Turnitin and Rhetoric

What do you think about Turnitin, the plagiarism detection service? Apparently quite a few people don't like it. A little less than a year ago, Charlie Lowe, Ellen Schendel, and Julie White on Cyberdash wrote,

Scholars and teachers in the field of Rhetoric and Composition, as well as other academics, have discouraged the use of plagiarism detection services.

Now, one would think that those trained as rhetoricians and compositionists would have good arguments. However, except possibily for one argument, that of intellectual property, their position is based mostly on unwarranted assumptions and faulty logic. I'll save the intellectual property argument for a later post and look at their other claims here.

According to the Cyberdash folks, two reasons not to use Turnitin are:

  1. "Turnitin Can Be Ineffective for Detecting Plagiarism"
  2. "Turnitin Discourages Good Pedagogical Practices Concerning Writing"

That word "can" is a nice sleight-of-hand term. Scholars use it because it lowers the strength of the claim sufficiently so that it can't be refuted. The major problem with the first claim, however, is its warrant: "anything that can be ineffective for detecting plagiarism shouldn't be used." By this logic, instructors shouldn't use Google or Yahoo (two search engines mentioned by these same scholars) for detecting plagiarsim because they, too, "can" be ineffective. Even the teaching practices they espouse "can" be ineffective. No single approach to detecting plagiarism is 100% effective, and so it's an unenlightening truism to say that a particular method can be ineffective.

Under the first claim, the authors write:

In “Defining and Avoiding Plagiarism: The WPA Statement on Best Practices,” the Council of Writing Program Administrators urges teachers to “use plagiarism detection services cautiously,” for they should “never be used to justify the avoidance of responsible teaching methods.” We recommend that teachers work toward implementing the WPA’s best practices as a long-term solution to eliminating plagiarism and building a culture of responsible participation in the creation and circulation of academic knowledge.

Certainly, Turnitin should be used with care, and certainly "responsible teaching methods" should be employed. Who would disagree with these assertions? But note that the authors have used values we agree with to reduce a complex situation to an "either-or" fallacy of opposing good teaching practices with the use of Turnitin. More appropriately, Turnitin should be considered as one tool that, when used in conjunction with other approaches, such as Google or Yahoo, along with "good" teaching practices, statistically increases one's overall effectiveness in detecting plagiarism and in teaching how to avoid plagiarism.

On the second claim, the authors specifically claim that turnitin

emphasizes the policing of student behavior and texts over good-faith assumptions about students’ integrity, and can shift attention away from teaching students how to avoid plagiarism in the first place.

So many fallacies are entangled in this sentence, it's difficult to know where to begin. On the first point, on "policing," many composition professors require photocopies of sources to be turned in with the papers to catch plagiarism. That common practice, which is used to help students with the process of writing, often has the additional purpose of deterring/catching plagiarism (i.e., "policing"), according to what I've gathered in casual conversations with other instructors. More importantly, the claim of "policing" depends entirely on the context and classroom environment. It is not the tool that emphasizes policing, it is the instructor and the environment established by the instructor. Of course, I can imagine that some instructors would establish a policing environment with Turnitin. If Turnitin is used appropriately, however, this claimed "emphasis" simply does not exist. Rather, using Turnitin appropriately is a valuable tool for teaching students "how to avoid plagiarism."

With respect to students' integrity, until I started reading about the "problems" with Turnitin, I assumed, based on my experience, that perhaps 5% of my students intentionally cheated or plagiarized. Since then, however, I've learned plagiarism and cheating is prevalent.

Donald McCabe, a professor at Rutgers, has conducted extensive research on cheating. In the last 4 years, his research has surveyed more than 50,000 undergraduate students and 18,000 high school students and has found that 70% of each group of students has admitted to cheating, that levels of cheating have been increasing over the last 4 decades, and that most students do not consider Internet plagiarism to be a "serious issue." In one study by McCabe,

Thirty-eight percent of the undergraduate students completing the survey indicated they had engaged in one or more instances of cut & paste plagiarism using the Internet in the past year - paraphrasing or copying a few sentences of material from the Internet without citing the source.

In an article by Emily Sachar, McCabe has also found that of more than 5,000 graduate students, 56% of MBA students and 47% of non-business students had cheated, noting,

You're asking kids to be honest about their dishonesty. So that suggests that kids are, if anything, underreporting their cheating activity.

To assume integrity when the evidence clearly indicates otherwise is unwarranted and naive. Even so, using Turitin has nothing to do with assumptions about students' integrity, unless, as a few commenters have said, someone also would claim that using metal detectors at airports assumes that all travelers are terrorists.

The assumptions being made reflect the ideological positions of those making them rather than reality. When I stayed in the hospital with my wife during the birth of our daughter eight months ago, every time I took my daughter from the nursery, the nurses always checked my ID bracelet and hers to see if they matched--even though I had accompanied the nurse with my daughter to the nursery! There was no assumption of my being a guilty baby snatcher. Instead, the environment was one of safety and security for the babies. That's the sort of environment our classes should have--not one of students being guilty of plagiarism but rather one of protecting the integrity of the grades of those who do their own work. In fact, some students see Turnitin in this manner (e.g., Aditi Banga and The Crimson staff).

On "shift[ing] attention away from teaching students how to avoid plagiarism in the first place," no evidence of this shift is provided here or anywhere else I have read. Now I can imagine teachers who are not tackling plagiarism to begin with might overrely, even soly rely, on Turnitin. However, it's difficult to imagine that instructors who have "effective" practices on avoiding plagiarism will stop teaching plagiarism avoidance simply because they started using turnitin. The authors' claim that "Turnitin Discourages Effective Pedagogical Practices" is, at best, no more than speculation.

Although these compositionists mean well, their claims are based not upon research but on faulty rhetoric that is driven by emotion not reason. Rather than assuming uncritically that using Turnitin projects an image that students are guilty or encourages bad teaching practices, it would be better to provide guidance on how to use Turnitin in a positive way. I'll post a little on that after I look at the issue of intellectual property in my next post.

Update: Responding to my assertion that "can" is "sleight of hand," one individual responded that this is a normal hedge in academic writing. Yes, it is. Generally speaking, evidence exists for a claim. Perhaps the evidence suggests that something occurs sometimes, frequently, or in some other manner. However, when there is no evidence for a claim, as in this case, then it's simply sleight of hand.

Related articles:
Turnitin and Intellectual Property
Using Turnitin
Turnitin Bibliography