Preventing collusion in online exams

Cheating has always been a concern for educators; even more so in a world of remote learning and online testing where collusion among students can be easily carried out.

Remote proctoring, or e-proctoring, solutions are available but these can present issues. Some are prohibitively expensive on a cost-per-head scale, and concerns have been raised around privacy standards and data security. For many students, the process can be stressful and more anxiety-producing than proctoring in a standard examination environment. 

In new research published in npj Science of Learning, engineers at New York's Rensselaer Polytechnic Institute suggest that the problem of student collusion can easily be overcome if individual competencies are known and taken into account. 

Distanced online testing

The study's authors utilise a strategy called 'distanced online testing' (DOT) to reduce the ability of students to help one another via phone or online methods. Key to the method is an understanding of student proficiency. 

Under DTO, participating students receive the same questions but at differing times depending on their known skill level. Those with a greater knowledge or competency are issued questions that relate to that knowledge or skill after students recognised as having less proficiency. The researchers say this reduces the incentive to receive help from others and additionally motivates increased study prior to exams as students are made aware that collusion will not be possible. 

The order of questions is based on an estimation of student competency derived from past exams, averages and other available student grade data. 

The researchers say the use of DOT technology for optimised sequenced questioning in synchronised timeslots "reduces collusion gain by two to three orders of magnitude relative to the conventional exam in which students receive their common questions simultaneously". They further assert that post-exam surveys and statistical tests deliver positive results. 

While obviously not applicable for every exam scenario, researchers say the DOT method offers a cost-effective solution to student collusion, which the study suggests represents around 80% of cheating events. 

The full research report can be found here. 

Image credit: ©stock.adobe.com/au/smolaw11