Students as Partners: Rethinking Student Voice in Course Evaluation

Bottom line up front: The European Standards and Guidelines (ESG 2015) place student involvement at the heart of quality assurance, and a substantial body of higher-education research — anchored by Healey, Flint and Harrington's influential 2014 framework — argues that students learn and quality improves most when students are treated as partners rather than survey respondents. Yet most course evaluation still operates on an extractive model: push out a form, harvest the numbers, file the report. This piece argues the gap is not philosophical but practical — partnership has been hard to scale — and that conversational, dialogic instruments are what finally make genuine student voice feasible across a whole institution.

What the frameworks already require

Student involvement in quality assurance is not a nice-to-have in the European context; it is structural. The Standards and Guidelines for Quality Assurance in the European Higher Education Area (ESG 2015), adopted across the Bologna Process, embed student participation throughout — in internal quality assurance, in review panels, and in the principle of student-centred learning. The European Students' Union has long pushed for students to be co-creators of quality rather than its subjects. The expectation, in short, is dialogue.

The students-as-partners model

The most cited articulation comes from Healey, Flint and Harrington's 2014 report for the Higher Education Academy, Engagement through Partnership: Students as Partners in Learning and Teaching in Higher Education — a work with well over a thousand scholarly citations. Partnership, they argue, is "a relationship in which all involved … are actively engaged in and stand to gain from the process of learning and working together." Their framework spans four domains, including curriculum consultation and the scholarship of teaching and learning.

The evidence base associated with partnership approaches reports gains in student engagement, metacognition, motivation and belonging — and, importantly, better-quality feedback, because students who feel ownership give more candid and constructive input than students filling in a form to be rid of it.

Why most evaluation still fails the partnership test

Despite the framework requirements and the research, the dominant practice remains extractive. The reasons are practical, not ideological:

Static forms are monologues. A fixed Likert questionnaire cannot ask a follow-up, cannot probe, and cannot show the student that anyone is listening. It is the opposite of dialogue.
The loop is rarely closed. Students complete evaluations, hear nothing back, and reasonably conclude the exercise is performative — which depresses both response rates and effort. Closing the loop is the single strongest driver of sustained participation.
Partnership has not scaled. Genuine dialogue — student-staff committees, focus groups, co-design workshops — is labour-intensive and reaches only a small, often unrepresentative, set of engaged students. Universities have faced a real trade-off between depth (a few partners) and breadth (everyone surveyed shallowly).

That last point is the crux. Partnership stalled not because institutions disagreed with it but because there was no way to have a genuine, two-way conversation with thousands of students at once.

"But isn't partnership just engaged students talking to themselves?"

This is the most important counterargument, and it cuts deep.

Critics rightly note that partnership initiatives often recruit the already-engaged — confident, high-achieving, frequently more privileged students — whose voices are then mistaken for "the student voice." Meanwhile the quiet, the struggling and the marginalised, whose feedback quality assurance most needs, remain unheard. On this reading, partnership can worsen representativeness compared with a universal survey, trading the breadth of a flawed instrument for the depth of an unrepresentative one.

The criticism is fair, and any honest advocate has to hold both truths: surveys are representative but shallow; traditional partnership is deep but skewed. The goal is not to pick one failure mode. It is to dissolve the trade-off — to make dialogue universal, so that depth no longer requires sacrificing breadth, and a conversational instrument reaches the disengaged student precisely because it lowers the effort and signals that someone is actually listening. Partnership becomes representative only when it scales to everyone.

What partnership looks like in practice

Partnership is not a single intervention but a spectrum, and most institutions already do the easy end of it. Student representation on programme committees, student-staff liaison meetings and the occasional focus group are all real, if limited, forms of partnership. The deeper end — students co-designing assessment, partnering in the scholarship of teaching, or helping shape the evaluation instrument itself — remains rare. The point is not that every course needs a co-design workshop, but that the stance should shift across all of these activities: from gathering data about students to making decisions with them. Even a conventional evaluation can be made more partnership-like by telling students in advance what last year's feedback changed, by letting them weight which topics matter most this term, and by reporting results back to the cohort rather than only upward to management. Each of these costs little and signals, unmistakably, that the feedback is part of a relationship rather than a compliance ritual.

Letting students shape the questions

One concrete and under-used move is to involve students in designing the evaluation itself. When the questions are written entirely by administrators, they encode administrative priorities — and students answer dutifully without ever raising what actually mattered to them. Allowing students to surface their own themes, in their own words, turns the instrument from an interrogation into a conversation and routinely reveals issues no fixed questionnaire would have probed. This is the practical core of treating students as partners in quality assurance rather than as the raw material of it: the people who experienced the course help define what counts as a good answer about it. Done well, it also improves data quality, because students engage more honestly with questions they helped frame. A conversational instrument does this natively, because an open, probing exchange lets the student introduce the agenda the form-writer never anticipated.

How Koji approaches it

Koji for Education is, in effect, an attempt to make partnership scalable — to give every student a short, genuine, two-way conversation rather than a form:

AI-moderated conversational interviews turn evaluation from a monologue into a dialogue at scale: the instrument asks follow-ups, probes vague answers, and acknowledges what the student says — the experience of being heard, delivered to an entire cohort, not just a committee.
Six structured question types (open-ended, scale, single- and multiple-choice, ranking, yes/no) let students contribute both comparable data and their own framing, in their own words.
Standardised, bias-aware AI moderation gives every student the same quality of conversation, which helps reach quieter and less confident voices that traditional partnership forums tend to miss — improving representativeness rather than narrowing it.
Automatic thematic analysis lets staff genuinely act on what thousands of students say, not just the loudest few.
Closing-the-loop action tracking records and surfaces what changed in response — the "you said, we did" step that turns a survey into a relationship and sustains participation over time.
GDPR/AVG-compliant handling keeps this dialogue appropriate to the European regulatory context.

The same conversational interview engine powers general user and customer research on the main Koji platform; for education it is tuned to the partnership and quality-assurance expectations of the European Higher Education Area.

If your institution is committed on paper to student voice but still relies on one-way forms, see how Koji for Education makes partnership scalable.

The takeaway

European frameworks already demand that students be partners in quality, and the research says partnership produces both better learning and better feedback. What has been missing is a way to do it at scale without sacrificing the representativeness a universal survey provides. Treating students as respondents was always a compromise forced by the limits of the form. Those limits no longer hold — and the institutions that move first from extraction to dialogue will get richer, fairer evidence and more engaged students at the same time.