Students report severe distress during academic misconduct investigations
Updated Sep 03, 2026
An academic misconduct allegation may look like a case to be processed. For the student waiting for a decision, it can bring severe anxiety, stress, and depression symptoms. At Student Voice AI, we treat comments on high-stakes procedures as evidence about student experience, not administrative noise. Guy J. Curtis and Amanda W. White's new Assessment & Evaluation in Higher Education paper, "The emotional impact of academic misconduct allegations on students", gives universities unusually direct evidence of that gap. For UK teams reviewing casework, assessment policy, and student anxiety about AI detectors, the message is clear: evidential thresholds, waiting times, communication, and support are part of the student experience, not separate from it.
Context and research question
Universities need formal processes for suspected academic misconduct. They also need to understand the harm those processes can cause, including when an allegation is not upheld. The evidence base has been thin: earlier qualitative research described fear, shame, and lasting anxiety, but had not measured the intensity of students' distress during an investigation. This matters in a sector where academic stress is already a major university-related distress signal and a formal allegation can arrive alongside assessment deadlines, financial pressure, or existing mental health difficulties.
Curtis and White asked how much depression, anxiety, and stress students recalled experiencing while facing a formal academic misconduct process. They surveyed 771 students at two Australian universities and invited the 23 who reported a previous formal allegation to complete a follow-up; 22 completed it. The researchers adapted the 21-item Depression, Anxiety and Stress Scale so respondents rated how they had felt at the time of the allegation. They compared those scores with established adult and student norms, then added three open questions about learning, timing, wellbeing, and the wider experience.
This design provides a focused view of an experience that is difficult to study without breaching privacy. It also requires caution. The main sample contained only 22 students, came from two institutions, and relied on retrospective reports rather than measurements taken during live cases. The results show a serious risk that UK institutions should investigate locally, not a prevalence estimate for the whole sector.
Key findings
The levels of recalled distress were high and concerning. Half of the 22 students reported severe or extremely severe depression symptoms while facing the process. More than half reported anxiety in the extremely severe range, and 40.9% reported severe or extremely severe stress. These classifications come from norm-referenced DASS thresholds. They do not establish a diagnosis or prove that every misconduct process will have the same effect, but they show why institutions cannot treat distress as a peripheral case-management issue.
For some students, distress was severe across several dimensions at once. Only three students scored in the normal range for depression, anxiety, and stress together. Four scored in the extremely severe range on all three measures. That combination matters because a student may be trying to understand an unfamiliar procedure, prepare a response, and continue studying while experiencing acute symptoms across mood, worry, and stress.
Delay and uncertainty appeared to make the experience harder. Fourteen students answered at least one open question. Seven of the 14 responses to the question about timing and wellbeing mentioned anxiety, stress, or low mood. One student said quick resolution reduced stress; another described being unable to function during a process that lasted about six weeks. Three responses described a lasting fear of later emails or allegations.
"I do remember feeling ... scared whenever I got an email a while after."
Support could reduce some of that pressure. One student identified help from a student association support person as protective. The wider pattern supports a practical principle: students need clear timelines, regular updates, an opportunity to present evidence, and visible routes to independent advocacy and psychological support. Those measures do not weaken academic standards. They make a high-stakes process more navigable and procedurally fair.
Students could find the process educational while still finding it harmful. Ten of the 13 students who answered the learning question said the process was educational or identified something they had learned, such as better citation practice. Other responses described a punitive tone or an inability to defend themselves properly. The important distinction is not education versus enforcement. It is whether universities can enforce standards through a process that is timely, intelligible, proportionate, and capable of helping students learn.
The findings also sharpen the debate about AI detection. The authors argue that institutions should be especially cautious about starting formal proceedings on the basis of an AI-detector score alone because known false positives expose students to a potentially severe emotional cost. In practice, a detector result should be treated as an uncertain signal requiring human review and corroborating evidence, not as a conclusion.
Practical implications
For UK universities, the first priority is to set and monitor case-duration standards. Students should receive a realistic timeline when an allegation is raised, an explanation when it changes, and scheduled updates even when there is no decision yet. Institutions should track both median duration and long-running cases, then collect short post-case feedback on where uncertainty became most difficult. This can reduce avoidable distress without rushing careful decisions.
Second, universities should build support into the first communication, not offer it only after a student reaches crisis point. The allegation notice should explain the process in plain language, name the student's right to respond, and signpost an independent adviser, students' union advocate, and mental health support. Staff also need escalation guidance when a student shows signs of acute distress. Students can then reach help at the moment when uncertainty and shame may otherwise stop them asking for it.
Third, institutions should raise the evidential threshold before uncertainty becomes a formal allegation. AI-detector outputs, text-matching scores, and staff suspicion may justify a closer look, but no single weak signal should carry the decision. Human review, assessment context, drafting evidence, and a genuine opportunity for explanation should come first. This lowers the risk of avoidable investigations and strengthens confidence in the cases that do proceed.
Fourth, universities should treat process-generated text as student voice evidence. Closed-case questionnaires, appeals, advocacy notes, complaints, and reflections can reveal repeated issues with delay, unclear rules, tone, support, or perceived fairness. Research on 3,070 student misconduct reflections shows how aggregation can turn individual accounts into evidence for prevention. Student Voice Analytics fits here by helping institutions separate recurring themes across sensitive comments, while keeping the purpose on system improvement rather than individual surveillance. Teams then have a clearer route from case experience to policy change.
Finally, this evidence needs careful governance. Small groups, international-student status, disability, or unusual case details can make apparently anonymous comments identifiable. Teams should set minimum reporting thresholds, restrict access, document which text sources are included, and follow a student comment analysis governance checklist before comparing groups. These controls help improve the process without exposing the students who described it.
FAQ
Q: How should a university apply these findings to its academic misconduct process?
A: Start with a case-journey review. Measure how long each stage takes, inspect the clarity of the first allegation letter, and check whether every student receives a named contact, regular updates, independent advocacy, and mental health signposting. Then ask students after closure which parts felt unclear, delayed, unsupported, or fair. This turns the paper's warning into a manageable quality-improvement cycle.
Q: Do these results show that misconduct investigations cause severe distress for most students?
A: No. The 22-person sample was small, retrospective, and drawn from two Australian universities. Students rated how they remembered feeling during a previous process, so the study cannot establish sector-wide prevalence or a precise causal effect. Its value is as a strong risk signal, supported by the severity of the scores and the open responses, that institutions should test against their own case data and student accounts.
Q: What does this paper change about student voice in academic integrity work?
A: It shows that student voice should cover the operation of high-stakes procedures, not only teaching and general satisfaction. Universities can uphold standards while asking whether students understood the allegation, could present their case, received support, and experienced timely decisions. That broader view of student voice helps institutions judge the quality of a process as well as its outcome.
References
[Paper Source]: Guy J. Curtis and Amanda W. White "The emotional impact of academic misconduct allegations on students" DOI: 10.1080/02602938.2026.2725802
Request a walkthrough
Book a free Student Voice Analytics demo
See all-comment coverage, sector benchmarks, and reporting designed for OfS quality and NSS requirements.
- All-comment coverage with HE-tuned taxonomy and sentiment.
- Versioned outputs with TEF-ready reporting.
- Benchmarks and BI-ready exports for boards and Senate.
UK-hosted · No public LLM APIs · Same-day turnaround