Deterring, Detecting and Tracing Plagiarism

RETURN
edited 9/25/05

Discouraging and detecting plagiarism is an unpleasant business. Instructors don't want their classroom to turn into a security state. But at the same time they know that, given half a chance, some students will not be honest. And they certainly don't want principled students doing the work, while their unprincipled classmates steal the same grade. What is to be done?

Deterring Plagiarism

Let's begin by considering the student perspective. For them things can seem a bit confusing. Disciplines require different types of citation. What one discipline considers "common knowledge" another does not. What one instructor requires, another waives. What one prefers, another abhors. Such variations, particularly when coupled with misunderstandings and inexperience, can lead to actions that look like plagiarism, yet are quite innocent.

With this caution in mind, here are some tested methods for deterring plagiarism.

General Procedures

Preventing plagiarism is clearly preferable to having to deal with it after it happens. Here are some general procedures that deter plagiarism.

Make sure students know what counts as plagiarism in your class. (The Plagiarism Book helps with that.)

Indicate in writing the penalties you are prepared to assess for plagiarism. But make sure you don't go out on a limb that might get sawed off.

Note:     Before pre-announcing penalties, consider that unknown consequences offer a different kind of deterrence: "uncertainty avoidance."

Imagine that students maintain files of your past tests and take the necessary precautions.

Enforce the same due dates in multi-section courses and be wary of granting extensions. James Brown (http://www.yorku.ca/admin/cst/pla.html) reports that one of his students repeatedly asked for deadline extensions. The reason? He was secretly waiting for a friend who had a different instructor in another section of the course to get her paper back. Brown's student stalled because he didn't want his plagiarized paper and the original to simultaneously be in the hands of instructors.

Let your students know that you are aware of the cheat sites on the Web. (The Plagiarism Book cautions that you have been provided with detailed information on these sites.) Initially, we worried about acknowledging such sites even existed. But the thousands upon thousands of hits these sites record suggests that students know about them already.

Assignment Sequence and Design

Here are assignment sequencing and design tips that deter plagiarism.

Have students write an early in class assignment, grade it, and keep it on file. (This provides a baseline to compare with work done out of class and discourages nascent dishonesty.)

Require a topic proposal, rough draft and, possibly, an interim working bibliography. (Keep in mind that if they've "shopped" early and already have a paper they plan to plagiarize, they can work backwards from it to create these things.)

Require specific application of things taught. (Suppose, for instance, you taught them to distinguish between informal, formal and technical levels of culture. Assign a paper that requires them to put these distinctions to use.)

Require personal reflection. (Students can still tack that on at the end of something they plagiarized; but at least they have to write something and you might pick up the difference in styles.)

Require narrative or descriptive writing, rather than argument or exposition. (This limits the usefulness of free term paper files because they tend not to stock these types of papers.)

Require students to list where they found their references. (Did you find The Idiot's Guide to Successful Alchemy in our library?)

Referencing Strategies

Referencing strategies can be a significant plagiarism deterrent. Try some of these.

Offer students practical advice on keeping track of their sources while doing research. Because of its volatile nature, this is particularly important to students using the Web. Remind them to note the sources of their facts and ideas as the go along. Returning to find things you've decided to use is often difficult.

Require distinctive references in papers. You can, for instance, require at least one Web source, one source from journal X, and so forth. (The more distinctive you make things, the less likely a free Web paper is to fit.)

Require an annotated bibliography. (This will nonplus "scholars" planning to use free papers from the Web.)

Require students to photocopy their references, highlight relevant sections and turn them in with the assignment. (This is a highly effective tool but you get an awful lot of attachments to look through.)

Detecting Plagiarism

Here are some of the better detection techniques we gleaned from the web, interviews with educators and personal experience. Some might be obvious to veteran instructors, but not to novices.

Look for uncommon fluency, or varying fluency, in a paper. That can be an indicator that the work was copied in whole or in part.

Be alert for vocabulary that seems too advanced for that student or for technical language the student is unlikely to know.

Note:     We know a student who was accused of plagiarism because she used "virtuoso" in a middle school paper. The teacher discovered, the hard way, that the student was a music academy student and something of a virtuoso herself.

Be suspicious of papers with sections that don't seem to match. It might be a patch job.

Double check doubtful papers by seeing if they use non-web based references that are unavailable locally. (A librarian who is death on academic fraud shared this one with us. It seems obvious now, but we never thought to check it.)

Be dubious of papers with a referencing style different from what you require. If you asked for APA, but got Chicago it might be because that was the style used in the purloined paper. Remember, though; write for hire shops customize reference styles.

Tracing Plagiarism

To nail a plagiarist you often need the original he or she counterfeited. Finding that can be quite a challenge. You might get lucky and remember where you originally saw the work in question. That's happened to us. But it usually isn't that easy. Here, then, are a few tracing strategies.

The Web not only facilitates plagiarism; it aids tracing it too. One of the simplest ways to trace something is to use the search engine at www.google.com by entering a string of words from the suspect text proceeded and followed by quotes. Google's vast library of documents often makes short work of the inquiry. We also recommend using a meta-search engines such as www.metacrawler.com www.dogpile.com or www.profusion.com to look for a distinctive text string such as a characteristic phrase or misspelling. This works surprisingly well.

Heyward Ehrlich, Rutgers University, Newark Campus, http://newark.rutgers.edu/~ehrlich/ says check on plagiarism from subject discussions in listservs and newsgroups on www.dejanews.com.

Note:     Even if you use meta-search engines, try more than one. Each of them indexes different search sites.

Ehrlich also recommends checking electronic encyclopedias both on-line and CD based. (He observes that Encarta was given away to millions of Microsoft customers as Microsoft Bookshelf. Many school libraries also purchase rights to Britannica online or a similar resource.)

Compare the student's "work" with the original texts of listed references. (Surprisingly, it's not unusual to have students reference their plagiarism.)

Get a top-notch librarian to help you track the suspect paper down. (This really works if you have access to as competent and hard working a librarian as we do.)

Try a bluff. Tell the student that you know it isn't their work and demand to know where it came from. However, you might get bizarre results. Darsie Bowden, in the English Journal tells a tale that might inhibit this practice. An instructor suspected that a student copied her paper on anorexia from several popular girls' magazines, though he could not find a source. So he tried a bluff. Under pressure the young lady admitted to academic thievery. However, it turned out she was no such thing. Her paper was really her own personal story, her "confession" a measure of her insecurity.

Unfortunately, none of this will work if the plagiarizing student has paid to access special parts of cheat sites that are available only for a fee. Search engines cannot access these files without paying the same fee, and they aren't going to do that. So it's nearly impossible to trace student larceny if the have "invested" in a purloined paper.

Phony References

Here we're not dealing with plagiarism but something that goes hand-in-hand. Students use inaccurate and concocted references to lend an air of authority or an appearance of substantiality. The trouble is, bad references are often difficult to detect because of the labor involved. Here's a suggestion for dealing with that.

The chart below summarizes a procedure by which pre-verified good and bad references are given to students to check out. The instructor can use this as a gradable activity since he or she knows which references are good ones. By incorporating into this activity references from actual student papers, the students themselves will end up doing your checking for you while all the time not knowing whether it is merely part of the activity or an actual reference verification. The educational justification is that it is a means to teach library research procedure and care in scholarship.

The first step is to develop a file of verified references. These might include actual samplings of texts - some deliberately rewritten with mistakes in them, such as the wrong year or edition indicated.

The set is broken up and a few items given to each student for them to verify. Providing each student with only one reference to check might be sufficient. You could duplicate entries so students would act as a check on each other. The initial difficulties to be overcome would be compiling the original list and regular updating. (This manual provides an initial set of real and mistaken references for your consideration.)

Our thought is that if several instructors adopted this procedure, each could provide references for the other to check. Certainly if students knew that reference checking occurred from time to time, this would discourage carelessness and dishonesty.

Reference-Check Procedure

Materials	Procedures	Educational Justification	Comments
A file of verified references	To be used to verify researchers skill and honesty	To teach research procedure	Initial compiling is a task. Update regularly
Research sheets containing source options, etc.	Given to research as task	Class or homework assignment	A simple common form will do
A rubric for plagiarism	One of the judgments to be made about reference	Teach meaning of plagiarism	Will need consensus among instructors working together.
References to be tested	Taken from real research papers	To acculturate to "reference-honesty."	Can be used as service among instructors.

These are the best tactics we've found to date. Of course, prevention is preferred. Even when you prove plagiarism beyond doubt, dealing with it is messy and unpleasant. Besides, if precautionary measures keep students on the straight and narrow, they just might end up learning something worthwhile. That's far less likely if they become cyberspace patrons of something like The Evil House of Cheat.

Summary

Few educators enjoy flunking plagiarists or hauling them before disciplinary boards. You often can avoid such unpleasantness with preventative strategies. Those we've just reviewed can be thoughtfully brought together into potent formulations that usually make sanctions unnecessary.