The Inner Workings of Plagiarism Detection Technology
There are a number of ways that technology can be used to identify potentially plagiarized content. This post examines the different ways, and how Turnitin uses search technology and content comparison algorithms to help educators help students learn how to use source attribution appropriately.
Plagiarism has always existed as a problem - the origins of the word date back to the 1st century. It's only of late, however, that plagiarism has become a significant concern not just for educators and researchers, but also in the public sphere. New instances of plagiarism seem to hit the news on a daily basis. Whether it's song lyrics, plagiarism by school officials, government ministers, speeches by political figures, or the plagiarism that happens in the classroom, incidents of plagiarism appear to be on the rise everywhere.
We have the internet to thank for that. With the rise of the internet, we've seen exponential growth of content created and made readily available, almost everywhere. The growth is happening on such a large scale that we don't even have a way to grasp how huge of a change in content creation we're witnessing. In 2013, factshunt.com pegged the amount of total internet content at 14.3 trillion pages (article). The growth is happening so fast, that we don't have a way to accurately determine the number of new pages created each day or the total amount of content that currently exists online. The best estimates suggest there are 47 billion indexed and searchable web pages (article). To put this number into perspective, it would take approximately 300 trillion sheets of paper to print out the entire internet, today.
With all of this information so immediately accessible, is it any surprise that we've seen a rise in plagiarism as well? Fortunately, the growth of the internet and our need to find ways to search that content has led to developments in web crawling and indexing technology (the latter of which is used to identify the content that is crawled) that has led to technology that quickly identifies copying and the potential plagiarism of content.
First off, it is important to clarify that plagiarism detection software doesn't specifically identify plagiarism. No software will ever be able to accurately determine intent. And intent is one of the factors that educators consider when looking at incidents of plagiarism in student work. The way that plagiarism detection software works is to identify content similarity matches. That is, the software scans a database of crawled content and identifies the text components and then compares it to the components, or content, of other work. Based on that comparison, the software will generate a report that highlights the content matches. Plagiarism detection software crawls and indexes content very similarly to the ways that search engines, like Google, crawl and index web content. The key difference here is that plagiarism detection software is crawling and indexing content not to make it keyword searchable, but to identify similar content stored in the database of crawled pages.
So how does the software scan for similarity?
There are, generally speaking, four different ways to go about doing this. The first way is through keyword analysis. What does that mean? Like a search engine, you enter in a keyword and the software scans documents to find instances of that word. Another way to scan text for similarity is to look at groups, or strings, of words. Rather than looking just at individual words, the software looks for strings or sequences of words (say 3-4 or more words ordered in such a way to create a sentence or sentence fragments). As you may be able to see already, these two approaches can be pretty effective for identifying the strict or exact copying of content within one document to others. The shortcoming of these approaches, however, is that it doesn't identify paraphrasing--where the ideas and meaning may have been copied--but the text is different enough that it doesn't get identified as a match.
A way to better get at this type of problem would be through a third way, which is to go about scanning for content matches by looking at the style of the writing and to compare that style to other documents. This is not a strict word-to-word analysis, but more of an approach that takes a look at the probability of certain word sequences ("phrases") that may appear in one document and then compare it to other documents. The challenge here is fine grained, word-for-word matches can get lost. Better yet, why not identify a document's unique "fingerprint," and then compare that fingerprint to others? This last approach, that we will discuss in this blog post, is what we largely do at Turnitin.
With "fingerprinting," Turnitin's technology scans and identifies the unique fragments and the ordering of word fragments that appear in a document. With this level of analysis, we can uncover word string matches ("fragments"), but also look at the unique sequences of those matches to create a fingerprint of the document.
Fingerprints are entirely unique and can be identified by the specific features displayed in a print. The same thing can be said for documents, each document has unique features such as phrasing, tone, style that if completely original is like a fingerprint, unique. If a document contains content that is unoriginal in its phrasing, the document will match to other document fingerprints that also contain this feature.
One issue this approach faces is how to avoid picking up very common words--like articles ("the," "an," "a") or conjunctions ("and," "but," "of")--and hone in on the strings of words that make a document unique. Fingerprinting gives us a way to exclude commonly-used words, while providing us with the ability to identify when content is poorly paraphrased. Because Turnitin was developed for use in academic contexts, our approach with fingerprinting is to focus on features of the text that are clearly relevant to the content or subject matter of the document.
For example, if you're finalizing your dissertation thesis you would want to make sure that all of the ideas you discussed were properly referenced and cited and that sections where you paraphrased were paraphrased properly. What gets less emphasized is strict word-to-word matches. If a more keyword search-biased approach were used here, we'd be unable to identify poor paraphrasing or selective word substitution--which incidentally is the majority of what academic and educators see in student work.
If you're looking for strict word-to-word matches, you could use a search engine (which is what everyone did before the advent of plagiarism detection software). If you're looking at comparing one author's style to another, there's an approach for that. As for identifying content matches in academic-type writing, Turnitin has developed a fingerprint-based approach that excels at finding the content that matters, when it matters. In other words, Turnitin is designed to support students who are learning how to use the internet to do research, use source materials, and take ownership of their own writing and ideas.
By David A. Tomar
For ten years, I made my living helping students cheat. I worked as a professional ghostwriter, completing homework assignments, producing essays, and composing senior theses for alternately desperate, lazy, or disengaged college and graduate students.
I worked as an independent contractor affiliated with various online paper mills and, between 2000 and 2010, spent nearly every day of my life immersed in academic research and compositional writing. Writing as many as 5,000 typewritten pages a year, I earned as much as many professors.
In November of 2010, I announced my retirement in a tell-all article published in the Chronicle of Higher Education. Using the pseudonym Ed Dante, I offered what was, for many, a first glimpse into the shadowy underworld of academic ghostwriting.
Since that time, I’ve had the opportunity to tell my story in various venues:
- my 2012 memoir, The Shadow Scholar: How I Made a Living Helping College Kids Cheat (Bloomsbury USA, 2012)
- my platform at the Huffington Post
- and my work with Turnitin.com, the International Center for Academic Integrity (ICAI), as well as other leaders and organizations striving to rebuild academic integrity.
Through each of these outlets, I’ve shared my experiences and drawn lessons for what these tell us about academic dishonesty, student cheating, the ghostwriting industry, and how we can collectively confront and redress these assaults on education.
This same purpose drives the present article, which gives a thorough state-of-the-art account of how ghostwritten papers may be detected and deterred. I offer this account as a resource to educators, administrators and schools who wish to better understand and more ably combat ghostwriting in their classrooms.
(For a companion piece by David A. Tomar, see The Ghostwriting Business: Trade Standards, Practices, and Secrets.)
Table of Contents
2 The Ghostwriting Business
Before it is possible to prevent and police ghostwriting, one must understand the industry. Though many educators are well aware of ghostwriting, how it happens and that it most likely has occurred in their own classrooms, just as many others have a limited or nonexistent sense of its impact.
Quite to the point, of the many reactions that greeted my original article in The Chronicle, doubt and skepticism were among the most common. Some truly dedicated, earnest, and otherwise astute educators refused to accept not only that wholesale cheating of this sort could be perpetrated but that it could be done so consistently and effectively without detection right under their noses.
I was actually a little shocked that they were so shocked. Too many educators were resistant to the idea (rather arrogantly, if I may editorialize) that a student could or would employ the services of a ghostwriter to complete a major assignment, a dissertation, or an entire course of study.
But I worked directly with students over the course of entire semesters to develop theses, respond to mentor feedback, make revisions, and ultimately produce final dissertations. I can confirm that this happens every day and that it’s easier to get away with than one might think.
We should first demystify this mode of cheating by analyzing it the way we would any traditional business, because, in most regards, it is a traditional business. Ghostwriting is legal, accessible and, with proper management, capable of driving a successful and sustainable company. Your ability to confront ghostwriting will depend on how well you understand the nature and norms of the business. What follows is a ground level overview of the industry.
We will also build on this discussion with a more comprehensive and self-contained analysis of the business in a separate account. Here, however, we begin with a brief overview of the ghostwriting industry.
It’s difficult to say for certain how many college paper mills are in operation today. But for a frame of reference, we look to the practice of medical ghostwriting, which is more readily allowed (read: overlooked) than standard academic ghostwriting, but which is nonetheless reviled by academics and their critics.
As with graduate and undergraduate ghostwriting, medical ghostwriting is an industry that operates partially in plain view and partially in shadow.
An article on its impact in medicine notes that “it is difficult to determine the extent of ghostwriting owing to both secrecy and a lack of research. Flanagin et al (1998) found that of 809 articles published in peer-reviewed medical journals, for 156 (19 percent) there was evidence that guest authors had not substantially contributed to the article, 93 (11 percent) included ghost authors, and for 13 articles (two percent) there was evidence of both.”
We have no way of knowing what percentage of students in any given academic context have submitted ghostwritten works but the findings above highlight the relative commonality of ghostwriting in one academic context (and a context reserved for elite academic performers at that). These findings also underscore just how difficult it is to gain a full perspective on the prevalence of ghostwriting in colleges and grad schools.
Still, we may be able to deduce a great deal just from the accessibility and ease-of-use of ghostwriting services. According to an article in the New York Times regarding rising rates of student cheating, “research has shown that a major factor in unethical behavior is simply how easy or hard it is.”
We can say with great certainty that it is easier than ever to employ an academic ghostwriting service. If a student has the money, he or she has the means.
The vast majority of students locate these services simply by doing a Google search for “Custom Paper Writing,” “essay help,” “term papers,” “homework services,” “essay writing services,’” or any number of other pertinent word combinations. Each of these terms will ultimately return dozens of pages of relevant search results.
These results provide us a glancing view of the online ghostwriting business but they also conceal layers of syndication; they obscure countries of origin; and they intermingle paper mills with legitimate education websites on the Internet landscape.
On the subject of legitimate sites, Coastal Carolina University provides us with a useful resource, though a relatively outdated one at this juncture. The site, last updated in 2009, offers a list of Internet paper mills containing several hundred distinct entities and their associated links. I can personally testify to the value of the site’s content as I have successfully used it as a resource for job hunting.
Today, I use it as it was intended, to help widen our knowledge of this industry. I would therefore call attention to a useful fact elucidated by the CCU site regarding the syndication or networking of ghostwriting sites. Though hundreds or thousands of links may appear for every search term entered, some of these sites will reflect overlap.
In many instances, multiple websites will function as part of a wider network. Each site will pull in orders through its distinct order form. Orders are consequently directed to a repository site to which independently contracted writers have access. One staff of writers and one set of umbrella policies may actually serve several dozen companies or sites.
For instance, ABC Papers, Amazing Papers, BookReportsRUs, Paper Campus, PaperResearch, and Paper Tutors all fall under the Paper Experts Inc. umbrella. CCU lists as many as 23 services affiliated with Paper Experts, though many appear to be defunct now. From this latter fact, we can also deduce that the shelf life varies considerably from ghostwriting service to the next. Admittedly though, this inference probably does not differentiate a ghostwriting startup company from any other web-based business.
By contrast, the best and most consistent paper writing companies are now more than a decade old. I can attest to the long-term survival of companies like custompapers.com (established 2000) and go-essays.com (established 2003), both of which employed my services in the past, both of which process many orders daily, and both of which still appear to be fully operational.
Ghostwriting companies also vary widely in size. Most rely on a stable of independent contractors. However, this stable can range from just two or three writers to thousands. For instance, at the time of writing, Crest Essays claims a database of 26,981 writers, of which 743 are listed as ‘active.’ All told, the service claims to have produced 167,594,264 words of material for paying customers, or over half a million typewritten pages (at roughly 300 words per page).
Crest Essays claims a 94 percent customer satisfaction rate but introduces its ‘About Us’ section by inquiring, “Are looking for help with your term papers, essays, research papers, and other coursework?”
So…I would take that unsubstantiated 94 percent with a grain of salt.
As we will explore in a separate and more exhaustive article about the paper mill business, the buyer must always be wary. Quality control can vary greatly from one company or writer to the next.
The point that we raise with Crest Essays is that its staff of writers and its general output are substantial. This is true of at least several dozen if not several hundred companies of comparable size. There is little in the way of a comprehensive scientific review of the industry, its size, and the economy it commands. The global nature of the business, the chosen anonymity of the customer base and the fact that at least some portion of vendors engages in only semi-legitimate business practices make it difficult to truly ascertain the reach of ghostwriting in education.
From what is immediately apparent though, we can conclude two things about the prevalence of ghostwriting:
- The inquiring student will find it easy to locate a desired service and begin using it; and
- The enterprising freelancer will find it easy to locate an employment opportunity and begin earning income from it.
In my experience, there are few freelance writing outlets where independent writers and unaffiliated clients are more readily, efficiently, and consistently paired with one another. In terms of prevalence, the qualitative takeaway for educators is basically this: It is extremely easy for students to cheat using ghostwriting services.
2.2 Pricing and Structure
Most companies operate using a similar pricing spectrum, charging between $10 and $50 per page depending on proximity of the deadline. For instance, Mypaperwriter.com prices its custom writing services between $17.55 and $45.85 per page. This is in line with the pricing spectrum and structure of the industry’s more lucrative companies.
The variance is usually determined by deadline. This is the measure used most frequently to define an assignment’s price. Papers due in a week or more are typically bound to the low end of the pricing spectrum. For anything due in less than a week, the cost per page will go up as the number of days goes down. A paper due in less than 24 hours will fall on the highest end of the cost-per-page spectrum.
Another feature that each of my past employers has in common is the use of a writer-manager or, in the case of larger companies, multiple writer-managers. Writer-managers serve a key role in the structure of most paper-writing companies, though responsibilities will vary depending upon the company’s operational model.
In nearly all cases, the writer-manager will serve as a liaison between the company and the independently contracted writer; and as a mediator in conflicts, disagreements or confusion between writers and customers. Given the linguistic challenges that so many ghostwriting clients face, the third of these occasions happens rather frequently.
Each company has its own way of processing orders and delegating assignments. The mode of delegation is usually a determining factor in the operational model.
Below is a brief outline of the three preeminent operational models. The models outlined here are drawn largely from personal experience. That said, most of the companies I’ve worked for are fairly representative of trends in the general marketplace.
A writer-manager will send assignments directly to a writer based on declared areas of expertise. Assignments will come with instructions, deadline, and amount of compensation. When a writer agrees to take on an assignment, he or she becomes responsible for submission by the deadline.
A writer-manager will send instructions to a select number of staff writers, often based on a writer’s area of expertise. Based on the deadline, length, and anticipated difficulty of an assignment, writers will quote prices for prospective assignments. Though writers create their own bids, relative market value still applies. Bids that overshoot the anticipated value are rarely matched with assignments. Another writer will likely quote a fairer price. When a writer’s bid is accepted, he or she becomes responsible for meeting the deadline.
2.2.3 Assignment Board:
This is the most sophisticated and efficient front- and back-end model. When an order is placed through a website, pricing is automatically determined per-page based on the length of time until deadline. The order is also automatically placed on a bulletin board visible to writers by way of username and password. This gives the writer an opportunity to visit the board and select orders at his or her discretion. Once a writer clicks the “Write It” button (or some equivalent), he or she becomes responsible for submitting the assignment by the deadline.
The ghostwriting industry enjoys a customer base comprised of three primary demographics. These are the likeliest perpetrators of ghostwritten plagiarism:
2.3.1 English Language Learners:
International students often arrive at American universities without a background or meaningful support in English composition. According to a 2003 study, these students lack the skills for paraphrasing and inferential use of sources possessed by their English-speaking counterparts. Many of these students will surmise, perhaps correctly, that they stand a far greater chance of getting away with cheating than mastering a new language.
International students comprise a ready-made source of revenue for the ghostwriting business. These students turn to ghostwriting services out of a combination of desperation and expedience.
2.3.2 Composition/Research deficient students:
A startling number American students—for whom English is a native language—will actually suffer from many of these exact same deficits. The viability of the ghostwriting industry shines a spotlight on the tragically overlooked prevalence of students at the undergraduate and even graduate levels who simply lack the skills or knowledge to produce university-level writing or research. These students turn to ghostwriting services out of desperation.
2.3.3 Lazy students:
Some ghostwriting clients simply lack the motivation and interest to complete their own work, a condition that Farnese et al. (2011) call “academic moral disengagement.” In many cases, a perfectly capable student will utilize an academic ghostwriting service as a way to cut down effort or improve his or her chances of receiving a better grade. In other cases, the lazy student may, in addition to being unmotivated, lack the necessary writing and research skills to complete the task at hand. These students turn to ghostwriting services as the path of least resistance.
As we proceed with the discussion of detection and prevention, it will be useful to consider that different strategies may apply to different demographics. We should recognize that there may, at times, be a distinction between students we might call “willful cheaters” and those we might call “reluctant cheaters.”
In most instances, members of the first two demographics described above will fall under the umbrella of “reluctant cheaters,” those who would prefer not to cheat but feel compelled to do so because they believe themselves to be in some way academically deficient or misplaced. In the conversation on prevention, the motivation to cheat is of considerable importance and can help educators to target students with preventative strategies.
As we will find, removing the sense of pressure or dread that comes with academic deficiency or misplacement may be an important part of contending with ghostwriting and with academic cheating on a more general scale.
3 The Ghostwriting Conundrum
Ghostwriting is hardly a new phenomenon nor is it one inherently dependent upon web proliferation. As I mentioned, I myself got started the old-fashioned way, taking orders from fellow Rutgers classmates and accumulating a growing independent business entirely by word of mouth.
However, the web has proliferated and simplified cheating, dramatically expanding the accessibility, visibility, and ease with which students can lift, recycle or otherwise claim authorship of work that is not their own. Consequently, the growth of this industry helped to provoke the growth of the plagiarism-detection industry of which Turnitin is a leading example.
Other notable sites include Viper, Plagscan, Plagtracker, Grammarly, Small SEO Tools, and Plagiarism Checker.
Turnitin represents the gold standard in plagiarism detection. Even so, given the limitations inherent in plagiarism detection, even Turnitin has no way to bring its extensive empirical data to bear on ghostwriting.
For instance a Turnitin (2012) White Paper on The Plagiarism Spectrum identifies “10 types of unoriginal work.” The statistics reveal that “Cloning” or turning in somebody else’s work as one’s own, and “CTRL-C” or copy-and-paste, account for the largest percentage of cheating students.
However, among the 10 types identified, ghostwriting is not included. This is not because Turnitin is unaware of the ghostwriting industry. To the contrary, the company has shown itself quite open and dedicated to better understanding the impact of ghostwriting. Still, where ghostwriting is concerned, educators lack a streamlined, technology-driven detection method like that which drives Turnitin’s originality detection software.
For many ghostwriting customers, this detection software actually provides a modicum of comfort, suggesting that the instructor may not be reading the work and simply be letting automated plagiarism checkers test for originality. Presuming the ghostwriter has created a wholly original piece, Turnitin software will not detect any irregularity.
Further, it is quite common for a customer to include a specification in their instructions that all completed work will be passed through Turnitin or a similar program. This means that the ghostwriter can often make practical decisions that will ensure no irregularities are detected. The motivation to do so is simply sound business practice and helps to facilitate repeat patronage.
The ghostwriter’s goal is to produce an assignment that will pass the student’s (often admittedly low) personal standards. But most ghostwriting companies also succeed on the strength of the return customer. Each company that I worked for provided the customer with a field in the order form by which to request a preferred writer. I would receive an automatic email notification telling me that a customer had requested my services.
I’ve even completed entire semesters, academic years, or courses of study for individual students across periods as long as two and three years. In addition to the implication that such repeat customers have clearly succeeded in passing ghostwritten work off as their own on multiple occasions, this repeat business denotes that ghostwriters are actively engaged in practices designed to evade detection.
These practices are what keep customers turning to the paper mill for help again and again.
With these conditions in mind, we point to a handful of detection and deterrence challenges that are unique to ghostwriting:
3.1 Original, non-plagiarized content:
Most ghostwriting companies are faithful to this service guarantee and will terminate independent contractors for failure to comply. Though the optimistic educator may take some comfort in the view that paper mills are not legitimate enough to constitute a threat, there are rules for professional paper writers and the more successful companies will enforce them.
Chief among these rules is the responsibility to provide completely original, never-before-used material crafted to respond to a specific assignment inquiry. This product is the cornerstone of this industry’s success.
3.2 Low likelihood of raising suspicion:
Ghostwriting places the onus on the educator to have initial cause for suspicion. This requires the individual grading a written assignment to sense a disconnect between the student and the assignment, which of course requires some initial familiarity with the student in question.
There are a great many educational contexts in which this is difficult if not impossible. When I began my ghostwriting career at Rutgers, there was no mystery as to why the semi-coherent, fraternity-bound, budding alcoholics who used my services were able to pass my work off as their own.
Courses are routinely taught in lecture halls containing hundreds of students; written assignments are often reviewed by teaching assistants; and any personal interaction between professor and student is strictly optional. Naturally, students who plan on defrauding the professor are distinctly less likely to pursue a personal relationship with him or her.
These challenges are compounded even further as a consequence of distance learning, online education and for-profit education (the last of which I mention here both because much of it is exclusively online and because ghostwritten cheating abounds in this context).
Learning contexts like this demonstrate just how unlikely it really is that a professor will have cause to question the authorship of an assignment once it has passed through plagiarism detection software.
That said, ghostwriting is not confined to the web class or giant lecture hall. I was also often hired to write in direct response to professor feedback, especially in post-graduate contexts. This suggests that some students are not dissuaded from employing ghostwriting services in more intimate settings, even as intimate as one-on-one interaction with an academic mentor.
3.3 Difficulty of translating suspicion into proof:
Cheating is, of course, a serious allegation and students have a lot riding on the completion of their education. So obviously, the average student will go to great lengths to deny any such allegations. Students are not afraid to get litigious if need be. The point is, as an educator, one must be very careful about levying the accusation without hard evidence.
This is true regardless of the chosen method of cheating. What differentiates ghostwriting is that, true to its name, it leaves a far less visible trail. Assuming that an instructor has cause to suspect an assignment, he or she now has the unenviable obligation to demonstrate guilt based on a well-informed and probably correct suspicion. The process is an uphill climb.
Plagiarism detection software or even a quick Google search can verify suspicion of a cut-and-paste job and yield incontrovertible proof. But when it comes to ghostwriting, unless the student accidentally leaves the receipt attached to an assignment (which I wouldn’t completely rule out as a possibility), the likelihood of producing evidence is low. Part of the discussion of detection and deterrence must logically revolve around finding ways of accumulating concrete proof based on suspicion.
Taken together, these challenges render many traditional methods of combating plagiarism even flimsier when applied to ghostwriting. This means that most professors have almost certainly graded ghostwritten material without ever realizing it. There are perhaps as many others who have graded ghostwritten materials with some level of suspicion but without the means, institutional support or emotional energy to translate this into proof.
The challenges specific to ghostwriting call for an approach that is not merely reactive (i.e., triggered only by suspicion of cheating) but proactive as well. That is to say that educators and institutions can take certain preemptive steps that both discourage cheating by way of ghostwriting and place educators in a more advantageous position to spot questionable work and acquire meaningful proof of its third-party authorship.
4 The Four D’s of Ghostbusting
I’ll introduce this section with a well-traveled and apocryphal anecdote that I remember hearing in my college days:
A college student spends an entire semester goofing off. He doesn’t show up for a single class. It’s in a 500-person lecture hall so he figures nobody will miss him.
Naturally, he shows up for the final exam with all the answers written on his arm. He finishes the exam and brings it to the front of the room. As he places it on the top of the test pile, the professor looks down at him from the lectern. He observes the cheat-sheet scrawled on the student and demands to see his arm.
The student responds indignantly: “How dare you! Don’t you know who I am!?”
The professor admits that he does not. So the student shoves his exam in the middle of the stack and quickly departs.
There is a lot about the system of higher education, as it is today, that makes cheating easy. The goal of the present account is to help educators create a campus atmosphere where cheating is neither as easy nor as desirable.
This calls for a combination of reactive and proactive measures, an encompassing approach that broadens and gives nuance to the fight against cheating.
There are far-reaching educational issues implicated by the viability of the ghostwriting industry. To reiterate the challenges facing students with learning or language deficiencies, policing alone does not get to the root of the problem. Some very real structural, practical (and even ethical) flaws in our educational system are reflected in these demographics and in students’ decisions to cheat.
The goal of this strategy, therefore, is to take on cheating both as a violation of academic integrity and as a symptom of some broader problems within our educational system. This underscores the 4 D’s Strategy, which confronts ghostwriting and its underlying causes through Design, Deterrence, Detection and Dedication.
Design refers to the way a professor constructs assignments, course materials, tests, classroom time and the semester-long curriculum. This is an area in education where quality control runs the gamut from excellence to criminal incompetence. There are plenty of professors who work tirelessly to tailor assignments, materials and examinations to remain in-step with constantly evolving subject matter, student culture and best practices. But there are also plenty of professors who recycle old materials without scrutiny and who depend wholly on text-based content which most students could acquire without professor mediation.
It’s not hard to guess which of these approaches is more vulnerable to cheating. With those professors who take the latter approach, students may find cheating not only easier but also more susceptible to rationalization. The attitude is essentially, “If the professor doesn’t care, why should I?”
According to a study from Teaching Sociology (2012), strategic design is one effective path to mitigating the risk of plagiarism. The approach that an instructor takes when creating assignments, course content and materials will have a direct bearing on a course’s susceptibility to ghostwriting.
An article compiled by the New Zealand Qualifications Authority (NZQA)(2013), which provides an overview of cheating prevention strategies in place in New Zealand’s institutions of higher learning, confirms this assertion.
The NZQA advises instructors to “vary (or rotate) assessment tasks from year-to-year and course-to-course. One of the most common forms of cheating involves submitting work produced by students in previous courses. Relatively subtle changes to assessment tasks can be enough to alert markers to cheating. Case studies, contexts, data sets, and actual items can be changed while assessing the same outcomes.”
When I worked as a ghostwriter, lazy students helped me to make my living but it was the lazy professors that made my life easier. The task of pretending to be a student in somebody’s class is greatly simplified when the professor takes no special steps to differentiate the course, its content, or its assignments from the many millions of other courses that have been taught on the same exact subject from time immemorial.
In other words, if I am assigned the same five page paper asking the same three or four general questions about Plato’s Republic that I’ve been assigned and asked four dozen times in the past, I can presume the professor is taking as much care to grade the assignment as he or she did to write it…that is to say, none.
Even just 20 years ago, the risks of reusing the same basic assignment or exam were modest. There have always been on-campus repositories of completed assignments available for student purchase. However, the accessibility of these repositories can’t compare with the information that can be gotten online. This is not to suggest that the ghostwriter will struggle to complete an assignment that is otherwise unique and original in nature. It’s only to point out that the less original and unique an assignment, the less likely that an instructor will have any way of differentiating the work of a student who’s been in the classroom from that of one who has not.
To say it simply, a generic assignment begets a generic essay. If this is all an instructor seeks from his or her students, said instructor makes it nearly impossible to differentiate between the work of a pupil and the work of a person who has never set a foot in the lecture hall.
If, by contrast, one designs materials, assignments and exams with thought, care, and specificity, one has much better odds of spotting the work of an outsider.
4.1.1 In-class writing:
Partial emphasis on in-class writing exercises, when supplemented by out-of-class assignments, is a powerful way of getting to know students’ writing capabilities and voices. Class time should be used to challenge students with unique and fun writing exercises.
There are a lot of ways to integrate this strategy into an existing curriculum. Assignments can be formal and woven into the grading structure or they can simply be loosely defined writing activities designed to promote critical thinking. The important thing is that these assignments will require students to write thoughtfully and uniquely on course-relevant subjects, providing the instructor with readily verifiable writing samples throughout a semester.
No matter how convincingly a ghostwriter writes on a given subject, this approach provides a document whose authorship is not in question as a point of comparison.
4.1.2 Multi-draft process:
For assignments of greater depth, a balance between in-class and out-of-class draft-writing offers a measure of oversight even as the student pursues an assignment independently. The caveat with this strategy is that there may be nothing stopping the student from outsourcing each and every draft of the assignment.
However, using the multi-draft process can stretch an assignment out across weeks or months. This results in a greater length of exposure for the cheating student. Instead of the once-and-done security of getting away with a single ghostwritten assignment, each student knows that his or her work will be held up to sustained and ongoing scrutiny.
By inserting one-on-one conferences into this draft process, the instructor can heighten this scrutiny by requiring each student to defend the approach, argument, and decisions comprising the written work.
This is also an opportunity to challenge students to craft unique arguments based on course content.
4.1.3 Personalization of subject matter:
Assignments that incorporate personal experiences and interests not only offer students a welcome reprieve from the rote, repetitive, or regurgitation-based work that makes up so many courses, they also make it more difficult for the ghostwriter to assume a student’s identity. This challenge may even strain the credibility of submitted assignments to the point of making them more detectable.
Professors might be shocked to learn that many of the customers I’ve served have requested personal essays without providing any personal information. In other words, the assignment instructs the student to write about a life-changing experience and the customer leaves me, the ghostwriter, with the power to define this life-changing experience. Knowing one’s students on a personal level might, in this case, provide more than enough information to peg suspicious assignments.
Naturally, not every subject or discipline lends itself to personalization. But in many of the liberal arts subjects, where cheating is decidedly rampant, there is an opportunity to create assignments that correlate to the everyday lives, individual cultures, geographical backgrounds, ethnic makeups, personal histories, and extra-curricular interests of one’s students.
Most ghostwriters have no problem with a little creative writing but, again, this approach gives an educator the chance to spot a mismatch between a student and the written work.
4.1.4 Original course materials:
Just as experienced ghostwriters become accustomed to seeing the exact same assignments over and over, they also get accustomed to seeing the same sources and course materials again and again.
Instructors can reduce the ease with which an outsider can replicate old assignments by updating course materials and creating new assignments on an annual basis. Ghostwriters rely on materials that are readily available online and prefer the efficiency of assignments that can be summoned from memory and experience.
Creating challenging course material that is not drawn verbatim from standard texts can serve as a distinguishing feature when it comes time to read students’ work.
4.1.5 Emphasis on class discussion:
In addition to creating unique course materials, a course should be designed to distinguish the in-class experience from the experience of reading about something on Wikipedia.
Assignments that rely strictly on standard texts make the ghostwriter’s job very easy. Most texts are readily available online. By contrast, a lecture, a class discussion and the experience of being a part of both should be something unique and impossible to replicate. Nobody’s saying every professor has to be Robin Williams in Dead Poets Society, but every professor should give his or her students something they can’t get on the Internet.
Naturally, if they can get it on the Internet, so can any ghostwriter. The goal is obviously not to eliminate, supersede, or overlook the standard texts. The classics are important. But differentiating what happens in a classroom and calling for written work that reflects this differentiation do create a genuine challenge for the ghostwriter.
4.1.6 Mixed-format testing:
Predictable testing methods (i.e., Scantron) make it easier for students to bypass the writing portion of a course altogether. Multiple-choice, machine-graded exams help the student differentiate between reporting to class on test day and outsourcing take-home writing work.
Merging essay-writing and test-taking removes this impression. Instructors can create mixed format tests that include multiple-choice, essay questions, short-answer questions, etc. This approach requires the student to demonstrate writing capacity during in-class testing.
Once again, this approach results in the creation of a written document that one can be certain originated from the student in question. This forms a good basis for comparison with suspicious take-home work and—insofar as it also might give the student pause before submitting ghostwritten work—also makes an excellent segue into our discussion on deterrence.
Deterrence refers to ways of diminishing the inclination, motive, or desire to purchase a ghostwritten paper. A 2011 Research in Higher Education study, seeking to understand the motives behind student cheating, finds that cheating is often rational in nature.
That is, students at least believe that they are cheating out of ease, normalcy, or necessity. The study finds that the onus falls on instructors to live up to certain student expectations regarding clarity and engagement of course content. The study identifies this as the best route to deterring the rationalized impulse to use a ghostwriting service.
In other words, deterrence should revolve around strategies that make cheating seem less rational. The NZQA suggests that one way to deter cheating is to be in a state of constant evidence-gathering. The NZQA notes that such evidence can extend from written work to a mix of formative and summative assessments as well as informal ongoing observations. Researchers also suggest having “a mix of assessment methods” that includes the appropriate combination of uncontrolled assessments (such as take-home essays), controlled assessments, oral presentations, supervised composition and “cross-checking assessment events.”
The article points out that this would serve as a beneficial way to help TAs or graders identify writing inconsistencies even without their having the opportunity to know the students personally.
The NZQA points out that while strategies such as one-on-one conferencing may seem time-consuming, this approach could actually occupy less time than setting, proctoring, and grading exams.
This research is compatible with my experience to some extent. I have also generally observed that students are more motivated to cheat in contexts where professors aren’t particularly motivated to teach. Students know when a professor is disengaged from the subject, the class, or the general profession of educating our youth. Those who are already inclined to do so, specifically those who fall into the ‘lazy’ demographic of potential ghostwriting targets, may take the professor’s cue and decide to disengage as well.
The Honor System is a useful way to ensure that students understand what is expected of their work in terms of originality. But it makes for a poor deterrent on its own. The customers who contact ghostwriting services out of laziness or a feeling of being overburdened by their various personal and academic responsibilities are generally not moved by the ethical quandary of cheating.
When they express concern to the ghostwriter about the process of purchasing and receiving a ghostwritten paper, all of this concern focuses on the fear of being caught. While many of the suggestions below revolve around student engagement as a way of deterring cheating, they work best when backed by sound detection strategies such as those outlined in the subsequent section.
Individualization of the educational experience can instill in the student a greater sense of commitment to course materials and to the knowledge and career opportunities thereby implied. Large lecture halls and online courses can create a sense of anonymity for the would-be cheater.
Instructors can remove this sense of anonymity by creating opportunities where students can advance along individual research paths while sharing the same larger educational experience. Strategies include greater research subject flexibility, more opportunities for independent study and the chance to complete work directly relevant to a chosen career path.
Students will be less inclined to employ the services of a ghostwriter if they believe that what they are researching will actually lead to pertinent knowledge and skills with value beyond university halls.
One thing that large universities and online courses have in common is that, if one desires, one can go an entire semester without ever once personally meeting a professor. There is comfort in this anonymity. Removing this comfort creates a deterrent that does not otherwise exist.
Strategies such as mandatory and regular one-on-one conferences can create a relationship between student and professor that can’t be avoided. With respect to online courses, videoconferencing applications are now readily available and standard for laptops, home computers, and smartphones. By building conferencing or videoconferencing into the semester, the instructor fosters a relationship with his or her students that will function as its own deterrent.
4.2.3 Emphasis on in-class participation:
Mandatory class participation heightens the imperative for students to become familiar with course content. Mandating contributions to class discussions gives students a strong incentive to establish a consistent voice and perspective on course subjects.
Class discussion helps to create a connection between the student and certain ideas, perspectives, and ways of expressing oneself on a given subject. The more intensive and exciting the in-class discussion, the harder it will be for students to divorce themselves from the ideas stated in class.
As a ghostwriter, I have completed countless assignments on controversial social issues like gun control, abortion, affirmative action, and the War on Terror. In every case, the student is likely to have a well-defined position and some fairly personal reasons for that position. By creating a safe educational context in which these positions can be shared, explored, defended, and held in civil contrast with one another, the instructor can deter the student from outsourcing the expression of these positions when it comes time to write. This approach forces students to own the opinions and ideas that will ultimately form their written work.
As a bonus, this approach is also more than likely helping students to refine their ideas in ways that make them better prepared to handle their written assignments.
4.2.4 Student engagement:
This one is really and truly up to each individual educator. It is within every educator’s power to be as creative, energetic, inspiring, original, unpredictable, and engaging as he or she wants to be. When the professor demonstrates passion for the material, this helps to create a moral dilemma about cheating that has more to do with the student/teacher relationship than with the notion of academic integrity. In reality, this relationship weighs heavier on the conscience.
Many students feel no remorse about cheating in a course from which there is a feeling of disengagement. Uninspired lectures, standard texts, and generic assignments serve as great ammunition for a student who wishes to rationalize his or her detachment.
These conditions also help to reinforce the commonly held conception among students that education revolves around grading rather than learning. If this conception is true (and in the least inspiring classrooms, it usually is), it really doesn’t matter how a student comes by his or her grade so long as it is a good one. Educators create a deterrent to this attitude by being their best selves in the classroom, during their office hours and beyond.
4.2.5 Miscellaneous strategies of deterrence:
- Course discussions where students are invited to share research experience and knowledge
- Professor lectures based on and attributed to content drawn from student assignments
- A requirement for students to occasionally present research findings or other written work to the class or professor
In each instance, the student has added cause to be intimately familiar with his or her subject and is therefore deterred from handing the subject off to a ghostwriter.
Detection is both a manual process driven by professorial experience and a technology-driven process with continued room for growth and improvement.
In either case, detecting ghostwritten materials requires familiarity with each student’s subject knowledge and writing style. In a study regarding authorship identification of online messages, Zheng et al. (2006) note that “People have different habits when organizing an article. These habits, such as paragraph length, use of indentation, and use of signature, can be strong authorial evidence of personal writing style.”
This denotes a detection strategy where a student’s portfolio of writing comprises a body of work to be used as a diagnostic tool. The NZQA points out that “seeing a collection of a student’s work within one portfolio will make inconsistencies obvious. It also enables managers/tutors to set tasks that rely on personal situations, unique data, etc.”
The NZQA study describes this practice as “triangulating evidence from a range of sources,” the goal of which is to spot irregularities.
Though it is tempting to think of a future where this level of manual detection is no longer necessary, the reality is that no degree of automated detection will replace the need for professorial vigilance. Regarding this point, we consider a “Tip Sheet” produced by the Center for Educational Resources (CER) in 2006, which offers an overview of Turnitin.com.
The Tip Sheet notes that “Turnitin.com does not provide a judgment of plagiarism; it provides data. It uses a set of algorithms to match text and, based on the results, produces an ‘originality report’ that shows the percentage of similar text and the source(s) of the match. But faculty must review and interpret the report.”
Even with the best detection software, human analysis is required. Only the critical eye of the professor will be capable of determining if academic dishonesty has occurred.
The goal is to find instructive and innovative ways of merging human analysis and technology-mediated assessment to detect ghostwriting.
In a research article entitled “Fifty Ways to Detect a Ghostwriter” (which we wishfully presume was inspired by Paul Simon), the author assesses a wide range of potential methods for identifying ghostwritten work.
Among the study’s recommendations, the approach that yielded the most compelling findings was the author’s investigation of linguistic similarities across a sample set of assignments. The author notes that “the best way to catch a ghostwriter is to compare writing styles in all essays” and notes that most students offer up certain distinguishing characteristics in their writing. Among them, the research identifies “frequency of words and short phrases consisting of at most five words; frequency of most frequent verbs; and frequency of conjunctions.”
Researchers suggest that these tendencies toward dependency on certain verbiage and phraseology can serve as fairly reliable indicators of authorship. The researchers aspire to take detection a step further, observing that though these methods of detection are useful for calling out a cheater in the classroom, “the professional outsourcer was never caught in the net.”
Accordingly, the researchers suggest that the work of specific professional ghostwriters could itself be stamped with distinguishing features. The study concludes by noting that “we have already started the creation of a plagiarism tool intended to integrate student essays with search engines, Google Translate, and the pool of previous essays. The tool will be soon enlarged with ghostwriter detectors. We do hope that it will discover the cheat and reduce it to the level of previous years.”
Kennesaw State University also identifies a study that is currently underway entitled “Challenges and Possible Solutions to Using Statistics to Detect Ghostwriting: A Preliminary Investigation.”
The study’s proposal points to the scholarly use of statistical analysis to determine authorship of historical documents and manuscripts. The study acknowledges the challenge of applying this strategy to the decidedly more modest bodies of work created by individual students.
The goal of the study is to determine how to adapt these analyses to smaller writing sample sizes. According to the proposal, “this study takes a preliminary look at how short text length and limited writing samples affect the attribution process. It is hoped that even as a precursor to a full-scale study, this project will be able to suggest steps that maximize the reliability of ascription of ghostwritten works.”
What differentiates customers of ghostwriting services from other cheaters is that they are willing to go the extra mile to avoid detection. Turnitin and other plagiarism-detection services have all but eliminated the threats of copy-and-paste, duplication, and other modes of self-directed online cheating. Of course, students attempt all of these methods anyway. But educators do have a resource at their disposal that absolutely works.
Students who use ghostwriting services know this quite well. As I noted earlier, assignment instructions often include the specification that Turnitin will be used. This is the equivalent of hiring a jewel thief and arming him with the vault’s security schematics. The ghostwriter will compose an assignment with every intention of evading this security.
But of course, the ghostwriter can only control what goes on outside of the classroom. It therefore falls on the professor to use in-class time wisely.
4.3.1 Assignment exit interviews:
Standardizing one-on-one conferencing with each student following assignment-submission requires each student to defend his or her writing. This is an especially attractive approach because it need not revolve around the suspicion of cheating. This healthy exercise can simply serve as a way of helping the student to reflect on the content of an assignment and the process involved in its completion.
That said, the approach would also serve inherently as a microscope fixed on the student’s work. Inconsistencies between the ideas, voice and knowledge expressed in the written work and in the interview will serve as red flags and as a point of entry for a more prying investigation.
4.3.2 Manual literary fingerprinting:
Of the many strategies outlined in this account, this may well be the most readily adaptable to any context where writing forms a portion of the coursework.
Here, the orientation process for any writing-intensive course will begin with an in-class writing assignment. This sample will serve as a literary fingerprint for each student to be used for the purpose of matching against voice, diction, and other distinguishing features when evaluating future assignments for signs of ghostwriting.
In addition to being adaptable to any number of learning contexts, this is a strategy that could also produce meaningful evidence of cheating or, at the very least, shift the burden of proof to the student.
Not to be overlooked is the opportunity that this strategy also offers to identify students who struggle with writing and to provide them with access to the support they need. More on that subject in the section below on “Dedication.”
4.3.3 File properties:
One way to improve the chances of detecting ghostwritten work is to simply be a savvier user of technology than the average cheating student. It’s easier than one might think.
Most instructors already require students to submit materials digitally. Certainly this is true for educators teaching online courses and for those who use plagiarism-detection software. By opening a student’s assignment in Microsoft Word and clicking on “Properties” or “File Properties,” the average educator might be surprised at the information that is readily available.
This should include information about the file’s author, the registered user of the machine or program on which it was created, the date it was created and some history on modifications made to the file. All of this information constitutes real, concrete, and usable evidence in making a case against a student suspected of cheating.
4.3.4 Computational literary fingerprinting:
Based on the effectiveness and value of Turnitin.com as a strategy for plagiarism detection of the non-ghostwritten variety, this strategy may best predict the future of ghostwriting detection.
According to a study published in the December 2013 International Journal of Computer Science and Network, stylometric [O1][BT2]software may be an effective tool in author identification and authentication.
Stylometry is a method of author identification with a long and sometimes problematic history in literary scholarship. Its use has historically pertained to authorship attribution for ancient texts and classical works of literature whose authorship is in question.
However, a compelling study from 2013 notes that a forensic interest in online authorship attribution has given new purpose to stylometric analysis. Researchers point to two major distinctions in the detection of online authorship: the error-prone and unstructured nature of online content and the brevity of most samples. The study’s focus on stylometric author identification through the 140-character medium of Twitter demonstrates a focus on redressing both challenges.
With improved data accuracy such as that sought in the above-noted study, Stylometric software could be the key to creating an algorithmic literary fingerprint for every student.
This fingerprint, molded by each individual’s unique tendencies in terms of phrasing, word-choice, and diction, could be used as a yardstick against which a student’s subsequent assignments are measured. This is an approach that could also be readily standardized as part of the assignment submission process, as is already the case with Turnitin’s detection software.
This would add a layer of detection against ghostwriting that does not yet exist. Such technology is largely in the research and development phase today but, once market-ready, could significantly ease the burden of detection placed upon the educator.
Detection is all well and good, but let’s face it, people good at detection are more likely to join a police force than a teachers union. Teachers are in the classroom to teach. This is where the fourth “D” comes into play. The instructor must be dedicated to the education of his or her students, not just to punching an academic time card.
It is the professor’s job to take a group of finely tuned intellects, honed through years of painstaking secondary school training, and stimulate them to think critically, to plumb the depths of their own assumptions and to emerge with new and nuanced perspectives.
Yeah … If only it were that simple.
Far too many students show up in the classroom or lecture hall without even the basic fundamental writing skills needed to compose a Tweet let alone plumb the depths of anything. Therefore, it becomes professor’s job to be more than an academic, an intellectual, a researcher, or an author. It becomes the professor’s job to be an educator.
Here I need to recall the existence of “reluctant cheaters.” For these students, deterrence is not just difficult to achieve. More than that, it invokes the idea of bringing a knife to a gunfight. For most, the fear of getting caught is substantially less than the fear of failing, mostly because the latter seems likelier than the former. No amount of threatening can stop a desperate student from taking the last resort.
There is a perspective today, fostered by scandals at silver-spoon academies like Stuyvesant High School and Harvard University, that privileged students are cheating more than ever before. And that is almost certainly true. But we should take a moment to acknowledge that their privilege doesn’t necessarily preclude them from the same challenges facing so many other struggling students.
Day in, day out, my work as a ghostwriter brought me into contact with students who truly and completely lacked the capacity to communicate through the written medium. I know. I corresponded with them via email. The writing deficiencies that students carry into college and beyond are nothing short of disturbing.
Some of them were wealthy students in reputable universities. Some of them were lower middle-class Phoenix University enrollees who were probably using some portion of their student loan money to pay for my services. What they had in common was that they couldn’t write, which was in many cases also an indication that they lacked any number of additional learning and research skills required to complete assignments on their own recognizance.
Students who are deficient either in learning or language face tremendous pressure to stay afloat in educational situations they simply aren’t qualified to handle. To deter them from hiring ghostwriters, it is necessary to do more than scare them. We need to help them.
4.4.1 Identify struggling students and see that they get help:
These are the students who are by far the most likely to employ a ghostwriter. In order to reduce the presence of the ghostwriter in the classroom, educators must take preemptive steps to identify those who are most likely to need his services.
Addressing their struggles reduces the single biggest motive for academic dishonesty. It is important to connect struggling students—graduate and undergraduate alike—to needed academic support. It is no longer reasonable to take for granted the idea that a student who has somehow advanced along the educational continuum to arrive in a university classroom must inherently have the academic skills needed to succeed there.
Instead, we must presume that struggling students are far more vulnerable to the temptation of cheating. If we’re being fair, we must also presume that most struggling students would rather not cheat but feel that they have been left with no suitable alternative. Finding that suitable alternative could be a game-changer for many would-be ghostwriting clients.
Alternatives may include the following:
- English Language Learner (ELL) support
- Writing assistance
- Research training
- Personal/Mental Health Counseling
The strategies I outline here are not one-size-fits-all. Higher education exists in all shapes, sizes, media, and models. It falls upon each educator to synthesize the strategies outlined here with their own ideas to create an approach that best suits a given educational context.
One’s approach will be determined by class size, content, subject matter, level of education, feasibility, institutional support, and one’s own personal discretion and creativity.
Of the many things that distinguish ghostwriting from other methods of cheating, there is no magic bullet that can be used to kill it. Instead, this resource aims to promote an educational strategy that weakens the ability of ghostwriters to do their job, that diminishes the appeal of ghostwriting services for students, that creates a more keenly aware educator and that takes a curative rather than punitive approach to students who are in need of help.
There is an impetus here throughout to promote learning strategies that are more varied, distinctive, creative, and enriching and which also foster a more familiar working relationship between students and educators.
The combination of these strategies will heighten the ability of educators to discourage and detect ghostwriting while also minimizing the factors that motivate students to use such services.
Many of these strategies are labor-intensive.
Many demand that the instructor spend more time working on course materials, interacting with students, and becoming familiar with students as writers and as individuals.
But these strategies come with their own rewards.
Even if the motive in taking any of these steps is to detect or deter ghostwriting in the classroom, the outcome will be a more positive and engaging educational experience for educators and students alike.
Bhargava, M.; Mehndiratta, P. & Asawa, K. (2013). Stylometric Analysis for Authorship Attribution on Twitter. Lecture Notes in Computer Science, 8302, p. 37–47. Ret. http://link.springer.com/chapter/10.1007%2F978-3-319-03689-2_3
Binongo, J.N.G. (2014). Challenges and Possible Solutions to Using statistics to Detect Ghostwriting: A Preliminary Investigation. Kennesaw State University: Center for Excellence in Teaching and Learning. Ret. http://cetl.kennesaw.edu/incentive-funds-research-and-creative-activity-jose-nilo-g-binongo
Bosch, X. (2011). Exorcising Ghostwriting. EMBO Reports, 12(6), 489–494. Ret. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3128288/
Brent, E. & Atkisson, C. (2011). Accounting For Cheating: An Evolving Theory and Emergent Themes. Research in Higher Education, 52(6), p. 640–658.
Center for Educational Resources (CER). (2006). Deterring and Detecting Plagiarism with Turnitin.com. www.cer.jhu.edu. Ret. http://www.cer.jhu.edu/pdf/turnitintips.pdf
Farnese, M.L.; Tramontano, C.; Fida, R. & Paciello, M. (2011). Cheating Behaviors in Academic Context: Does Academic Moral Disengagement Matter? Procedia-Social and Behavioral Sciences, 29, 356–365.
Farnworth, D. (June 7, 2013). The Brutally Honest Truth About Ghostwriting. Squawk. Ret. http://squawk.im/content-marketing/truth-about-ghostwriting/
Finn, K.V. & Frone, M.R. (2004). Academic Performance and Cheating: Moderating Role of School Identification and Self-Efficacy. The Journal of Educational Research, 97(3), 115–121.
Heckler, N.C.; Forde, D.R. & Bryan, C.H. (2012). Using Writing Assignment Designs To Mitigate Plagiarism. Teaching Sociology, 41(1), 94–105.
Kimbel Library. (2009). Presentations, Cheating 101: Internet Paper Mills. Coastal Carolina University. Ret. https://www.coastal.edu/library/presentations/mills2.html
Lakshmi et al. (2013). A Study on Author Identification Through Stylometry. International Journal of Computer Science & Communication Networks, 2(6), 653–657.
New Zealand Qualifications Authority (NZQA). (2013). Effective Practice In Preventing and Detecting Academic Fraud. www.nzqa.govt.nz. Ret. http://www.nzqa.govt.nz/assets/Providers-and-partners/NZQA-Effective-practice-guide.pdf
Perez-Peña, R. (Sept. 7, 2012). Studies Find More Students Cheating With High Achievers No Exception. New York Times. Ret. http://www.nytimes.com/2012/09/08/education/studies-show-more-students-cheat-even-high-achievers.html?_r=0
Tomar, D. (Nov. 12, 2010). The Shadow Scholar. The Chronicle of Higher Education.
Turnitin.com. (2012). The Plagiarism Spectrum: Tagging 10 Types of Unoriginal Work. iParadigms, LLC. Online at http://www.turnitin.com/assets/en_us/media/plagiarism_spectrum.php
Yamada, K. (2003). What Prevents ESL/EFL Writers From Avoiding Plagiarism?: Analyses of 10 North-American college websites. System, 31(2), 247–258.
Zdravkova, K. (2011). Fifty Ways to Detect a Ghostwriter. University Ss. Cyril and Methodius in Skopje, Macedonia.
Zheng, R.; Li, J.; Chen, H. & Huang, Z. (2006). A Framework for Authorship Identification of Online Message: Writing-Style Features and Classification Techniques. Journal of the American Society for Information Science and Technology, 57(3), 378–393.
1. Bosch, 2011.
2. Perez-Peña, 2012.
3. Kimbel Library, 2003.
5. Yamada, 2003.
6. Finn & Frone, 2004.
7. Farnese et al., 2011.
8. Turnitin.com, 2012.
9. Heckler et al., 2012.
10. NZQA, 2013.
12. Brent & Atkinson, 2011.
14. NZQA, 2013.
16. Zhang et al., 2006.
17. NZQA, 2013.
19. CER, 2006.
20. Zdravkova, 2011.
24. Binongo, 2014.
27. Lakshmi, 2013.
28. Bhargava et al., 2013.