The Self-Evaluation Report (SER): Turning Course Evaluation Evidence into Accreditation-Ready Documentation

The self-evaluation report — variously called the self-assessment report (SAR), self-evaluation report (SER), or critical reflection — is the single most important document a programme or institution produces for an external quality review. In almost every European accreditation and periodic-review process, the review panel forms its judgement primarily from the SER plus a site visit. And in nearly every SER, student and course evaluation evidence appears repeatedly: to demonstrate that you listen to students, analyse what they say, and act on it. This guide shows exactly where evaluation evidence belongs, maps it to the ESG standards reviewers check, and explains how to make that evidence survive a skeptical panel.

One clarification up front: there are two different "self-assessment reports" in the European system. Agencies write an agency-level SAR when they seek ENQA membership or EQAR listing (assessed against ESG Parts 2 and 3). This guide is about the other one — the programme- or institution-level SER you submit to an agency (NVAO, ANECA, HCERES, ASIIN, QAA-affiliated bodies, etc.) for accreditation or revalidation, assessed against ESG Part 1. That is the document deans, QA directors and programme leaders actually prepare.

What the SER is — and what panels do with it

The SER is a structured, evidence-backed reflection in which a programme describes its aims, provision and quality processes, then honestly analyses how well they work and what it is doing to improve. Reviews are based on the information in the SER and an external assessment that normally includes a site visit; the strongest SERs are analytical rather than descriptive and are written collaboratively, drawing on staff, students and management. Panels are not looking for a marketing document. They are looking for evidence of a functioning quality cycle: you collect data, you interpret it, you decide, you act, and you check the action worked.

Course evaluation is the most continuous, student-facing data stream feeding that cycle — which is why weak evaluation evidence is one of the fastest ways to lose panel confidence.

Where course evaluation evidence maps to the ESG

The Standards and Guidelines for Quality Assurance in the European Higher Education Area (ESG 2015, ENQA) define what agencies expect. Several Part 1 standards are directly fed by course evaluation evidence:

ESG standard	What it requires	Where course evaluation evidence goes in the SER
1.3 Student-centred learning, teaching & assessment	Delivery that respects and responds to student needs	Evidence that student experience of teaching and assessment is systematically gathered and used
1.6 Learning resources & student support	Adequate, accessible resources and support	Student feedback on resources, workload support and learning environment
1.7 Information management	Collect, analyse and use relevant information, including student satisfaction	Your evaluation data pipeline: what you collect, how it is analysed, how it informs decisions
1.9 On-going monitoring & periodic review	Programmes reviewed regularly involving students and other stakeholders	The heart of the SER — evaluation results, trend analysis, and the actions taken in response

Standard 1.9 is where most SERs live or die: it explicitly asks that programmes are monitored and revised with student involvement, and that the information collected is analysed and acted on. Standard 1.7 asks specifically about how you manage and use student-satisfaction data. If your evaluation evidence is a stack of end-of-term averages with no analysis and no visible actions, you are not meeting the spirit of either standard.

The five failure modes reviewers flag

Across published review reports, the same weaknesses recur. Test your evaluation evidence against them before the panel does:

1. Satisfaction-only data. Mean Likert scores ("4.1/5 overall") show contentment, not learning or specific problems. Panels ask what the numbers mean and what you learned from them.
2. No closing-the-loop evidence. You collected feedback — but did anything change? The most common single gap is an inability to show the action that followed the data, and whether it worked.
3. Low or unreported response rates. A 15% response rate with no acknowledgement of non-response bias invites the panel to discount your evidence entirely.
4. Un-analysed free text. Hundreds of student comments quoted raw, or worse ignored, instead of themed into findings. Reviewers want the themes and representative quotes, not the dump.
5. No longitudinal view. A single snapshot cannot show improvement. Panels look for trends across cohorts and evidence that interventions moved the numbers.

Requirement-to-output mapping

The practical question is: how do you produce evidence that answers all five without a heroic manual effort each review cycle? This is where the instrument you use to evaluate matters. Mapping common accreditation requirements to concrete outputs:

Accreditation requirement	Concrete evidence output	How Koji produces it
Systematic, comparable evidence (1.7)	Standardised evaluation across courses/cohorts	AI-moderated interviews ask every student the same core questions in a neutral, standardised way, reducing framing variance
Analysed qualitative findings (1.9)	Themed findings with representative quotes	Automatic thematic analysis converts open responses into themes and exemplar quotes
Depth beyond satisfaction (1.3)	Specific, probed insight into teaching/assessment	Conversational follow-up probes vague comments in the moment for actionable specifics
Closing the loop (1.9)	Documented action + verification	Built-in action tracking records what changed in response to feedback and whether it worked
Longitudinal / cohort reporting	Trend evidence across terms	Institution-level, cohort-comparable reporting over time
Defensible data handling	Anonymity + GDPR compliance	Designed-in anonymity and EU-hosted, GDPR-first processing

Represent this honestly to your panel: an AI-moderated tool standardises and automates evidence production — it does not manufacture quality. The judgement about what to change, and the genuine act of changing it, remain yours.

A realistic timeline

Evaluation evidence cannot be retrofitted the month before a visit. A workable rhythm:

• 18–24 months out: ensure standardised evaluation runs every term with acceptable response rates; start capturing actions taken in response to feedback.
• 12 months out: build the longitudinal picture — trends per programme, themes across cohorts, and a log of closed-loop actions with outcomes.
• 6 months out: draft the SER analytically, using themes and trends (not raw averages), and cite specific closed-loop examples against ESG 1.3, 1.7 and 1.9.
• Site visit: be ready for the panel to ask students directly whether the actions you claim actually happened. Your evidence and the student voice must agree.

When your existing setup is already enough (honest note)

Not every institution needs a new tool to pass. If your current evaluation process already produces analysed, longitudinal, action-linked evidence with defensible anonymity — for example a well-run EvaSys or Explorance Blue deployment with a disciplined QA office behind it — you may simply need to present it better in the SER. A dedicated conversational platform earns its place when your bottleneck is qualitative depth and closing-the-loop documentation: when you have plenty of scores but little insight, and cannot readily show what changed. Buy for the gap you actually have, not for novelty. The same underlying AI interview engine powers the main Koji platform at koji.so for customer and user research; the education product applies it to course and programme evaluation.

Description versus reflection: what panels actually reward

The most common SER writing error is describing a process instead of reflecting on it. "We run end-of-semester evaluations for all modules" is description. "Our evaluations showed persistent dissatisfaction with assessment-feedback turnaround in Years 1–2; we introduced a 15-working-day feedback policy in 2024, and the following cohort's feedback theme shifted from 'slow feedback' to 'clear rubrics', with the relevant satisfaction item rising accordingly" is reflection — it names the evidence, the action, and the verified effect. Panels reward the second because it demonstrates the full quality cycle in a single paragraph. Aim for every major claim in your SER to follow that arc: evidence → interpretation → action → check.

The student voice must corroborate your SER

During the site visit, panels routinely ask students directly whether the improvements the SER claims actually happened. If your report says you closed the loop but students have never heard of the change, the contradiction damages your credibility more than the original problem would have. Involve students in drafting the relevant sections, and make sure the actions you document are ones students can genuinely recognise. Evidence that the institution and its students tell the same improvement story is the single strongest signal a panel can receive.

Related Resources

• Course evaluation evidence for NVAO accreditation
• Student feedback as ESG accreditation evidence
• Course evaluation evidence for programme review and revalidation
• Course evaluation evidence for UK TEF and QAA
• Course evaluation evidence for AACSB and EQUIS
• Course evaluation evidence for German accreditation

Bottom line

The SER is where your quality culture is tested on paper. Course evaluation evidence runs through it — but only analysed, longitudinal, action-linked evidence with defensible anonymity satisfies an ESG panel. Score sheets do not. Whether you upgrade your instrument or simply present existing data better, build the SER around the quality cycle the standards actually ask for: collect, analyse, act, verify.