Your Nursing Students Spend Half Their Degree on Placement. Your Evaluation Never Goes There.

Bottom line: A nursing or midwifery student spends roughly half of their degree not in a lecture theatre but on clinical placement — under EU rules, at least 2,300 of the minimum 4,600 training hours must be clinical. That placement, supervised and assessed under the Nursing and Midwifery Council's 2018 framework, is where a student becomes safe to register. Yet the instrument most universities use to evaluate the programme — an end-of-module satisfaction survey built for a taught module — almost never asks about the practice learning environment, the quality of supervision, or the student's readiness to practise. You are evaluating the half of the degree that is easiest to survey and ignoring the half that decides whether your graduate is fit to care for patients.

The regulated shape of a nursing degree

Nursing is one of the professions that receives automatic recognition across the EU under Directive 2005/36/EC. Article 31 sets a hard floor: the training of a nurse responsible for general care comprises at least three years of study or 4,600 hours of theoretical and clinical training, with clinical training making up at least one half of that minimum. In practice, a European nursing programme is close to a 50/50 split between the classroom and the ward, the community placement, the care home, the clinic. This is the same regulated-profession logic that governs medicine and pharmacy: the qualification is a passport, and the placement is where it is earned.

The classroom half you evaluate to death. The clinical half — the harder, higher-stakes, more variable half — usually gets a single line on a module survey, if that.

What the 2018 supervision model actually changed

In the UK, the NMC's 2018 Standards for student supervision and assessment (SSSA) replaced the old mentorship model. Three roles now shape a student's placement: a practice supervisor who supports day-to-day learning, a practice assessor who assesses practice and cannot be the same person as the supervisor, and an academic assessor at the university who collates evidence and confirms progression. The standards also removed the old requirement that a mentor spend 40% of their time working directly alongside the student.

That is a profound change in how learning happens on placement — and it introduced new failure modes worth evaluating. Is the supervisor–assessor separation working, or has it become a paperwork exercise? Is the student getting consistent supervision, or being passed between staff who never see enough of them to judge? Has removing the 40% co-location rule improved flexibility or diluted the relationship? None of that is visible in "How satisfied were you with this module?"

The practice learning environment is a measurable thing — you are just not measuring it

Nursing education research has spent two decades building instruments to capture exactly what a satisfaction survey cannot. The most widely used, the Clinical Learning Environment, Supervision and Nurse Teacher (CLES+T) scale, measures the pedagogical atmosphere of a placement, the supervisory relationship, the role of the nurse teacher, and the ward's leadership — because these, not the lecturer's slides, are what determine whether a placement teaches. The existence of a validated, internationally used placement-environment instrument is itself the argument: the sector already knows the practice environment is the variable that matters, and already knows the generic module survey does not reach it.

The stakes are not abstract. The placement is where the theory–practice gap opens or closes, where students form professional identity, and — bluntly — where many decide whether to stay in the profession or leave it. A programme that cannot see which placement sites are supportive and which are corrosive is flying blind on its single biggest retention and safety lever. This is the same problem faced by every programme where the workplace is half the curriculum, only with patient safety on the line.

"But students cannot assess clinical competence"

The strongest objection to student evaluation in nursing is a fair one: a second-year student is not qualified to judge whether their own clinical practice is safe, and is certainly not a reliable rater of a supervisor's clinical expertise. That is true, and any honest system has to concede it. Self-assessed competence is weakly correlated with demonstrated competence, and a student mid-placement is poorly placed to certify their own proficiency — which is precisely why the NMC hands assessment to a trained practice assessor, not to the student. The same caution applies in medical education, where programmatic assessment deliberately refuses to let any single self-report decide anything.

But that objection defeats a claim nobody serious is making. The student is not the judge of clinical competence. The student is the only continuous observer of the learning environment: whether they were supervised or abandoned, whether they were included in the team or used as a spare pair of hands, whether they felt able to raise a safety concern, whether feedback actually happened. On those questions the student is the primary source, and the practice assessor is not in the room to see them. The fix is not to stop asking students; it is to ask them the right questions, triangulate their answers with assessor and outcome data, and keep the two data streams distinct.

A second real objection: placement feedback is rarely anonymous in practice. On a ward with two students, a critical comment identifies its author to the very assessor who signs their competency document. Fear of reprisal silences exactly the feedback that matters most. This is a design problem, and it has design answers: collect formatively and mid-placement rather than only at the end, route feedback through a channel the local assessor does not control, and report at the site and programme level rather than the individual-student level.

Where Koji fits

Koji was built for the kind of evaluation the placement demands. Its AI-moderated conversational interviews probe beyond a number — when a student says supervision was "fine," the interview follows up on what "fine" meant, surfacing the specifics a Likert item flattens. Because moderation is standardised, every student gets the same careful, bias-aware follow-up, with no local assessor in the room to chill the conversation. Mid-placement, formative collection catches a failing placement while there is still time to intervene, rather than in a retrospective survey after the student has moved on. Automatic thematic analysis turns hundreds of open-text placement accounts into site-by-site patterns, and programme- and institution-level reporting lets a lead compare placement environments across a whole partnership of trusts and clinics. Closing-the-loop tracking records what was changed at a site and whether the next cohort noticed. Data handling is GDPR/AVG-compliant, which matters acutely when feedback touches named staff and identifiable incidents.

Koji does not replace the practice assessor or claim to measure clinical competence — it surfaces the environment and the relationship that determine whether competence can develop. Many of the same teams also run wider stakeholder and staff research on the main Koji platform, which shares the same AI interview engine.

The nursing and midwifery degree is half classroom, half practice. It is time the evaluation stopped being all classroom.

Ready to evaluate the placement, not just the lecture? See how Koji for Education helps nursing and midwifery programmes hear the practice learning environment — and act on it.

Frequently asked questions

Does EU law really require half of a nursing degree to be clinical? Yes. Directive 2005/36/EC, Article 31, sets a minimum of three years or 4,600 hours of theoretical and clinical training for a nurse responsible for general care, of which clinical training must be at least one half — roughly 2,300 hours on placement.

What changed under the NMC's 2018 SSSA standards? Mentorship was replaced by three distinct roles: a practice supervisor, a practice assessor (who cannot be the same person as the supervisor), and an academic assessor at the university. The old requirement that a mentor work directly alongside the student for 40% of the time was removed.

Can students validly evaluate placements if they cannot judge clinical competence? They cannot certify their own competence — that is the practice assessor's job. But they are the only continuous observers of the learning environment: supervision, inclusion, psychological safety, and whether feedback happened. Ask them about those, and triangulate with assessor and outcome data.

Why not just reuse the end-of-module survey for placements? It was designed for taught modules. It does not capture the supervisory relationship, the practice environment, or fitness-to-practise readiness, and it typically arrives too late to rescue a failing placement.

How does Koji handle the anonymity problem on small placements? Standardised AI moderation with no local assessor in the loop, formative mid-placement timing, and reporting at the site and programme level rather than the individual-student level — so honest feedback does not identify its author to the person signing their competency document.