Detection is possible!
Despite common misperceptions and uncertainty around detection, it is possible to detect work which has not been completed by the student. This can be done by markers and other staff trained in the area, additionally, software may be able to help detect cases.
Key things to look for may include metadata in the assignment which is unusual (for example the author name is different to the student name); the template is unusual (for example, the proof language is Indian English, rather than Australian English, or it does not follow the provided template given by the University); the subject matter discussed does not respond to the question posed and seems to be responding to a different question. Items within the work may also indicate outsourcing, such as strange or hidden textboxes; reflections given by students don’t relate to the workshop or practical session held in the class.
Staff and software can help detect cases of contract cheating
In addition to these general indicators, where possible, additional indicators which may alert markers to a suspicion of contract cheating is different in language style or expertise when compared to the student’s prior work. It is acknowledged that in large cohorts this is not something which is always possible to be detected, particularly where there may be a large pool of tutors or makers managing a subject.
A study by Dawson and Sutherland-Smith (2017) found that markers detected contract cheating 62 per cent of time accurately, when knew that they should be looking for it. Any evidence collected should be referred to the student to allow them to respond, so any indicators found should be collated and summarised. For in-depth detail on detection indicators, refer to the substantiating contract cheating guide.
- Clare, Walker, and Hobson 2017
Can we detect contract cheating using existing assessment data? Applying crime prevention theory to an academic integrity issue
- Dawson, Sutherland-Smith, and Ricksen 2020
Can software improve marker accuracy at detecting contract cheating? A pilot study of the Turnitin authorship investigate alpha
- Dawson and Sutherland-Smith 2017
Can markers detect contract cheating? Results from a pilot study
- Rogerson 2017
Detecting contract cheating in essay and report submissions: process, patterns, clues, and conversations
- TEQSA 2020
Guide to substantiating cases of contract cheating, with list of indicators to help with detection
Local area experts
Local area experts can not only help in the determination of cases or evidence gathering in relation to the detection of potential outsourcing, but may also be able to train other staff in the key signs to look out for. It is recommended that all providers have staff who have undertaken some training or development in this area.
- Dawson and Sutherland-Smith 2019
Can training improve marker accuracy at detecting contract cheating? A multi-disciplinary pre-post study
- Detecting and addressing contract cheating in online assessment
- Substantiating contract cheating guide
Evidence and processes
Gathering and determining suitable evidence
There are different methods to detect contract cheating. Various indicators present in student assignments can raise suspicion that the work has been outsourced. Essential to the process is gathering evidence in a clear way that allows decision-makers to come to a determination of whether a case is substantiated or not; and for students (and any support person or advocate) to be able to appropriately respond to the allegation.
Gathering evidence in a clear way is essential
Evidence might include a variety of objects, such as screenshots, edited files, summary reports, images, or statements from persons with evidence of the outsourcing, or emails. Often the evidence will include a report from a marker, or local area expert who determined the original suspicion based on some of the indicator factors, or from using any available technology.
As with text-matching software, for providers who do use technology to assist in their detection processes, human interaction to quality control outputs of system reports is important. An expert eye can provide a clear intervention in data which may or may not be a substantial enough level of evidence for local processes. This is equally applicable whether determining evidence for submitted assignments or considering suspicions of outsourcing in online exams.
- Chuang, Craig, and Femiani 2017
Detecting probable cheating during online assessments
- Juola 2012
Using technology to determine authorship attribution: Large scale experiments in authorship attribution
Standards of evidence and burden of proof
Evidence that is presented to students needs to be clear. The standard of proof for providers is ‘on the balance of probabilities’. This means that the burden of proof for a decision-maker (which may be an individual, panel of people, or committee) is that it is more probable than not that the evidence presented before them indicates that the student submitted work which was completed by someone else.
The evidence considered must include the evidence collated by the institution, and any evidence submitted or presented in response to the allegation and original evidence by the student, or the student’s representative.
Evidence should be detailed
Any outcome or determination letter which informs the student as to whether the allegation was substantiated or not, must refer to the evidence that was used by the decision-maker/s to come to that outcome. For example, during any committee hearing, it may have been found that a piece of evidence was refuted by the student, and the counter evidence accepted by the committee. The evidence which related to the determination of the case (whether substantiated or not substantiated) should be detailed.
Provided no details of the decision-making process are shared (for example do not share minutes or case reports), it is recommended that outcomes are given to those reporting suspicions and allegations. This provides confidence to persons throughout the institution that cases are carefully considered and matters with substantial evidence are managed appropriately.
- Ellis, Haeringen, and House 2020
Technology and standards of evidence in contract cheating case management
Managing caseloads, procedural fairness, and natural justice
Providers should ensure that they have sufficient systems, structures, and processes to manage cases which are brought to them in a timely manner, and one in which procedural fairness and consistency is applied at all times.
Sufficient systems, structures and processes are required to manage contract cheating cases
Training for staff involved in collating evidence, or considering evidence, is important to ensure parity of treatment in regard to evidentiary standards, as well as in considering counter-evidence and responses provided by the student whose work is being questioned. Decision-makers must be able to interpret case reports or data presented. Where applicable, this training should also be provided to student advocacy services.
Providers may have interim steps which allow for informal or formal conversations or interviews, which may alleviate the need for a formal hearing. Irrespective of the processes, all students should be afforded the same opportunities to defend their work (natural justice). Templates and standard practices can assist providers in maintaining this consistency, where processes for investigation and decision-making are decentralised. Clear accountability for each step of the process should be defined.
Site type awareness and blocking
TEQSA will be developing further advice and guidance on this matter over coming months.
Sites and companies can be ruthless!
Not only are the marketing campaigns from these sites predatory and ruthless in targeting students, they are also very persuasive when it comes to promoting themselves as legitimate methods for students to get help in their studies. Messaging often focusses on the key drivers previously outlined that may lead students to approach an outsourcing provider in the first place.
Contract cheating companies often promote themselves as legitimate methods for students to get help
Companies may also criticise educational institutions as not providing students with enough help, support, or time from tutors, and that they (the company) may be a great option for the student to get help. These targeted campaigns are usually found in the more commercial companies, contract cheating companies and bidding sites, but similar tactics may be found in the peer-sharing sites.
Business models may often have a parent company with several child companies, all of which share information and may use work a student uploads onto a peer-sharing site to sell on to another person through one of their essay mill sites. This has even been seen in the detail found in the small print of a site’s T&Cs.
Companies can also trawl public pages and social media feeds for keywords to offer their services to people posting that they are working on an assignment, and stressed, for example. As seen in research undertaken by Amigud in the US, companies are not put off in advertising their services even when this may be illegal.
- Amigud 2019
Analysis of Twitter targeting and promotion of contract cheating
- Ellis, Zucker, and Randall 2018
The infernal business of contract cheating: understanding the business processes and models of academic custom writing sites
- Rowland et al 2018
Marketing campaigns of sites: ‘Just Turn to Us’: the persuasive features of contract cheating websites
Direct targeting and security considerations
What can be alarming to students are the direct targeting to personal phone numbers or email accounts. This can happen for a variety of reasons, unrelated to an account being hacked. For example, Facebook groups set up for students may show the list of group members. Students who include their phone number on their Facebook profile may have their phone number accessed this way. Companies may set up general targeted mailing campaigns based on the known end of university email addresses or may have gained access to student class lists. Unfortunately, when engaging some contract cheating providers, students have allowed providers access to their unit sites to complete work for them, giving over their email address and password. This gives the third-party access to class lists or email listings.
Educating students on cyber security issues will help protect their contact details
In order to address direct targeting, it is advised that providers communicate with their students about these issues and that messaging includes appropriate cyber security measures. Educating students on security issues related to third party provider access to their own student accounts will help protect those individual students and reduce impact on others if student class lists are accessed.
Blocking of sites
The blocking of sites identified as being a contract cheating (or other form of assignment provider) can be one method to prevent access to these sites. The action has been promoted as a method to send a message to students that these types of sites are not condoned. However, there are some drawbacks with site blocking.
Firstly, students will only receive the message when on campus and logging on from institutional Wi-Fi, or when using a provider’s computer (in a computer lab for example). Messaging will not work when students are off-campus, and given the mass shift to online delivery to adhere to COVID-safe distancing measures across Australia, this will be applicable to the vast majority of the student body.
Site blocking lists will need to be regularly maintained
Secondly, it is recommended that any blocking of sites is coupled with an educative response to refer students to local support services and study guides or referring them to academic integrity definitions in regard to breaching expected standards.
Finally, the list of sites itself changes frequently. Sites may often have a parent and child site and these child sites are known to change their names or redirect to other sites. New sites are regularly appearing and will need to be added to sites to be blocked. Any list which is utilised at an institution will need to be regularly maintained.
- Seeland, Stoesz, and Vogt 2020
Effectiveness of blocking sites: Preventing online shopping for completed assessments: Protecting students by blocking access to contract cheating websites on institutional networks