suspicious than a high one.
3. The Careful Rewrite
The idea behind this one is pretty straightforward. All plagiarism detection software is really, at a fundamental level, copy copy detection software and they wall work by matching strings of text. Therefore, theoretically, if you edit an article enough, replacing enough words, no strings will be detected and the software will not be able to match it to anything, making it appear original.
Why it Doesn’t Work: There’s nothing wrong with the theory on this one, but a lot wrong with the practice.
While it is true that plagiarism detection is, more or less, copy matching, it’s very advanced copy matching when it’s done well. While it’s definitely possible to rewrite something enough to the point that an automated system can’t detect it, the amount of rewriting needed to do that is very high. In some cases, you need to edit one out of every three words to guarantee that you don’t get dinged.
On a 500 word paper, that means changing at least 125 words, no easy task. You can also rewrite and rearrange, but once again you have to do a lot of work.
The truth is that it’s much easier to just do an assignment legitimately than it is to try and cheat your way through it this way. You can automate the process using spinning software, but that just produces poor-quality work that needs to be rewritten by a human regardless.
In short, you either waste time or fail anyway.
4. The Essay Mill
One of the more common tricks for circumventing plagiarism detection is to purchase a paper from an essay mill. The reason this works, at least in theory, is because whatever you get from the essay mill should be original and not in the database of any plagiarism detection software.
As such, even though the plagiarist didn’t write it, the software won’t know better because it hasn’t seen the work before.
Why it Doesn’t Work: First, much, if not most, of the content from essay mills are themselves plagiarized. This is out of necessity. The rates essay mills charge and the turnaround times they offer and the breadth of subjects they cover make it impossible to write completely original work for every customer.
In short, there’s no honor among plagiarists.
But even if the essay is plagiarism-free and gets past automated dection, instructors are likely to feel as if something is up when they detect the change in writing voice and style. Humans are still the first line when it comes to plagiarism detection and one of those is not likely to be fooled.
5. Avoid the Database
Similar to the essay mill approach, other plagiarists have realized that detection software can only find what they’ve already seen, meaning that something not in their database can’t be found.
So some plagiarists spend a great deal of time studying what is and is not in the database and try to only lift content that’s outside of that checker’s scope.
Why it Doesn’t Work: You have no way of really knowing what is or is not in that database.
Not only do most plagiarism detection companies not reveal everything in their database, most have an element of unpredictability as to what is in there. For example, academic services routinely put all of the previous essays they’ve scanned. Some of the larger databases will also have an archive of the Web that can go back ten years or more, meaning things that aren’t online now may still be matched against.
You can never be 100% sure what is or what is not in the database and even probing it doesn’t always help. Not only do things change, often very quickly, but some services will only reveal their full report to instructors and editors.
The goal of plagiarism detection software isn’t to be an impervious barrier to plagiarists. In fact, it’s not really intended to catch plagiarists at all
Human beings, namely instructors, editors, supervisors, etc. are still the first and best line against all forms of cheating, not just plagiarism, and technology can only assist them.
A plagiarist’s best hope at escaping detection is not in trying to beat the softawre, but in hoping that the tools will be unsed improperly or not at all. Nearly all of the failures in detecting plagiarists come not from failed tools, but in people who did not do their jobs adequately and faulty systems that allowed them to thrive.
In short, the human factor is still by far the weakest link in detecting plagiarists and it’s likely to stay that way for a long, long time.