Students as Partners, Not Just Respondents: Co-Designing Course Evaluation

In brief: The dominant model of course evaluation treats students as data sources: they answer, the institution analyses, and the loop rarely closes back to them. The students-as-partners literature — anchored by Healey, Flint and Harrington's (2014) Engagement Through Partnership — argues that engaging students as co-inquirers in designing, interpreting and acting on evaluation produces both better data and stronger engagement. The evidence base is more conceptual and case-study than experimental, so partnership should be read as a well-theorised, values-driven practice rather than a proven metric-mover. Done well, it directly addresses the single most corrosive fact about course evaluation: students stop responding honestly, or at all, when they believe no one listens.

What the research says

The anchor is Healey, Flint and Harrington (2014), Engagement Through Partnership: Students as Partners in Learning and Teaching in Higher Education (York: Higher Education Academy). Drawing on a wide international evidence base, the report reframes student engagement from participation (students taking part in activities others design) to partnership — "a relationship in which all involved are actively engaged in and stand to gain from the process of learning and working together." It sets out four overlapping areas where students can be partners, one of which is directly relevant here: learning, teaching and assessment, including the design and interpretation of feedback and evaluation. Partnership, the authors stress, is a process and an ethos built on reciprocity, not a specific technique — and it explicitly rejects the consumerist framing in which students are customers rating a service.

The complementary practical guide is Cook-Sather, Bovill and Felten (2014), Engaging Students as Partners in Learning and Teaching: A Guide for Faculty (San Francisco: Jossey-Bass). They define student–faculty partnership as "a collaborative, reciprocal process through which all participants have the opportunity to contribute equally, although not necessarily in the same ways" to pedagogical decision-making and inquiry, and anchor it in three principles — respect, reciprocity and responsibility (trust). Their operational advice is deliberately modest: start small, keep participation voluntary, be clear about shared aims, and learn from mistakes.

Why this connects to the evaluation-methods evidence: a recurring, well-replicated finding is that students are willing to give feedback but doubt it is used (the Spencer and Schmelkin tradition), and that the strongest predictor of whether evaluation improves teaching is whether the loop is genuinely closed with consultation and visible action. Partnership is, in effect, the structural answer to that finding — it builds the "we heard you, here is what changed, help us decide what next" cycle into the design rather than bolting it on afterward.

Why it matters for course evaluation in practice

It attacks the response-rate and honesty problem at its root. Declining response rates and satisficing are, in large part, a trust problem: students who believe feedback vanishes into a void either stop responding or answer carelessly. When students have helped design the questions and can see prior feedback acted upon, the instrument stops being an imposition and becomes something they have a stake in. This is a more durable lever than another reminder email or a prize draw.

It improves the questions themselves. Students often understand better than administrators which items are ambiguous, double-barrelled or irrelevant to their actual experience. Involving them in item design — or cognitively pre-testing items with them — surfaces problems before the instrument is fielded, improving validity at source.

It changes what "closing the loop" means. In a partnership model, students are not just told what changed; they help interpret the results and prioritise responses. That shifts evaluation from a summative audit toward a formative, developmental conversation — closer to the mid-semester-feedback and utilization-focused traditions that the evidence shows actually improve teaching.

It reframes the student, and that matters for data quality. Healey et al. warn specifically against the student-as-consumer framing. That framing is not just philosophically contested; it is associated in the wider literature with more transactional, grade-focused evaluation behaviour. Positioning students as partners in a shared enterprise is a plausible counterweight.

Limitations and honest caveats

• The evidence is conceptual and case-based, not experimental. The partnership literature is rich in frameworks, principles and illustrative cases, but thin on randomised or quasi-experimental studies showing that partnership causally raises response rates, validity or learning. It should be presented as a well-argued, values-aligned practice with promising case evidence — not as a proven metric intervention.
• Selection and representativeness. The students who volunteer to partner are, by definition, a self-selected and often already-engaged minority. Their priorities may not represent the quieter majority, and partnership can inadvertently amplify the already-heard. Deliberate effort is needed to bring in students who would not normally step forward.
• Power and tokenism. Partnership can slide into tokenism — consulting students for legitimacy while real decisions stay elsewhere. Cook-Sather et al.'s emphasis on genuine reciprocity is precisely a guard against this, but it is easy to breach in practice.
• Cost and scalability. Deep partnership is resource-intensive and hard to scale across every module in a large institution. It is often best deployed at the level of programme-level evaluation design and interpretation rather than for every individual course each cycle.
• It does not fix measurement bias. Partnership improves relevance, trust and use; it does not, on its own, remove the gender, attractiveness, discipline or workload confounds documented elsewhere. It is complementary to good psychometrics, not a substitute.

How Koji incorporates this

Koji for Education is built around the same reciprocal, loop-closing logic that the partnership literature calls for.

• Voice, not just a rating. Koji's AI-moderated conversational interviews give students genuine agency: instead of ticking boxes someone else wrote, students explain their experience in their own words and are asked thoughtful follow-ups. The experience is closer to being consulted than being surveyed — the felt difference partnership depends on.
• Student-informed question design. Koji supports the full range of structured item types (open_ended, scale, single_choice, multiple_choice, ranking, yes_no), which makes it practical to co-design an instrument with a student panel and pilot items before fielding them — the item-quality benefit partnership delivers.
• Closing-the-loop action tracking. Koji is designed to turn feedback into visible follow-up: themes are surfaced through automatic thematic analysis, actions can be tracked against them, and results can be shared back with the cohort. This operationalises the "you said, we did — now help us prioritise" cycle that the Spencer and Schmelkin finding shows is decisive for engagement.
• Formative and mid-cycle collection. Because Koji supports mid-semester and formative rounds, not only end-of-term summative surveys, it enables the developmental, in-course conversation that partnership favours over a one-off retrospective verdict.
• Programme-level triangulation. Koji can aggregate across cohorts and cycles, supporting the programme-level, students-as-co-interpreters model that is more scalable than course-by-course partnership.

Koji does not claim that a tool creates partnership — partnership is a relationship and an ethos, and no software supplies that. What Koji provides is the infrastructure that makes genuine partnership practical at scale: a voice-giving instrument, co-designable questions, and a real closing-the-loop mechanism. Institutions that extend this participatory ethos to their wider research use Koji's core platform at koji.so, applying the same AI-moderated interview engine to customer and product studies where treating participants as partners, not data points, similarly improves what you learn.

Getting started: a minimal viable partnership

Partnership does not require an institution-wide programme to begin. Cook-Sather, Bovill and Felten's advice — start small, keep it voluntary, be clear about shared aims — points to a lightweight first step that most programmes can run within a single cycle. Convene a small, deliberately diverse student panel (not only the usual keen volunteers) to review the draft evaluation instrument before it is fielded: which items are ambiguous, which miss what students actually care about, which feel like box-ticking. After the results are in, bring the same panel back to help interpret the themes and prioritise which two or three actions the programme will commit to. Then — and this is the step most institutions skip — publish a short "you said, we did, and here is what we are still deciding" note back to the whole cohort before the next round. Even this minimal loop converts students from anonymous data sources into stakeholders with a visible return on their effort, which is precisely the mechanism the evidence links to sustained, honest engagement. Scale up only once the loop is genuinely closing.

Related resources

• Does Closing the Feedback Loop Actually Matter? The Evidence on Acting on Student Evaluations
• Students Are Willing to Evaluate — They Just Doubt Anyone Listens
• Do Student Evaluations Actually Improve Teaching? The Feedback-Intervention Evidence
• What Can Open-Text Student Comments Tell You That Likert Scores Cannot?
• Designing Course Evaluation for Use: The Utilization-Focused Approach
• How to Design Open-Ended Course Evaluation Questions That Get Useful Answers

References

• Healey, M., Flint, A., & Harrington, K. (2014). Engagement Through Partnership: Students as Partners in Learning and Teaching in Higher Education. York: Higher Education Academy (Advance HE). https://www.advance-he.ac.uk/knowledge-hub/engagement-through-partnership-students-partners-learning-and-teaching-higher
• Cook-Sather, A., Bovill, C., & Felten, P. (2014). Engaging Students as Partners in Learning and Teaching: A Guide for Faculty. San Francisco: Jossey-Bass.
• Bovill, C., Cook-Sather, A., Felten, P., Millard, L., & Moore-Cherry, N. (2016). Addressing potential challenges in co-creating learning and teaching: overcoming resistance, navigating institutional norms and ensuring inclusivity in student-staff partnerships. Higher Education, 71(2), 195–208. https://doi.org/10.1007/s10734-015-9896-4