Your Course Survey Closes in Week 12. The Growing Season Runs All Year.

Bottom line up front: Land-based higher education — agriculture, agronomy, horticulture, land and estate management, agricultural engineering — is taught on a calendar the end-of-semester survey cannot see. A crop cycle, a lambing season or a machinery-maintenance year does not fit inside a twelve-week module, and the compulsory sandwich placement on a commercial farm is where professional competence is actually forged. A generic satisfaction score measures the classroom weeks and misses the seasons and the placement that decide whether a graduate can do the job. This is not an argument against student feedback; it is an argument for feedback shaped to the way the discipline is actually taught.

The sector you are evaluating for

Land-based education serves an industry that is far larger than its cultural profile suggests. The UK agri-food sector contributed £146.7 billion — 6.5% of national Gross Value Added — in 2022, and employed just under 4.4 million people, around 14% of all employees in Great Britain, in 2023, according to DEFRA's Agriculture in the United Kingdom 2023. Across the European Union, agricultural knowledge is treated as strategic infrastructure. The Common Agricultural Policy is built on Agricultural Knowledge and Innovation Systems (AKIS), defined in Regulation (EU) 2021/2115 as "the combined organisation and knowledge flows between persons, organisations and institutions who use and produce knowledge for agriculture and interrelated fields" (EU CAP Network). A programme that trains the people who will work inside that system deserves an evaluation instrument that understands how they learn.

Why the semester survey misreads land-based education

Two structural features of the discipline defeat the standard module survey.

First, seasonality. Most academic quality machinery assumes a self-contained twelve-week teaching block that produces a clean end-point for a satisfaction survey. Agriculture does not run on that clock. A student learns arable agronomy across a sowing-to-harvest cycle, livestock husbandry across a breeding and rearing season, and soil and machinery management across a full year of weather. The pedagogically important events — the disease outbreak in week 3, the failed establishment, the harvest that came in late — are distributed across months that straddle module boundaries, vacations and, often, the placement year. Close the survey in week 12 and you sample the lecture theatre while the field, the herd and the glasshouse are still teaching.

Second, the placement year. In specialist land-based providers the commercial placement is not an optional enhancement; it is the spine of the degree. Harper Adams University requires its main-campus undergraduates to complete "365 days (or equivalent) spent on compulsory placement," and describes itself as number one in the UK for Graduate Outcomes — against a sector average of 53.4% of full-time first-degree graduates in full-time employment fifteen months out (Harper Adams, 2024 figures). The Royal Agricultural University, founded in 1845 as the first agricultural college in the English-speaking world and granted full university status in 2013, is built on the same premise. If the placement year is where employability is manufactured, an instrument that never surveys it is measuring the wrong twelve months. This is the same blind spot we have flagged for work-integrated learning and internships and for hospitality and tourism industry placements.

The signature pedagogy is "learning by doing"

The mismatch is not incidental; it is baked into the discipline's pedagogy. In their study for Renewable Agriculture and Food Systems, Valley, Wittman, Jordan, Ahmed and Galt (2018, 33(5), 467–480) identify an emerging signature pedagogy for sustainable food-systems education built on four themes: systems thinking, cross-disciplinary integration, experiential learning, and participation in collective-action projects. Three of those four are, by definition, not lecture-hall activities — and none of them are captured by a Likert item asking whether the lectures were well organised. When the field is the classroom, a survey designed for classrooms under-samples the curriculum.

Competence is certified after the degree — and it never stops

Perhaps the strongest signal that the degree is a beginning, not an end-point, is that the sector certifies professional competence after graduation, on a continuing basis. Lantra, the awarding body for the land-based and environmental industries, has developed qualifications for the sector's workforce for over forty years. BASIS runs a Professional Register that keeps crop-protection advisers current through audited continuing professional development, and administers FACTS, the Fertiliser Advisers Certification and Training Scheme. An agricultural engineer becomes a Chartered Environmentalist or Chartered Engineer through the Institution of Agricultural Engineers — a route requiring a minimum of four years' relevant experience plus a commitment to ongoing CPD. None of these bodies asks whether the graduate enjoyed a module. They ask whether the practitioner can demonstrate current competence. Course evaluation should be feeding that trajectory, not sitting apart from it. It is the same "the market grades the degree" pattern we have described for actuarial science and regulated professions under Directive 2005/36/EC.

But isn't this just special pleading?

The strongest objection is that every discipline believes it is a special case, and if quality assurance indulged them all, cross-institutional comparability would collapse. There is real force to this. A regulator or a benchmarking exercise needs comparable signals, and "our subject is different" is exactly what a weak programme says to dodge scrutiny.

Two responses. First, the claim here is not that land-based education should be exempt from evaluation — it is that the timing and unit of evaluation should follow the pedagogy. Comparability is preserved by asking comparable questions (about workload, support, feedback quality and belonging); it is the schedule that needs to change, not the standards. Second, the seasonality and placement-centrality of land-based education are documented features of the field, not rhetorical conveniences. Adjusting when and where you collect feedback to match a year-long cycle is not lowering the bar; it is aiming the instrument at the thing it is supposed to measure. A survey that fires in week 12 and never again is not more rigorous for being ill-timed.

What to measure instead

• Collect across the cycle, not at a single end-point. Run short formative touchpoints at pedagogically meaningful moments — after a key seasonal operation, mid-placement, on return from placement — rather than one retrospective survey after teaching stops.
• Survey the placement itself. Ask about the quality of workplace supervision, the range of tasks, the integration between placement and academic study, and whether the student could apply what they learned — the placement is half the degree and deserves its own evidence.
• Ask about applied competence, not just satisfaction. "Could you carry out this operation unsupervised?" is a more useful question for a land-based programme than "How satisfied were you with this module?"
• Read the free text thematically. Seasonal and placement learning surfaces in stories — the failed establishment, the difficult lambing, the machinery breakdown — not in scale points.

Where Koji fits

Legacy survey tools were built to average a Likert score at the end of a semester. That is precisely the shape of instrument this discipline defeats. Koji for Education runs AI-moderated conversational interviews that can be scheduled formatively — mid-placement, on return, after a seasonal operation — and probe beyond a number to ask what a student could actually do, and what the placement taught that the classroom could not. Its automatic thematic analysis reads the seasonal and placement stories buried in open text at scale, and its programme-level reporting connects module-, placement- and year-level signals so a course team can see the whole cycle rather than a single week-12 snapshot. The standardized, bias-aware moderation stays consistent across cohorts and placement types where a human interviewer could not. Koji mitigates and surfaces; it does not pretend to eliminate the judgement a course team still has to exercise. (The same AI interview engine powers general user and customer research on the main koji.so platform.)

The point is not that land-based education is too special to evaluate. It is that evaluating it well means following the growing season and the placement year — the parts of the degree the semester survey was never built to see.

Frequently asked questions

Why can't a standard end-of-semester survey work for agriculture programmes? Because the learning it is meant to measure is distributed across a full seasonal cycle and a year-long commercial placement, not a self-contained twelve-week block. A survey that closes in week 12 samples the lecture weeks and misses the harvest, the breeding season and the placement that decide whether a graduate is competent.

Isn't asking for different timing just a way of avoiding scrutiny? No. The proposal keeps the standards and the comparable questions (workload, support, feedback, belonging) and changes only when and where feedback is collected, so that it aligns with a year-long pedagogy. Aiming an instrument at the right months is more rigorous, not less.

What does the placement year have to do with course evaluation? In specialist land-based providers the placement is the spine of the degree — Harper Adams requires 365 days of compulsory placement — yet most survey systems never ask about it. Surveying supervision quality, task range and placement-to-study integration captures half the degree that otherwise goes unmeasured.

How is professional competence in this sector actually judged? After graduation, on a continuing basis: the BASIS Professional Register and FACTS keep advisers current through audited CPD, and chartered status through the Institution of Agricultural Engineers requires years of demonstrated experience plus ongoing development. The degree is the start of that trajectory, and evaluation should feed it.

Can AI-moderated interviews really help with seasonal, placement-based learning? They can be scheduled at pedagogically meaningful moments rather than one retrospective end-point, and they probe what a student could actually do rather than only how satisfied they were. Automatic thematic analysis then reads the seasonal and placement stories in open text at a scale manual coding cannot match.

Does Koji claim to remove bias from evaluation? No. Koji is careful to say it mitigates and surfaces — through standardized, bias-aware moderation and thematic analysis — rather than eliminating bias or replacing the course team's professional judgement.

---

Evaluating a land-based or agriculture programme and want feedback that follows the season, not the semester? See how Koji for Education works.