title: Is plagiarism rational? A case study in Computing author: Module 2 / 33347617 mainfont: Helvetica Neue linestretch: 1.5 papersize: a4paper geometry: margin=1.5in header-includes: \renewcommand{\abstractname}{Declaration}

abstract: '\noindent By submitting this, I declare it to be my own work. Any quotations from other sources have been credited to their authors. Approximate word count: 1150'

At the Designing Learning Landscapes conference, Prof. Andrea Sella included in the course of his presentation on “My Learning Landscapes” a tangential remark that (paraphrased) plagiarism might be a rational response from students to the situation academics put them in. This claim, even as an aside in a highly entertaining keynote, is shocking: are we, as academics, complicit in the construction of an environment where one of the cardinal sins of academia – plagiarism, or passing off others’ work as one’s own – is effectively encouraged?

This case study considers Sella’s statement in more detail, in the context of the Computing discipline. In particular, we will examine both halves of the claim: the situation that academics and the University system construct for their students, and whether severe academic misconduct is a rational response of students to that situation.

One particular aspect of the Computing discipline is its multiple audiences, as undergraduate study aims to be both direct vocational training for the craft of computer programming (and related jobs such as system administration) and a precursor to the study of computation and computational thinking as a ‘pure’ research area – while of course also aiming to address all the points between these two extremes. For those seriously intending to pursue academic study of computation, there is unlikely to be a problem with plagiarism, as a serious intent to study is largely incompatible with submitting the work of others as one’s own without understanding – and with understanding, there is no need to plagiarise.

For the more pragmatic undergraduate, for whom a degree in Computing is perhaps seen as the ticket to a lucrative career (or more prosaically an entry-level job) – and 66% of undergraduates in general are in this category (Newstead et al., 1996), so likely more in Computing specifically given its vocational aspect – the temptation of plagiarism is more clear. In particular, if the degree qualification is seen by the student as the end, rather than the learning during (and after) the course of the degree, then the student will attempt to minimize resources spent on obtaining the degree (the degree’s opportunity cost) in order to spend those resources elsewhere, for example to learn other skills, earn money, or even enjoy their relative youth.

The particular situation which makes this risk most acute is when the academic system emphasizes learning which is seen by the student as not relevant for their future. One danger in the discipline of Computing, with its vocational and academic components, is that the teachers of the material, by and large, will have focussed on the academic aspects of the discipline – necessarily, since they have chosen the academic path. Without a clear and convincing narrative, which may not be easily articulable, it might be that the curriculum can be seen by students to fail to address their needs as apprentices in the craft of Computing, hence reinforcing the perception of the degree as a sequence of hurdles to overcome and effectively incentivizing shortcuts, if those shortcuts can be taken with impunity.

So based on the above narrative, we can see that there is danger in the Computing discipline that the academic situation does not discourage plagiarism for the majority of students; this is without even considering the pressure of repaying student debt or funding a habitable existence in inner London. The question of whether plagiarism is a rational response to the situation, though, involves an examination of the cost/benefit trade-off, or (more generally) the expected utility of a slate of possible actions: which involves a consideration of the likelihood of detection and the effect of possible sanctions against the individual.

Computing as a vocation has a peculiarity in that the major fruits of labour in the application of the discipline have general applicability and minimal cost of reproduction: computer programmers, and workers in related jobs, are encouraged to construct reusable software artifacts whose source code can be distributed with effectively zero marginal reproduction cost. Hence, the only barriers to reuse of products are those of intellectual property law, particularly copyright (protecting the reproduction of expressions) and patents (protecting against the reproduction of inventions).

Practitioners in the software industry have constructed ways of working with such reproducible artifacts, such as Open Source licences and community knowledge-exchange websites; a substantial amount of the modern programmer’s work – typical for a graduate in Computing – will involve reuse of existing code, configuration and binding to Open Source libraries, and troubleshooting by tapping into communities such as StackOverflow; these are part of the life of the Computing professional.

Given this background, it is worth considering in detail a particular case of plagiarism detected this year: a finalist submitted, for their 60 CATS project, work that derived substantially from material in three 30-minute video tutorials freely-available on YouTube. While the student did write an original report, the submission as a whole was clearly plagiarised, the student attempting to pass off design decisions in the tutorial as the result of his own careful experimentation. The decision to penalise the plagiarism, once it was discovered and proved, was (in my opinion) completely valid, as the dishonesty was rampant; one can only speculate what the pressures on the individual were to make the plagiarism seem like the best option. However, one aspect bears some consideration: what outcome would the student have achieved if the submission had been totally honest and transparent about its reuse of available material, particularly given the working practices in industry?

Given the limited amount of original content in the student’s project, even with full referencing of external sources it would have been impossible to argue for a passing grade for the project. It would probably have been considered a valid attempt, however, which in general might have allowed for a retake over the Summer – except that this student had also failed to make a valid attempt at one of his other modules, and hence would have had to resit in the following academic session in any case. Given that, it is possible that the calculation made was that attempting to pass the project through plagiarism had a higher expected utility than either attempting to pass honestly or from honestly documenting an insufficient attempt, and was hence rational: the worst case (the plagiarism attempt being caught) was not substantially worse than the situation the student was already in.

How could this situation be improved? The combination of the credit framework (graduation and late Summer retakes impossible without a valid attempt) and the weak expected sanction for the first detected case of plagiarism (a mark of 0 awarded for that module) leads to the situation where plagiarism is not discouraged; either making the Summer retakes generally available, or communicating clearly to students (Carroll, 2002) that even first offences of plagiarism can have significant sanctions and following through on that threat, would probably discourage plagiarism of this sort sufficiently.

Carroll, J (2002) ‘A Handbook for Deterring Plagiarism in Higher Education’, Oxford Centre for Staff and Learning Development

Newstead, S E, Franklyn-Stokes, A and Armstead, P (1996) ‘Individual differences in student cheating’, Journal of Educational Psychology, 88, pp 229–241.