The tabletop roleplaying game (TTRPG) designer Monte Cook recently discussed a dilemma that anyone who is trying to build new TTRPG rulesets faces. On the one end, you can to maximize the opportunities for fun, while at the other end you also need to armour your design against players who will just try to exploit rules and twist the game to their whims, regardless of how that would effect fun at the table:
But offensive phrasing aside, it’s a decision that every designer’s got to make. For every table of great players interested in having fun, there’s often one player who insists on arguing about the phrasing of a rule in the book. This player will find (or attempt to find) a loophole or contradiction and exploit it to their own advantage. So you have to decide, do you design for that one player, or for all the others? And of course, for some tables and some games, every player is that one player.
You know the person. That guy. If you've spent some time playing TTRPGs in different groups, you have definitely come across the type. Even if it's a woman, it's that guy. As Monte Cook explains, designers who starts limiting their design to make it resilient against that guy are also sacrificing opportunities for fun, so there's a real design choice in not caring about him (or her).
There is an analogue in course design, where educators can focus on the many ways in which students could in principle cheat, plagiarize or otherwise sabotage their learning process. Like TTRPG players should arrive at the table to have fun, students should arrive in the classroom to learn, but course designers try to be resilient against those who will try to score easy points instead. And there's of course a reason for this – there is a subtle mismatch between the goal of a course (to have students learn) and the goals of learners (to learn and get a good grade), which can under some circumstances (e.g. student anxiety, high workload, misperceptions of norms) lead to fraud or plagiarism.
This is the case for only some circumstances and some students, but when we're designing assessment forms, we keep them in mind. We are basically creating assessment for that guy. By doing so, we run the same risk as game designers do, although instead of opportunities for fun we risk opportunities for learning. The classical example here is having students write an essay at home as a graded assignment. This is an activity that fosters reflection on a topic, allows a student to play with ideas taught in class and can reward critical and creative thinking. To my knowledge it is a valid and reliable measure of student competence, as long as you can be transparent about what the norms for good performance are.
Yet it is also a plagiarism risk. That guy would definitely exploit the liberty and depending on their disposable income would either hire someone to write the essay for them (there are dedicated Discord servers for finding brains-for-hire) or they would go for the softer version where they would write something and then have an accomplished student give feedback on it. In any case, that guy would game the assignment!
And so the course designer who cares about that guy has to outright remove the assignment or change it. Perhaps the essay should be written in-class, without internet access (thereby depriving students from important reflective and fact-checking opportunities) or the essay assignment is weighed trivially for the final mark (thereby incentivizing students not to do too much deep work for it). Whatever the solution, something is lost for everyone who came to the course to get as much learning as possible from it.
Whether that is a rational approach depends on a couple of things:
- Are there other ways to combat plagiarism and fraud? I think there are, but they are much softer and often more systemic than changing assessment forms. They include managing student anxiety, demonstrating the value of doing assignments and diminishing the importance of grades.
- How much quality in assessment forms are you willing to sacrifice for how much protection against plagiarism and fraud? Ideally, we'd formulate a destructive distrust index that tracks the gains in protective measures, as well as any losses for learning and assessment quality. A a critical cut-off point, the sacrifice is just too much.
- What is the base rate of plagiarism anyway? If it's a minor problem to begin with, you don't want the destructive distrust index to tip towards sacrificing assessment quality.
- Could actions against plagiarism and fraud actually motivate students to engage in plagiarism and fraud? Or put differently, would trust in students elicit trustworthy behaviour?
I can't say I have the full answers to these questions, but I have the strong suspicion we have been designing courses with fraud and plagiarism n mind much too often and for much too long. So why not try to not consider that guy for once and create assessment that works perfectly for the rest?