The Cheater’s Dilemma
To cheat or not to cheat. Students at universities across the nation are faced with the decision of whether or not to uphold academic integrity. During in-person exams, personal morals—along with the repercussions of getting caught—are typically strong enough to deter the vast majority of students from cheating. Furthermore, in an in-person test environment, students can have confirmation that their peers are not cheating; when looking around a room, a student can visibly see whether or not their peers are taking their respective tests honestly.
When the in-person environment switched to online exams in dorms or at home during the pandemic, students’ mindsets dramatically changed; the only incentive to uphold academic integrity was students’ personal morals. It is well known that for many students, this was not a sufficient incentive, and more students than usual broke their vows of academic integrity. This phenomenon can be modeled as an application of the Prisoner’s Dilemma.
The conditions of online exams are quite similar to those present in the Prisoner’s Dilemma. Let us consider a class where only two students—Student A and Student B—are enrolled. During an exam, cheating is defined as looking at notes when not allowed by the professor. A student may select one of two strategies: cheating or not cheating. It is also important to note that the professor assigns a student’s grade relative to their performance against the other student; the better student A does, the worse Student B will do and vice versa.
If Student A and Student B are both honest and do not cheat, then they score the same on the exam and both receive a B. If one student cheats and the other does not, then the student who cheated does significantly better on the exam and receives an A. The other student now does comparatively worse and receives a D. Lastly, if Student A and Student B both decide to cheat, then the professor knows that cheating occured and gives both students a C.
When Student A and Student B are deciding whether or not to cheat, they must consider what the other student will choose as well. Like in the Prisoner’s Dilemma, the two students are isolated and have no way of knowing what the other student will choose (this also differentiates online conditions from in-person ones, where a student could visibly see if their peers were cheating or not). When looking at all scenarios, it is clear that both students have the dominant strategy of cheating; whether Student B decides to cheat or not, Student A always has higher payoff when Student A chooses to cheat.
This simplified scenario of online test-taking can be broadened to fit all online classes at universities. Most classes are graded on the basis of a curve, meaning that the mean or median grade of the class is assigned a letter grade, and the rest of the grades are assigned accordingly. That being said, during online exams, students often assume that their peers are cheating and using their notes. As a result, that student often also cheats in order to prevent themselves from being put at a disadvantage. This thorough understanding of why students cheat during online exams is important to developing a solution to the cheating problem; if all students whole-heartedly believed that their peers were not cheating, then there would be less of an incentive to cheat. Finally, another way to minimize cheating would be to change the grading scheme by making classes not median based and instead letter grade cutoffs based on a student’s raw score (this has already been done in many of my Cornell classes). This removes the incentive of cheating by making your grade reflective of your own coursework rather than that of other people.