Defending Assessment Security in a Digital World

Preventing E-Cheating and Supporting Academic Integrity in Higher Education

Phillip Dawson

• Publisher: Routledge
• Year: 2020
• Print ISBN: 9780367341541
• eBook ISBN: 9781000201161
• Edition: 1
• Topic: Education
• Subtopic: Education General

Defending Assessment Security in a Digital World explores the phenomenon of e-cheating and identifies ways to bolster assessment to ensure that it is secured against threats posed by technology.

Taking a multi-disciplinary approach, the book develops the concept of assessment security through research from cybersecurity, game studies, artificial intelligence and surveillance studies. Throughout, there is a rigorous examination of the ways people cheat in different contexts, and the effectiveness of different approaches at stopping cheating. This evidence informs the development of standards and metrics for assessment security, and ways that assessment design can help address e-cheating. Its new concept of assessment security both complements and challenges traditional notions of academic integrity.

By focusing on proactive, principles-based approaches, the book equips educators, technologists and policymakers to address both current e-cheating as well as future threats.

1
E-Cheating

We are in the midst of a global panic about student cheating. Concerns about students outsourcing their work to ‘contract cheating’ websites, which produce bespoke assignments for a fee, regularly appear in the media. Alarming statistics are presented, such as 16% of students admitting to having paid someone online to do their work and only 1% of those cheating being caught (Lee, 2019). Cheating devices are being sold to students online; machine learning tools are being used by students to do their homework (Dawson, in press); and YouTube stars make big money selling product placement advertising for cheating services (Jeffreys & Main, 2018).

One thing unites this new context for cheating: technology. Whether it’s managing the logistics of providing cheating services, like an essay mill’s e-commerce site, pushing cheating to students through social media, or sophisticated artificial intelligence producing papers for students, technology is pervasive in the current cheating landscape.

On the flip side, technology is also heralded as part of the solution to cheating, ranging from locking down students’ computers while they complete assignments (Dawson, 2016) to flying drones above exam halls to spot cheaters (Reuters, 2017). Technology poses both great promise and great peril for the problem of cheating. And this is without considering the potential benefits of future technologies to both cheaters and those who wish to stop them.

This is a book about technologised cheating, or ‘e-cheating’, and what we can do about it. The book is primarily targeted at researchers, educators, technologists and leaders in the field of higher and professional education. But it’s also a topic of basic human interest; who doesn’t love a good story about cheating? The book explores the current and likely future states of e-cheating, what can be done to secure assessment in this new world, and some of the challenges we are likely to face along the way.

Why a book specifically focused on e-cheating? Firstly, because with new technologies come new ways to cheat. The rise of the World Wide Web in the late 1990s was associated with a rise in copy-paste plagiarism. Similar panics about new technologies can be found as far back as Socrates complaining about the emergence of writing as producing students who appear to know things they actually do not know (see Plato’s dialogue ‘Phaedrus’, sections 257c–279c). New technologies challenge norms about what is acceptable in education. The first step in addressing new cheating technologies is a systematic interrogation of technology and cheating.

In addition to interrogating technologised cheating, it is also essential to understand and question technologies that are being used to address cheating. In my work, I have tried to empirically test the claims of vendors of anti-cheating technologies. I’ve published about my successful hacking and cheating attempts at computer-based exams (Dawson, 2016). I’ve tested out new machine learning tools to see if they really can build an authorship profile on a student to see if the student has contract cheated (Dawson, Sutherland-Smith, & Ricksen, 2020). I’d like to give you a third example here, about replicating the types of cheating approaches I’ve seen used successfully in online exams, but legally I’m not allowed to do so. This is infuriating. Just as we need to scrutinise cheaters and cheating businesses, we need permission to question and study those on the other side, who profit from attempts to stop cheating. Do their anti-cheating technologies actually work?

The study of e-cheating benefits from its adjacent technological disciplines, such as artificial intelligence and cybersecurity, which offer us new metaphors and methodologies to study cheating. These include the notion of ‘penetration testing’ a system to find its vulnerabilities; the holding of hacking competitions with bounties for those who can demonstrate successful hacks; and the study of ‘social engineering’ to find non-technical breaches (Schneier, 2018). As Schneier notes, with the integration of computers into all facets of society, ‘computer security will become everything security’ (p. 7), which means that the lessons of cybersecurity apply to cheating as much as anywhere else.

Why care so much about cheating? I could ask the same question of you reading this book – what motivated you to read about cheating? There are many answers to this question, often quite personal ones. Cheating might just feel fundamentally unfair or unethical. You might believe cheating damages learning and the reputations of educational systems. Philosophers have argued extensively about the moral wrongness of cheating (Bouville, 2009). But the biggest concern I have about cheating is a fairly pragmatic one: it leads to people being accredited as having met particular learning outcomes they have not actually met. When I go to the doctor or drop my kids off at school, I place great trust in the capabilities of professionals whom I hope have earned their qualifications honestly. Cheating creates an unsafe situation where incompetent and dishonest professionals walk amongst us.

This is a book about the minority of students who e-cheat; the vast majority of students do not e-cheat or even cheat. However, for reasons of fairness, public safety, trust in education systems and productivity, we need to do our best to stop the small minority that do. So what proportion of students e-cheat? Little data is available that specifically targets e-cheating. Some e-cheating behaviours are considered in large-scale self-report surveys such as Bretag et al.’s (2018) survey of Australian university students, which found that roughly 6% of students admitted to having engaged in ‘contract cheating’, such as the use of websites that provide custom-made assignments (Bretag et al., 2018). A synthesis of the literature by Newton (2018) found a much higher proportion on average, as high as 16% since 2014; however, there was a very significant range in the findings of individual research studies, with ten studies finding prevalence of 20% or higher, and almost double that number of studies finding prevalence of 1% or less. There is much less data about newer e-cheating behaviours, such as the use of unauthorised electronic tools, or the employment of computer-based exam impersonators. But what we know from contract cheating tells us that prevalence probably varies significantly between contexts and over time.

You might have noticed this chapter uses the word ‘cheating’ a lot, and it has not yet used the term ‘academic integrity’. This is intentional and unashamed. The field of academic integrity has tended to focus on the positive: education about the importance of integrity; about citing sources; on awareness campaigns and honour codes; and on cultural change (Fishman, 2014; McCabe, Treviño, & Butterfield, 1999, 2002). This is important work and I do not in any way wish to diminish it. However, the field’s discomfort with the systematic study of the ways students cheat, including the seedy underbellies of cheating sites and hacker forums, has led to a somewhat dichotomous situation: you’re either pro-integrity or anti-cheating. This is a false dichotomy. We can, and we must, both promote academic integrity and seek to detect cheating. In my Australian context, doing both is legally required of universities (Australian Government, 2015). This book’s focus on e-cheating – in addition to the promotion of academic integrity – allows us to explore the negative, with a focus on technology.

Before continuing, it’s important to acknowledge that cheating is contextual and socially constructed. In that spirit, here’s a bit about me, my biases and my worldview. Firstly, I come from the research area of assessment and feedback. I’m currently Associate Director of the Centre for Research in Assessment and Digital Learning (CRADLE) at Deakin University. I view cheating as an educational assessment problem, and situate my work on cheating and academic integrity within my broader work on assessment in higher education. I take a pragmatic approach (Badley, 2003) to research and am comfortable working with qualitative and quantitative methods; I think that with a bit of lateral thinking, evidence from a range of paradigms can be compatible. I’m located in Australia, which seems to have engaged rather extensively with academic integrity. When it comes to cheating, I am something of a sinner: my mum helped me cheat on a poster when I was in grade four; my dad gave me too much help on a story I was writing in grade seven; and I have Googled the answers to some online compliance quizzes. But apart from that I think I’m clean. Some things that others consider cheating (e.g. reading a study guide rather than the primary text, which McKibban & Burdsal, 2013 mention as a cheating behaviour) I don’t consider cheating, and some things that I think are cheating (e.g. using an online auto-paraphrase tool rather than paraphrasing manually) others might think are ok. I’m an online privacy obsessive: I use Virtual Private Networks (VPNs) to protect my Internet traffic; I have a sticker over my webcams to physically stop spying; and I’ve been a member of the Electronic Frontier Foundation. Finally, Turnitin has provided funding to my research centre for the direct costs of some of my research (but not for this book). I’m aware of some of the tensions in the above positioning. All of this is to say that you should read what I have to say in this book with a critical eye.

This chapter introduces and defines e-cheating, which is important as it’s a concept that is used across the whole book. The chapter goes on to explore ways in which e-cheating is currently happening, and what the best evidence is about its prevalence. The chapter includes a taxonomy for e-cheating, which is used later when discussing what can be done to prevent or detect different types of e-cheating. It also includes a roadmap for reading the rest of the book.

Defining e-cheating

This is not the first text to use the term e-cheating. Other authors have studied technologised cheating before, and prepending an ‘e-’ to cheating is a somewhat obvious move – think e-assessment, e-bikes and e-waste. The term e-cheating has been used rather loosely, often with an implicit, straightforward definition that e-cheating means technologised cheating (e.g. Gao, 2012; K. O. Jones, Reid, & Bartlett, 2008a, 2008b; Khan, 2010; Khan & Balasubramanian, 2012). Some other researchers have been more specific, defining e-cheating as:

• ‘student violations of academic integrity through the use of any technology oriented device’ (D. L. King & Case, 2014, p. 21);
• ‘using information technology (IT) to aid in the process of cheating in a class. This includes the use of personal digital assistants (PDAs), camera or picture cell phones, two-way pagers, programmable calculators, computers, the Internet, and so on to gain an unfair advantage’ (D. L. King & Case, 2007, p. 71);
• ‘Cheating through the use of electronic resources’ (Styron & Styron, 2010, p. 38);
• ‘the use of the World Wide Web to assist with cheating’ (Rogers, 2006, p. 207); and
• ‘the habit of students engaging ICT devices to indulge in examination misconduct’¹ (Omonijo, 2012, p. 388).

Discussion of e-cheating has largely been restricted to relatively niche journals. At the time of writing, there is little use of the term e-cheating amongst the top educational technology journals: no mention in the British Journal of Educational Technology, one...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• List of Tables
• Acknowledgements
• 1 E-Cheating
• 2 Assessment security
• 3 The E-Cheating lifecycle and how to disrupt it
• 4 Cybersecurity, E-Cheating and assessment security
• 5 Lessons from E-Cheating in games and gambling
• 6 E-Cheating, assessment security and artificial intelligence
• 7 Surveillance and the weaponisation of academic integrity
• 8 Metrics and standards for assessment security
• 9 Structuring and designing assessment for security
• 10 Conclusion: Securing assessment against E-Cheating
• References
• Index