Max. Classroom Capacity: What Should We Do About Cheating?

Loren J. Naidoo, California State University, Northridge

Dear readers,

By the time you read this the FIFA World Cup will have just finished—as I write, it hasn’t started yet. Congratulations Brazil, France, Argentina, Spain, Germany, or England? I love watching soccer, but World Cup games can be frustrating when players engage in excessive “simulation” and “embellishment.” Simulation is the fancy soccer term for pretending to be tripped, kicked, head butted, or assaulted with a deadly weapon by one’s opponent when little or no contact has occurred. Embellishment is the fancy soccer term for the protracted rolling on the ground, clutching of the ankles, and mimicking of a 2-year-old’s temper tantrum, which sometimes involves the team’s medical staff running onto the field and spraying the player’s injury with, well, something (WD-40? Drakkar Noir? Eau de la colère?) and always ends with the player’s miraculous and complete physical recovery mere moments later. Question: Are these players cheating? One might say that trying to fool the referee is a fair part of the game. However, if such shenanigans cost your team a chance to advance in the tournament, then you may feel differently about it.

To paraphrase South Park elementary’s Mr. Mackey: “Cheating is bad, mmmkay.” Yet cheating inevitably is a part of sport, work, and education. As with the VAR (video assistant referee) in soccer, we now have technological tools to help us identify and perhaps reduce cheating in our classes. I first tried Turnitin, the plagiarism detection software, a little more than a decade ago. I had a writing assignment I had used for the past several years, and I wanted to try Turnitin as a way to discourage future cheating. I was also curious to see how much cheating had gone undetected in prior classes. I uploaded all 736 of the papers that past students had submitted. I was shocked to find that two were exact matches—one student had submitted a paper written entirely by another student the year before, changing only the name on the paper! How dare they cheat and in such a blatant fashion! Insolence! Impudence! I was experiencing that “hot and burning” sensation memorably described by Dr. Robert Bies (2001, p. 90) in his seminal chapter on interactional (in)justice.

As I noted in a prior Max. Classroom Capacity about teaching ethics in I-O psychology, we are living through an epidemic of ethical scandals in multiple domains (education, business, politics, sports—I’m looking at you, FIFA!). I’m certain we can all quickly bring to mind many examples of prominent people cheating and getting away with it, at least for a time. The sheer absurdity of wealthy celebrities paying gobs of money to cheat their kids into exclusive universities, despite the existence of many high-quality, lower cost institutions that probably do as good a job (or better) at providing students a diverse and useful education. Do these highly publicized ethical scandals tacitly tell our students that everyone is cheating; therefore, they should cheat too? Do our students think of cheating in school in the same way that soccer players think of fooling the referee, as part of a game? And what should we, as teachers of I-O psychology, do about it?

Before we get into the details, let me quickly say that I will focus on only cheating of the relatively unambiguous variety—a student copying another’s work wholesale, for example. I won’t discuss more ambiguous cases, such as a failure to cite others’ work properly, the line between student collaboration and cheating, or the use of artificial intelligence-based text writing software (as discussed in another prior Max. Classroom Capacity column).

Like anyone who has been teaching for a few years, I’ve encountered multiple incidents of student cheating. Each time, my initial reaction was to (a) feel angry/betrayed/disappointed, (b) identify how the student had been able to cheat, and (c) consider actions to prevent future students from doing the same thing. Early in my career I also heard stories about students cheating in my colleagues’ classes and took what I thought to be appropriate actions. For example, a colleague told me about a student who wore a baseball cap to a closed-book exam with notes written on the inside of the brim! As a result, during exams I started asking students wearing baseball caps to turn them around so the brim faced backwards.

In sum, in response to cheating I generally focused on trying to prevent future cheating. There are lots of ways to do this (see top half of Table 1). We can (a) increase our efforts to detect cheating on written work by using software like Turnitin, which identifies word-for-word matches in phrases in students’ work compared to a large database of work submitted by students in the past. During in-person exams we can diligently patrol and scrutinize our students’ every move, admonishing them to keep their eyes on their exams. We can (b) change the assessment context to make cheating more difficult. For example, during online exams we can use software like LockDown browser, which prevents students from changing windows, copying and pasting, accessing other apps, taking screenshots, printing, and so on. We can restrict students’ access to exam questions and answers after the exam is complete. We can (c) increase the salience of punishments for cheating by reciting the institution’s academic integrity code and the dire consequences of violating it. We can also (d) change our assessments to prevent or reduce cheating by using different assignment prompts or exam questions each semester; by varying the order of questions and response options on exams; and by creating difficult, timed or closed-book exams that make it difficult to look up answers using outside sources. However, it is likely that students who are sufficiently motivated to cheat will find new and innovative ways around whatever preventive measures you adopt. Moreover, although widespread cheating can undermine the validity of assessments, some of the aforementioned solutions can have the same effect. For example, using closed-book rather than open-book exams may reduce the predictive validity and reliability of the assessment by increasing the role of memorization and recall in determining results, whereas in most work contexts, workers are allowed (encouraged, even) to review references and notes when performing their jobs.

Table 1

In all candor, I’ve tried most of these cheating prevention approaches at times in my career, and perhaps they are all reasonable things to do in isolation, depending on the circumstances. Some of these approaches entail a greater investment than others. For example, creating new exams or assignments, and their associated grading rubrics, each time we teach a class would entail an enormous amount of work. In reality, most tenure-track faculty have limited time to devote to teaching (as opposed to research and service), and it seems inadvisable to allocate a large chunk of that time to developing ways to thwart the (likely) very small minority of students who cheat. Even those instructors who focus primarily on teaching may wonder whether their time and energy would be better invested in measures designed to maximize student learning as opposed to reducing cheating.

Many of these cheating-prevention approaches may have significant hidden costs as well, including the effect of infantilizing students by treating them like naughty children who cannot be trusted, and may reduce students’ trust in their instructors and institutions. Discussions of academic integrity often focus on the validity of students’ grades, fairness toward fellow students who do not cheat, and high-minded principles of trust and honor. In my experience, it is rare indeed for instructors or institutions to acknowledge their own roles in the incidence of cheating. For example, if institutions create “gatekeeper” courses with large enrollments for the purpose of “weeding out” substantial numbers of students who are seeking entry into a major, for example, and these classes use high-stakes, high-pressure assessments (e.g., a single final exam worth 100% of a student’s grade) that are perceived as unfair and deliberately tricky, and for which students do not feel they receive the support they need to perform well, then students will have more reason to cheat. Indeed, if the primary purpose of the course is to weed out students, then there may be perverse incentives that may lead instructors to design assessments on which many students will do poorly and to provide as little guidance to students on how to perform well. In this mindset, student cheating would undermine the validity of the assessment and downstream gatekeeping decisions and therefore must be stopped. However, to my mind, the primary problem with cheating does NOT concern measurement validity. To my mind, we should be concerned about cheating because it represents a student’s choice to prioritize their class performance over their learning, which we know from the achievement goals and goal orientation literatures is bad for learning and many other desirable outcomes (e.g., Elliot & McGregor, 2001; Payne et al., 2007).

I started this column by describing my outrage at discovering that two of my students had cheated. In retrospect, this was completely the wrong reaction. Although I was “in my feelings” about the two students who blatantly cheated, what I failed to realize in that moment was that the vast majority of students (99.7%, to be exact) had NOT cheated (at least, not in a way that could be identified by Turnitin). In fact, from an academic integrity standpoint this assignment was a big success. I believe the main reasons for students’ lack of cheating were that the paper was highly personal—it required students to observe humans interacting and interpret their behaviors using theories from social psychology. They had considerable latitude to decide what interactions to observe and what theories to write about. Students tended to enjoy the assignment. The paper was also a relatively low-stakes assessment—worth 25% of their grade—where none of their other assessments were high stakes either. In sum, students likely were less motivated to cheat on the paper. Introducing new punitive measures designed to prevent cheating would be unlikely to produce any tangible reduction in cheating as the base rate was so low to begin with yet could undermine students’ enthusiasm for learning by emphasizing compliance with anti-cheating rules.

Now when I consider measures to address cheating, I try to ask myself the core question that I think we all should be asking ourselves as educators: How can I maximize student learning? I humbly offer three categories of approaches designed to cope with cheating based on an orientation that prioritizes student learning (see the bottom half of Table 1). First, use multiple low-stakes assessments so students do not feel as much pressure to perform on any single assessment, reducing their motivation to cheat. Yes, it may be more work to develop multiple low-stakes assessments and rubrics, but this seems like a better investment of your time than developing complex means of detecting and preventing cheating. Second, when it comes to making decisions and building policies around cheating, involve students in the process. This might include initiating a discussion with students about what does and does not constitute cheating on an assessment, negotiating with students the conditions under which certain assessments happen (e.g., Will exams be open book?), and discussing the standards or expectations regarding citation and collaboration on written work. Following such discussions, I now ask students to read and sign an academic integrity agreement, but I don’t require them to do so. Recently, I administered an exam via canvas online to my in-person class where I had chosen the option of displaying only one exam question at a time, without being able to return to prior questions, in order to make it more difficult to cheat. However, in a discussion with students afterwards, many described how stressful and discouraging this was for them. We agreed that on the next exam I would remove this constraint with their pledge not to cheat and my expectation that I could effectively deter cheating by simply walking around the room more during the exam. Similarly, my students asked if they could be allowed to refer to their notes during their next exam, which I agreed to as well. I saw no evidence that cheating went up, but students’ experiences were much more positive, and the quality of their answers increased. Third, build learning assessments that are more personally meaningful to students. Provide students with as much choice as possible regarding assessment topics. Build assessments that go beyond simple memorization of definitions or concepts by requiring students to apply concepts to real-world issues in ways that mimic what they may do in their professional or personal lives. Incorporate other job-relevant assessments such as presentations, group discussions, and business simulations that students may find more engaging and valuable than traditional academic assessments like multiple-choice exam and research papers.

I’m all for incorporating reasonable measures to prevent and reduce cheating, but let’s make sure we don’t throw out the baby with the bathwater. We are educators—remember the GOOOOOOOOAL!

Readers, as always, please email me with comments, feedback, or just to say hi! Loren.Naidoo@csun.edu.

References

Bies, R. J. (2001). International (in)justice: The sacred and the profane. In J. Greenberg & R. Cropanzano (Eds.), Advances in organization justice (pp. 89–118). Stanford University Press.

Elliot, A. J., & McGregor, H. A. (2001). A 2 × 2 achievement goal framework. Journal of Personality and Social Psychology, 80(3), 501–519. https://doi.org/10.1037/0022-3514.80.3.501

Payne, S. C., Youngcourt, S. S., & Beaubien, J. M. (2007). A meta-analytic examination of the goal orientation nomological net. Journal of Applied Psychology, 92(1), 128–150. https://doi.org/10.1037/0021-9010.92.1.128