The Satisfaction Trap: What the Student-as-Consumer Model Does to Course Evaluation

Bottom line up front: The dominant framing of the modern university — students as fee-paying consumers buying a service — has reshaped course evaluation into something closer to a customer-satisfaction survey. That is a problem, because the goals of a good education and the goals of a satisfied customer are not the same, and there is evidence that a stronger consumer orientation is associated with lower academic performance. If your evaluation instrument rewards comfort, ease, and delivered expectations, it will quietly push teaching away from the difficulty that produces learning. The answer is not to stop listening to students; it is to stop mistaking satisfaction for quality.

How marketization rewired the question

Across much of European and Anglophone higher education, the past two decades installed a consumer relationship at the centre of the student experience: tuition fees, league tables, "student experience" as a competitive metric, and — in the UK — an explicit regulatory recognition that students are consumers protected by consumer law. The Competition and Markets Authority's Consumer law advice for higher education providers (2015, updated 2023) formalised institutions' obligations to students-as-consumers around information, fair terms, and complaints handling. Whatever its merits for protecting applicants, this framing has a gravitational pull on evaluation: if the student is a customer, the natural instrument is a satisfaction survey, and the natural summary statistic is a satisfaction score.

Sociologists of higher education anticipated the tension. Molesworth, Nixon and Scullion, in work on The marketisation of higher education and the student as consumer (2009), argued that consumerism nudges students from a mode of being (transformation, becoming a more capable person) toward a mode of having (acquiring a degree, a grade, a credential). Course evaluation sits precisely at the point where that shift becomes measurable — and enforceable.

The uncomfortable evidence: consumer orientation and worse outcomes

The most cited empirical anchor here is Louise Bunce, Amy Baird and Siân Jones, The student-as-consumer approach in higher education and its effects on academic performance (2017, Studies in Higher Education, 42(11), 1958–1978). Surveying 608 undergraduates in England, they found that a stronger consumer orientation was associated with lower academic performance, and that consumer orientation mediated the usual relationship between learner identity and achievement. Their blunt summary — reported widely, including in Times Higher Education — is that degrees cannot be bought; they require effort and engagement that a consumer stance actively undermines.

If that finding holds, it has a sharp implication for evaluation. A student in a strong consumer frame evaluates a course the way they would evaluate a purchase: did it deliver what I expected, with minimal friction, for my money? But friction — desirable difficulty, cognitive effort, being made to struggle productively — is often exactly what learning requires. We have documented the specific version of this in the classroom: the active-learning penalty, where students learn more from effortful methods yet rate them lower because they feel harder. Consumerism sharpens that penalty into an institutional incentive.

Satisfaction is a real signal — just not the one people think

Here is the strongest counterargument, and it needs to be met honestly: are we not just being snobs about student happiness? Surely a miserable course is a bad course, and satisfaction data catches that.

Partly, yes. Satisfaction is not worthless. A course that leaves students confused, unsupported, and demoralised is failing, and satisfaction data will often (not always) register it. Students are experts on their own experience — whether they could get help, whether assessment was fair, whether the workload was survivable. Dismissing that as mere consumer whining is both arrogant and empirically wrong.

The error is not listening to satisfaction; it is equating it with quality and optimising for it. Satisfaction measures whether expectations were met (see the expectation gap) — a real thing, but not the same as learning, and sometimes inversely related to it. When a satisfaction score becomes a target for promotion, funding, or league-table position, Goodhart's Law takes over: the measure stops being a good measure. The rational response for a rated academic is to reduce difficulty, inflate grades, and manage expectations — none of which serves the student's actual interests. The UK's reform of the National Student Survey — moving away from a blunt "satisfaction" item — is a sector-level admission that the framing had gone wrong.

So the honest position is neither "ignore students" nor "satisfaction is truth". It is: satisfaction is one input, it is systematically biased toward comfort, and building your whole quality apparatus on it institutionalises the consumer frame the evidence warns against.

From customers to partners — and what to measure instead

The constructive alternative reframes students not as customers rating a service but as partners in a shared enterprise of improvement. That is not a slogan; it changes what you ask. Instead of "How satisfied were you?", you ask what students found difficult and why, whether the difficulty felt purposeful, what they can now do that they could not before, and what specifically would improve the course. Those questions treat the student as a witness to their own learning rather than a consumer scoring a transaction. They are also far harder to game by lowering standards.

The European quality-assurance angle

There is a specific tension worth naming for a European audience. The Standards and Guidelines for Quality Assurance in the European Higher Education Area (the ESG, adopted in 2015) frame students not as customers to be satisfied but as active participants in a shared quality culture — partners in shaping and improving their own education. Student involvement is written into the standards as a matter of governance and design, not as a satisfaction metric to be maximised. The consumer-satisfaction framing sits in quiet tension with that ideal: one treats the student as a co-producer of learning, the other as the end-user of a delivered product.

This matters because the two framings pull evaluation instruments in opposite directions. A quality-culture orientation wants evidence that supports dialogue and improvement — rich, specific, oriented toward what to change. A consumer orientation wants a comparable score for accountability and marketing. Many European institutions have, in practice, imported the metric of the second while professing the values of the first, and course evaluation is where that contradiction is most visible: a satisfaction number dressed in the language of partnership. National quality bodies that emphasise enhancement over mere compliance are, in effect, asking institutions to resolve the tension in favour of the ESG's participatory model — which means collecting feedback that improves teaching rather than feedback that merely ranks it. The instrument has to match the philosophy, and a satisfaction score does not match a quality-culture philosophy no matter how it is branded.

Where Koji fits

A satisfaction number is the native output of the consumer model: fast, comparable, and exactly wrong for the job. Koji for Education is designed to collect the partner-model evidence instead.

Rather than reducing a student to a satisfaction score, Koji runs AI-moderated conversational interviews that ask what happened and why — probing whether a hard course was hard in a productive way or a badly designed one, a distinction a Likert item cannot make. Its six structured question types (open-ended, scale, single- and multiple-choice, ranking, yes/no) let you ask about learning, effort, and specific improvements rather than only global satisfaction, while automatic thematic analysis turns thousands of open responses into the actionable detail that a mean satisfaction score erases. Because the moderation is standardised and bias-aware, it does not simply reward the courses that felt easiest. And closing-the-loop action tracking operationalises the partner model directly: students see that their feedback drives change, which is the antidote to the transactional, one-way consumer relationship. Koji cannot make satisfaction stop mattering — nor should it — but it stops satisfaction from being the only thing you measure and optimise.

The same conversational engine powers the main Koji platform for customer and user research, where, tellingly, the mature practice long ago moved past a single satisfaction score toward understanding the why behind it.

If your course evaluation is really a customer-satisfaction survey wearing an academic badge, you are optimising for the wrong thing. See how Koji measures learning, not just satisfaction.