Factors That Can Undermine the Credibility of Assessment

This material provides a theoretical introduction to the challenges of ensuring the credibility of assessment in digital environments, along with recommendations on how to mitigate them. It serves as a basis for analysing case studies in accordance with the task and the provided template.

The terms digital assessment and online assessment are often used as synonyms, but they in fact refer to somewhat different concepts. Assessment in a digital environment (digital assessment) is a broader term that refers to any form of assessment involving the use of digital technologies, regardless of whether it takes place in a physical classroom or remotely. Online assessment is a narrower term, referring specifically to assessment conducted via the internet, in virtual spaces, often as part of online teaching or hybrid teaching models.

Online environments and digital assessment offer numerous advantages, well-documented in the literature. At the same time, assessment in a digital environment, and especially online assessment, is associated with challenges such as the potential compromise of assessment credibility and reduced opportunities for communication between teachers and students.

Assessment credibility may be compromised because some learning outcomes are difficult to assess reliably in digital settings due to the expanded range of opportunities for students to cheat in online environments, but also because not all students have adequate technical conditions for online assessment.

With the rapid development of artificial intelligence and the almost universal availability of generative AI, opportunities for cheating in digital environments are increasing day by day.

Cheating in online environments

In online learning environments, students may be more inclined to cheat due to feelings of anonymity, weak connections with teachers and peers, and the lack of direct supervision. According to Albert Bandura’s theory of moral disengagement, individuals employ cognitive mechanisms—such as moral justification—to rationalise unethical behaviour and reduce feelings of responsibility for cheating (Bandura, 1999).

Sometimes, the boundary between cheating and acceptable forms of assistance is unclear, and teachers or institutions must clearly define what is permitted and what is not.

However, some behaviours are almost always perceived as cheating.
Examples of such behaviours (Sevnarayan-Maphoto, M, 2024) in online learning include:

• copying material from the internet without citing the source
  
• using generative artificial intelligence to produce content without citing the source or prompt
  
• using unauthorised online resources or tools during assessment activities
  
• seeking answers from others via social media during tests
  
• copying another student’s work
  
• unauthorised collaboration during exams
  
• false identity representation during online exams
  
• purchasing ready-made essays or projects.

It should be noted that many of these cheating practices also appear in traditional in-person teaching.

Students generally attempt to cheat in summative assessments because such assessments carry higher stakes related to the risk of failure.

Noorbehbahani et al. (2022), in a systematic review of the literature, state that the primary reasons for cheating in digital exams are the opportunities offered by the internet and online platforms. Other factors include deadline pressure, lack of motivation, and the perception that cheating does not harm others.

Similarly, Rüth et al. (2024) emphasise that a human-centred approach (focusing on the needs and perceptions of students) can reduce academic dishonesty in online exam contexts not through surveillance technologies, but through thoughtful exam design and clear communication regarding the purpose and benefits of assessment.