High-stakes teaching evaluations can damage academic wellbeing
Updated Jul 08, 2026
A feedback system can fail even when students answer honestly. If evaluation scores and open comments are treated as routine evidence, but the process leaves academics more anxious, stressed, or defensive, universities have a governance problem as well as a survey problem. At Student Voice AI, we see the same tension in student voice: students need space to speak candidly, but institutions also need processes that keep feedback usable, fair, and proportionate. That is why Shaylen E. Stone, Gerard Jefferies, Megan Lee and Cindy Davis's Assessment & Evaluation in Higher Education paper, "Impacts of student evaluations of teaching on Australian university academics", matters for UK universities using student evaluations to inform teaching review, enhancement, and staff decisions.
Context and research question
Student evaluations of teaching, often shortened to SETs, are usually justified as evidence about teaching quality or the student experience. Once scores and comments are used in promotion, performance management, or contract renewal, they stop being just feedback. They become employment evidence, often without the same methodological caution universities would apply elsewhere. That problem becomes sharper when institutions already know that student evaluation scores are not perfectly reliable, and when a few sharp comments can carry more emotional weight than a page of middling scores.
Stone and colleagues ask a direct question that matters well beyond Australia: how do quantitative scores and qualitative SET comments affect academics' wellbeing, and how strongly are those perceived impacts associated with depression, anxiety, and stress? The study used an anonymous online survey with 574 participants from 41 Australian universities. Participants completed a Health and Wellbeing Scale on the perceived impact of quantitative and qualitative SET responses, alongside the DASS-21 mental health measure. For UK higher education teams, that makes the paper useful because it moves beyond anecdote and tests the issue at sector scale.
Key findings
Both the scored and written parts of SETs were associated with poorer wellbeing. The abstract states that the study found significant implications for academics' health and wellbeing from both quantitative and qualitative SET responses. That matters because institutions sometimes treat the numerical instrument as the serious evidence and the open comments as informal colour. This paper suggests both can carry real psychological weight.
"both quantitative and qualitative SET have significant implications on academics' health and wellbeing"
The paper links SET impact to depression, anxiety, and stress rather than general dissatisfaction alone. The regression models remained significant after controlling for age, gender, years in academia, and employment position, with R-squared values between 0.19 and 0.21 and p < 0.001. The abstract does not report every coefficient separately, but it does make clear that SET-related wellbeing impacts were meaningfully associated with mental health outcomes. For UK institutions, that raises the stakes. This is not only about whether staff like evaluation systems. It is about whether those systems create avoidable harm.
Scale matters because the study is not describing one department's bad culture. The sample covers 574 participants across 41 universities, which gives the findings more weight than a local case study of a single institution. The transfer question still matters, because Australia and the UK are not identical, but the core mechanism is familiar: anonymous student feedback becomes harder to absorb when it is tied to performance narratives, career progression, or repeated personal criticism.
The policy warning is about high-stakes use, not about silencing students. The authors conclude that institutions should reconsider how far SETs are used in promotion, performance evaluations, and contract renewal. That is a practical distinction UK universities need to hold onto. Student voice is still valuable, but a feedback tool becomes much riskier when raw comments and marginal score differences are treated as employment evidence without stronger safeguards.
Practical implications
For UK higher education teams, the first implication is to separate developmental feedback from high-stakes judgement more clearly. If module evaluations are meant to improve teaching, they should mainly support reflection, dialogue, and course enhancement. If they are also used in appraisal or promotion, institutions should define exactly what weight they carry, what corroborating evidence is required, and how context is documented. That reduces the risk that survey data becomes a blunt proxy for teaching quality.
Second, universities should govern qualitative comments as carefully as they govern scores. Clear guidance on constructive feedback, triage for abusive or personal remarks, and rules on who can see raw text are not optional extras. They are part of responsible evidence use. A practical student comment analysis governance checklist is relevant here because it sets out the operational controls needed when open-text feedback may affect both staff wellbeing and institutional decisions. The benefit is straightforward: teams hear student concerns without normalising harm.
Third, institutions should report patterns rather than isolated remarks. A single hostile comment can dominate how a lecturer remembers a survey, even when it is unrepresentative. Aggregating repeated themes across modules, cohorts, and time gives staff a fairer starting point for interpretation. It also makes student feedback easier to act on. That is why a defensible workflow for analysing open-text comments systematically matters, especially when universities want evidence that is reproducible and usable beyond one reading of a comment sheet.
Finally, universities should treat staff wellbeing as part of student voice design, not a downstream HR issue. This paper suggests that better evaluation practice is not only about getting more student responses or better survey wording. It is also about deciding how feedback is released, discussed, contextualised, and supported. When institutions build those protections in, they make evaluation data more trustworthy for staff and more useful for quality enhancement.
FAQ
Q: How should a university review its SET process after reading this paper?
A: Start by mapping the full route from collection to use. Ask who sees raw comments, whether abusive remarks are screened, how much weight scores actually carry in appraisal, and what other evidence is considered alongside them. Then check whether staff are given a structured discussion and support process after results are released. The goal is to make feedback developmental by default and high-stakes only with explicit safeguards.
Q: What are the methodological limits of this study?
A: This is a cross-sectional, self-report survey from Australia, so it cannot prove that SETs directly caused depression, anxiety, or stress. It measures perceived impacts and their association with mental health outcomes. That said, the sample is large, spans 41 universities, and addresses a question many institutions prefer to treat as anecdotal. UK teams should read it as a strong warning signal, especially when representativeness problems such as non-response bias in student evaluations already make high-stakes use harder to defend.
Q: What does this change about student voice practice more broadly?
A: It reinforces that student voice systems should be designed to produce candid, constructive, and interpretable evidence. If the process harms staff or encourages defensive reactions, the institution has made student feedback less useful for everyone. The practical aim is not to soften critique, but to make sure comments are analysed, contextualised, and acted on in ways that improve teaching rather than distort it.
References
[Paper Source]: Shaylen E. Stone, Gerard Jefferies, Megan Lee, Cindy Davis "Impacts of student evaluations of teaching on Australian university academics" DOI: 10.1080/02602938.2026.2696400
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