Turning Course Evaluations into NVAO Accreditation Evidence (Netherlands & Flanders)

Short answer: The NVAO (Nederlands-Vlaamse Accreditatie Organisatie) accredits higher-education programmes in the Netherlands and Flanders against frameworks that reflect the European Standards and Guidelines (ESG 2015). Course evaluations are not assessed for their own sake; they are evidence that a programme systematically gathers stakeholder feedback, reflects on it, and acts — the quality cycle. This guide maps NVAO standards to concrete evaluation outputs and shows how a structured, AI-moderated approach like Koji produces audit-ready, longitudinal evidence. It also notes honestly where a traditional survey tool may be sufficient.

How NVAO assessment is structured

NVAO operates two linked levels:

• Institutional level. In the Netherlands, the Instellingstoets Kwaliteitszorg (ITK, institutional quality-assurance audit) examines whether the institution has an effective, self-correcting quality-management system. In Flanders, an institutional review plays an analogous role. A central expectation is that the institution organises effective feedback from stakeholders and uses it to realise its policy — in other words, a working closing-the-loop mechanism.
• Programme level. For an existing programme, the limited assessment framework in the Netherlands contains four standards: (1) intended learning outcomes, (2) teaching-learning environment, (3) student assessment, and (4) achieved learning outcomes. (Institutions without a passed ITK are assessed against an extended set of standards.)

Student-evaluation evidence is most relevant to Standard 2 (teaching-learning environment) and, more indirectly, to Standard 4 (achieved learning outcomes) and the institutional closing-the-loop expectation. The NVAO frameworks are explicit that quality assurance is cyclical: an ongoing dialogue with students, alumni, staff and the professional field, drawing on both quantitative and qualitative feedback, followed by reflection and action.

What an NVAO panel actually looks for

Assessment panels are made up of peers — including a student member — who weigh evidence against the standards. For the parts touching student feedback, panels typically want to see:

• that feedback is collected systematically, not ad hoc;
• that it is representative (not undermined by low or biased response);
• that the programme reflects on what the feedback means;
• that reflection leads to concrete actions, and that those actions are followed up in the next cycle (the loop is closed);
• that the evidence is traceable over time, so a panel can see a multi-year pattern rather than a one-off snapshot.

This is where many programmes struggle: they have plenty of Likert averages but a thin trail showing what changed as a result.

Mapping NVAO expectations to Koji outputs

NVAO expectation	What a panel wants to see	Concrete Koji output
Standard 2 — teaching-learning environment	Systematic, representative student feedback on teaching and curriculum	Standardized AI-moderated interviews run every term, with consistent probing across all students
Cyclical quality assurance (institutional / ITK)	Dialogue then reflection then action then follow-up	Themes linked to tracked actions, with status carried into the next cycle
Representativeness	Response rates and non-response addressed	Engagement-friendly conversational format; response-rate reporting per cohort
Standard 4 — achieved learning outcomes	Student perspective on attainment, alongside other evidence	Exit and programme-level interviews probing perceived attainment and gaps
Reflection quality	Genuine qualitative depth, not just scores	Automatic thematic analysis surfacing the why behind ratings
Traceability over time	Multi-year, comparable evidence	Longitudinal cohort reporting on stable themes
Stakeholder breadth	Students, alumni, professional field	Same interview engine reused for alumni and, via the main Koji platform, other stakeholder research

The point is not that an evaluation tool "passes" accreditation — panels assess the programme, not the software. The point is that the right tool makes the programme quality cycle legible: it turns scattered feedback into a documented, standardized, longitudinal trail that a panel can follow.

Why closing the loop is the hard part

NVAO frameworks (like the ESG they reflect) put weight on acting on feedback, not merely gathering it. A spreadsheet of end-of-term averages does not show a loop. What shows a loop is a record like: theme identified (assessment feedback too slow) then action taken (introduced a rubric and a 10-day return deadline) then re-measured next cohort (concern fell from a dominant theme to a minor one).

Koji is built around that record: each theme carries forward as a tracked action, so the closing-the-loop documentation is a by-product of normal operation rather than something reconstructed under deadline before a site visit. For the conceptual background, see our companion guide on turning student feedback into ESG / ENQA accreditation evidence, which covers the ESG standards that NVAO mirrors, and the AACSB & EQUIS guide for business-school programmes pursuing dual accreditation.

Representativeness: a real NVAO risk

Panels are sensitive to whether feedback represents the cohort. Very low response rates, or responses skewed toward the most satisfied or most aggrieved, weaken the evidentiary value of any evaluation — a problem we examine in response rates and non-response bias. Conversational formats tend to engage students who would skip a long grid of Likert items, but no method guarantees representativeness; report response rates honestly per cohort and triangulate with other evidence (assessment data, alumni surveys, employer input). Over-claiming representativeness in a self-evaluation report is a credibility risk with an evidence-literate panel.

Where a traditional survey tool may be enough

Honesty matters with this audience. If your programme already runs a validated, well-understood Likert instrument, achieves healthy response rates, and — crucially — has a disciplined manual process for documenting actions and follow-up, a conventional tool such as EvaSys or Qualtrics can satisfy NVAO requirements. Accreditation does not mandate AI. The case for a tool like Koji is stronger when:

• your qualitative evidence is thin because students rarely write in comment boxes;
• your closing-the-loop trail is hard to assemble at audit time;
• you want comparable, standardized evidence across many programmes;
• you are moving from summative-only ratings toward formative, improvement-oriented feedback.

If none of those apply, a survey platform plus rigorous documentation may be the pragmatic choice. We compare approaches in Koji vs EvaSys and Koji vs Qualtrics.

A practical preparation checklist

1. Map your evidence to the four standards before the panel does — explicitly tag which evaluation outputs speak to Standard 2 and Standard 4.
2. Show the cycle, not the snapshot. For two or three priority themes, document theme then action then re-measurement across at least two cohorts.
3. Report response rates honestly per programme and per cohort; explain mitigations.
4. Keep it GDPR-clean. NVAO operates within the EU; ensure data handling, consent and retention are documented (Koji is built EU-first for this reason).
5. Triangulate. Pair student-voice evidence with assessment and alumni data so no single source bears all the weight.

Flanders, programme review and revalidation

While the Netherlands and Flanders share NVAO and the same ESG foundation, the procedures differ. In Flanders, much programme-level quality assurance has shifted toward the institutions themselves under a system of programme accreditation tailored to the control an institution exercises, with NVAO conducting institutional reviews. The practical implication is the same: an institution that owns more of its own quality assurance must show an even more convincing internal closing-the-loop system, because there is no external programme panel re-checking every cycle. Standardized, longitudinal student-voice evidence becomes more valuable, not less, when the institution is trusted to self-regulate.

A short worked example shows what panels reward. A Master programme notices, through end-of-term interviews, that international students repeatedly describe group-work assessment as opaque. The programme acts: it publishes a marking rubric, adds a mid-point check-in, and trains tutors on calibrated feedback. The next cohort interviews are re-analysed against the same theme, and the concern drops from a dominant theme to an occasional mention; a residual sub-issue (uneven peer contribution) is logged as the next action. In a self-evaluation report, that single tracked thread — observed, acted upon, re-measured, with a remaining issue carried forward — demonstrates the quality cycle far more convincingly than a page of stable Likert averages. It is also exactly the kind of artefact a student panel member can interrogate and find credible.

Note the documentation burden either way: NVAO does not reward the existence of a tool, only the evidence of a functioning cycle. The advantage of a structured platform is that the artefact is generated continuously rather than assembled retrospectively under deadline, which lowers the risk of gaps surfacing during a site visit.

Related Resources

• Turning Student Feedback into ESG / ENQA Accreditation Evidence
• Course Evaluation Evidence for AACSB & EQUIS Business-School Accreditation
• Response Rates and Non-Response Bias in Course Evaluations
• What Open-Text Student Comments Tell You That Likert Scores Cannot
• Koji vs EvaSys: A Fair Comparison