Written feedback judged as a credibility test, not neutral advice, study argues

Every written comment a university student receives on an essay carries more weight than it appears to. According to a new theoretical paper published in the journal Higher Education, written evaluative feedback in English-medium higher education should not be understood as an epistemically neutral exchange about academic performance. Instead, argues Tony Myers of Zayed University in Abu Dhabi, feedback is a credibility encounter: a moment in which the student is positioned as a more or less legitimate knower, someone whose contributions to knowledge count or do not count. The claim reframes one of the most routine practices in academia as a site where deep questions about justice, identity and the purpose of education are decided, often invisibly.

The paper addresses what Myers identifies as a blind spot in contemporary feedback research. Over the past two decades, scholars have built sophisticated accounts of feedback as dialogue, emphasising student agency, feedback literacy and the active uptake of comments. Students are increasingly encouraged to seek, interpret and use feedback rather than passively receive it. Yet, Myers contends, this literature has had much less to say about what feedback does to students in relation to knowledge and knowing. Feedback does not merely convey information about how well a task was performed; it communicates judgements about whether the student’s thinking is worth taking seriously, and those judgements can have consequences that outlast any single assignment.

To build this argument, the study brings three theoretical traditions into conversation. The first is recent scholarship on epistemic agency in assessment, which examines how assessment practices shape students’ capacity to act as knowers rather than merely as recipients of knowledge. The second is Gert Biesta’s influential account of the three purposes of education: qualification, socialisation and subjectification. Qualification concerns the knowledge and skills students acquire; socialisation concerns their induction into traditions, norms and practices; and subjectification concerns the emergence of the student as an independent thinking subject, someone capable of existing in the world in their own distinctive way. The third tradition is Miranda Fricker’s theory of epistemic injustice, which describes wrongs done to people specifically in their capacity as knowers.

Fricker’s framework distinguishes two main forms of epistemic injustice. Testimonial injustice occurs when a speaker’s testimony is given deflated credibility because of prejudice related to their identity, so that the speaker is wronged as a giver of knowledge. Hermeneutical injustice occurs when gaps in shared interpretive resources leave people unable to make sense of their own experiences. Myers focuses primarily on the testimonial form, arguing that written feedback is a paradigmatic site for it: the marker is, in effect, a credibility assessor whose judgements about a student’s claims are shaped by perceptions of the student as well as by the content of the work itself. When those credibility judgements are driven by structural prejudice rather than by genuine engagement with the student’s thinking, the student suffers a distinctive kind of harm.

Read together, these perspectives reveal what Myers describes as an unresolved tension at the heart of feedback theory. Feedback is supposed to serve socialisation: it inducts students into disciplinary norms, showing them how arguments are constructed, what counts as evidence, and how a field writes and thinks. But the very same judgements that accomplish this induction can simultaneously foreclose the student’s emergence as an independent epistemic subject. A student whose contributions are systematically discounted, whose voice is heard as deficient rather than different, learns not only the discipline’s conventions but also a corrosive lesson about their own entitlement to know. In Biesta’s terms, socialisation can colonise subjectification, crowding out the educational purpose that many educators regard as the most important one.

The paper grounds this abstraction by drawing on documented feedback policy to specify the institutional conditions of the encounter. Feedback is not delivered in a vacuum; it is shaped by institutional frameworks, quality assurance agendas, turnaround times and rubrics that predefine what legitimate academic performance looks like. Within these structures, the marker acts as an institutionally authorised credibility judge. The authority of the judgement derives not only from the marker’s expertise but from the institution’s backing, which means that a deflating credibility assessment carries official weight. When structural prejudice, whether related to language, race, gender or cultural background, shapes that assessment, the injustice is not an individual failing of a careless grader but a systematic feature of the encounter itself.

The argument has sharp relevance for research on bias in academic judgement. Empirical studies have documented, for example, that student evaluations of teaching exhibit gender biases, sometimes detectable even in controlled online settings where the same instructor is presented under different identities. Scholarship on race and assessment in higher education has similarly conceptualised barriers facing minoritised students and argued for measurable change. Myers’s framework offers a theoretical account of why such biases matter beyond their effect on marks: they operate as testimonial injustices within credibility encounters, distorting the student’s standing as a knower and thereby undermining subjectification in Biesta’s sense. The injustice is not simply that the grade may be wrong, but that the student is wrongly positioned in relation to knowledge itself.

One of the most timely dimensions of the paper concerns automation. Myers specifies how the mechanism he describes operates when credibility judgements are delegated to automated instruments, a situation that has become increasingly common with the spread of artificial intelligence tools in assessment. When a judgment about credibility is produced by a machine, the resulting credibility deficit acquires the appearance of measurement, of objective fact rather than contestable interpretation. This matters because automated systems can inherit and amplify structural prejudice. Research has shown, for instance, that GPT-based detectors are systematically biased against non-native English writers, flagging their authentic writing as machine-generated more often than that of native speakers. Scholars of AI ethics have explicitly analysed epistemic injustice in generative AI, and assessment researchers have begun developing recognition-based frameworks for understanding the place and value of AI-generated feedback.

In the automated case, the epistemic wrong is layered. The student is subject to a deflating credibility judgement, but the source of that judgement is an instrument whose authority is anchored in the aura of quantitative objectivity. Contesting the judgement means contending not with an individual marker but with an opaque system, which raises the barriers to challenge precisely as the stakes rise. Myers’s point is that this is not merely a technical problem to be solved with better algorithms. Epistemic injustice in feedback, he argues, is not simply a failure of technique but a distortion of educational purpose. A system can be efficient, consistent and well designed, and still function unjustly if its credibility judgements are structurally prejudiced or if they foreclose rather than foster the student’s development as an independent thinker.

The paper’s conclusion carries implications for institutions, educators and researchers alike. If feedback is a credibility encounter, then improving feedback means attending not only to the clarity, timeliness and actionability of comments, but to the conditions under which students are recognised as legitimate knowers. That requires scrutinising the structural prejudices that shape judgements, interrogating the delegation of those judgements to automated instruments, and asking hard questions about whether current feedback cultures serve socialisation at the expense of subjectification. Recent work on feedback cultures in universities, including analyses of strategy documents from more than a hundred UK institutions, suggests that accountability and quality assurance agendas already shape how feedback is enacted, often in ways that sit uneasily with dialogic ideals. Myers’s contribution is to name precisely what is at stake when those agendas collide with students’ standing as knowers: nothing less than whether higher education fulfils its promise of bringing new thinking subjects into the world, or quietly teaches many of them that their voices do not count.

Subject of Research: Epistemic injustice in written feedback and student credibility in higher education

Article Title: Feedback as a credibility encounter: epistemic injustice and the foreclosure of subjectification

Article References: Feedback as a credibility encounter: epistemic injustice and the foreclosure of subjectification. (n.d.). https://doi.org/10.1007/s10734-026-01797-w

Image Credits: AI Generated

DOI: 10.1007/s10734-026-01797-w

Keywords: epistemic injustice, feedback, higher education, assessment, testimonial injustice, subjectification, epistemic agency, credibility, Gert Biesta, Miranda Fricker, artificial intelligence, feedback literacy