The Student Voice in Quality Assurance: ESG-Ready Engagement Evidence

Short answer: European quality assurance treats students as partners, not just as respondents. The European Standards and Guidelines (ESG 2015) expect students to be involved in developing quality-assurance policy, in the design and review of programmes, and in the external review process itself - and reviewers registered with EQAR routinely include a student panel member and meet students during a site visit. Course evaluation is the most systematic, repeatable channel for that "student voice," but only if it produces evidence that students were heard and that something happened as a result. This guide maps the student-engagement expectations in the ESG to concrete course-evaluation outputs, and shows where a structured, AI-moderated approach like Koji strengthens the evidence - with an honest note on where a committee minute or a survey tool already does the job.

Student engagement is a standard, not a nicety

The idea that students co-own quality assurance is written into the EHEA's core documents. It runs through the ESG 2015 in several places:

• Standard 1.1 (Policy for quality assurance) expects internal stakeholders - students among them - to take responsibility for quality and to engage in quality assurance "at all levels of the institution," alongside the involvement of external stakeholders.
• Standard 1.2 (Design and approval of programmes) states that programmes are designed "by involving students and other stakeholders."
• Standard 1.3 (Student-centred learning) asks that programmes are delivered in a way that "encourages students to take an active role in creating the learning process," and that delivery is regularly evaluated and adjusted.
• Standard 1.7 (Information management) lists "students' satisfaction with their programmes" among the information institutions should collect, and adds that "students and staff are involved in providing and analysing information and planning follow-up activities."
• Standard 1.9 (On-going monitoring and periodic review) says programmes are reviewed "involving students and other stakeholders," and that resulting action is communicated to all those concerned.

External quality assurance mirrors this. Part 2 of the ESG expects agencies to involve students in external review, which in practice means a student member on the review panel and a dedicated meeting with students during the site visit. Agencies registered on the European Quality Assurance Register (EQAR) are expected to demonstrate this. So when a panel arrives, "the student voice" is not a soft theme - it is a line of evidence the reviewers are specifically tasked to test.

What reviewers actually look for

A review panel is not satisfied by the existence of a feedback survey. Across ESG-aligned frameworks, three questions recur:

1. Is the student voice systematic and representative? One-off surveys with a 15% response rate, skewed toward the most and least satisfied, are weak evidence. Reviewers want a repeatable mechanism that captures a defensible cross-section of the cohort.
2. Are students partners in the process, or only its subjects? Being surveyed is not the same as being involved in analysing feedback and planning follow-up (Standard 1.7). Panels look for students on programme committees, in review teams, and in the interpretation of results - not just at the end of a questionnaire.
3. Did the loop close, visibly to students? The strongest single piece of evidence is a documented chain: students raised a theme, a change was made, and students were told what changed. "You said, we did" communication is the artefact reviewers ask to see, because it demonstrates that engagement was consequential.

A common failure mode is worth naming, because reviewers see it often. Institutions collect a great deal of feedback, analyse it capably, and even act on it - but the evidence of student partnership is missing from the record. The survey happened, the committee met, the change was made, yet nothing in the file shows students helping to interpret the results or being told what their feedback produced. The mechanism worked; the documentation of engagement did not. Under the ESG that gap reads as a weakness even when the underlying practice is sound, because a panel can only credit what it can see. This is why the form of the evidence - traceable, representative, and visibly returned to students - matters as much as the feedback itself.

Mapping the ESG student-engagement expectations to evaluation outputs

ESG expectation	What the panel wants	Traditional SET survey output	AI-moderated evaluation output (e.g. Koji)
Students engaged in QA policy (1.1)	Evidence students shaped, not just answered	Response data only	Structured student accounts feeding governance, exportable for committees
Students involved in design/review (1.2, 1.9)	Student input into programme changes	Aggregate scores handed to staff	Themed student priorities mapped to specific review decisions
Active role in learning (1.3)	Students helped adapt delivery	Fixed engagement items	Open, standardized probing of what students would change and why
Students analyse information & plan follow-up (1.7)	Students in the loop, not just the sample	Manual analysis by QA office	Shared thematic summaries students and staff can jointly act on
Representative student voice	A defensible cross-section	Low, biased response rates common	Conversational, mobile format designed to raise engagement depth
Closing the loop, communicated (1.9)	"You said, we did," re-measured	Year-on-year mean comparison	Theme -> action -> re-measure tracker with student-facing summaries

Where AI-moderated evaluation genuinely helps

Koji uses the same AI interview engine as the main koji.so research platform, adapted for course and programme evaluation. Three properties map onto the student-engagement evidence a panel wants:

• Depth without an interviewer. A conversational, adaptive format lets students explain why - the reason behind a rating - which is exactly the partner-level input Standards 1.2 and 1.9 ask for. It does this through consistent, standardized probing rather than a human interviewer whose style varies from room to room, which mitigates the bias and comparability problems that make qualitative "student voice" evidence hard to defend across cohorts.
• From comments to committee-ready themes. Automatic thematic analysis turns hundreds of open responses into ranked themes that can go straight to a programme committee or a student-staff forum, so that students and staff are analysing the feedback together (Standard 1.7) rather than the QA office doing it alone weeks later.
• A visible, longitudinal closed loop. Actions can be logged against themes, communicated back to students, and re-measured next cycle - producing the "you said, we did," re-measured record that is the most persuasive student-engagement evidence at a site visit.

Stated precisely: the platform standardizes how the voice is collected, surfaces themes for shared interpretation, and documents the loop. It does not replace student membership on committees or the panel's student meeting - those are governance choices the institution still has to make.

When a survey tool or a committee minute is enough

For this audience, the honest boundary matters. If your student voice already flows reliably through active student-staff liaison committees, well-attended student union representation, and a survey with healthy response rates that your QA office can analyse and act on, you may not need an AI-moderated layer at all - a competent survey platform plus disciplined committee minutes will satisfy the ESG. Traditional tools such as EvaSys, Explorance Blue or Qualtrics are well established in European institutions and integrate deeply with student information systems. AI-moderated evaluation earns its place specifically when qualitative depth and closing-the-loop documentation are your weak points: when response rates or comment quality are too thin to be representative, when hand-coding student comments before every review is the bottleneck, or when you cannot easily show reviewers a traceable, re-measured loop. Match the tool to the gap in your evidence, and be candid in your self-evaluation about which parts of the student voice are already strong.

A practical checklist for student-voice evidence

1. Document the mechanism, not just the results - show the panel how students feed into policy, design, and review, referencing ESG 1.1, 1.2 and 1.9.
2. Report representativeness (response rate and known biases), because an unrepresentative voice is a finding against you.
3. Put students in the analysis, e.g. a joint student-staff review of themes, to satisfy Standard 1.7.
4. Keep "you said, we did" artefacts and show them to students, not just to reviewers.
5. Re-measure each acted-on theme in the next cycle and retain the before/after pair.
6. Brief your student panel representatives so the student meeting at the site visit is consistent with your written evidence.

Frequently asked questions

What do the ESG 2015 say about student involvement in quality assurance?

Student engagement runs through Part 1: students help develop QA policy (Standard 1.1), are involved in programme design and periodic review (1.2, 1.9), take an active role in learning (1.3), and are involved in providing and analysing information and planning follow-up (1.7). Part 2 expects agencies to involve students in external review.

Does an external review panel actually meet students?

Yes. Under ESG-aligned external quality assurance, review teams typically include a student panel member and hold a dedicated meeting with students during the site visit. Agencies registered on EQAR are expected to demonstrate student involvement, so the student voice is a line of evidence reviewers are specifically tasked to test.

What makes course-evaluation data strong student-voice evidence?

Three things: it is systematic and representative (a defensible cross-section, not a skewed low-response survey); students are partners in analysing it, not only its subjects; and it visibly closes the loop, with a documented chain from a student-raised theme to a change to a communication back to students.

Is being surveyed the same as being engaged in quality assurance?

No. ESG Standard 1.7 asks that students are involved in analysing information and planning follow-up, not just answering a questionnaire. Panels look for students on programme committees and in the interpretation of results, and for evidence that engagement was consequential - the you-said-we-did record.

When is a survey tool or committee minute enough for student-voice evidence?

When your student voice already flows reliably through active student-staff committees and a survey with healthy response rates your QA office can analyse and act on, a competent survey platform plus disciplined minutes will satisfy the ESG. AI-moderated evaluation earns its place when qualitative depth or closing-the-loop documentation is the weak point.

Related Resources

• Turning Course Evaluations into ESG Accreditation Evidence
• Course-Evaluation Evidence from Student-Staff Liaison Committees
• Course-Evaluation Evidence for an Accreditation Site Visit and Peer-Review Panel
• Course Evaluation, Learning Outcomes & the Bologna Process
• Institution-Level Evaluation Reporting for QA Audit Evidence