Course Evaluation Evidence for Romanian ARACIS Accreditation: A Buyer's Guide

In one sentence: For ARACIS authorisation, accreditation and five-yearly periodic evaluation, Romanian universities must show that student feedback on teaching is collected systematically through their internal quality-assurance structures and acted upon — and Koji turns that feedback into standardised, longitudinal, exportable evidence that maps cleanly onto Romania's two-tier evaluation model and the European Standards and Guidelines (ESG).

This guide is written for QA directors, vice-rectors for quality, CEAC members, deans and institutional-research leads at Romanian higher-education institutions who are preparing for an ARACIS evaluation and want their student-feedback evidence to be defensible rather than improvised.

What ARACIS is and why its evidence bar is high

The Romanian Agency for Quality Assurance in Higher Education (ARACIS, Agenția Română de Asigurare a Calității în Învățământul Superior) was established in 2005 and is the national body responsible for the external quality evaluation of higher education in Romania. It is a full member of the European Association for Quality Assurance in Higher Education (ENQA) and is registered on the European Quality Assurance Register for Higher Education (EQAR) — its EQAR registration was renewed in December 2023 for a five-year period (to 30 September 2028) following an external review confirming substantial compliance with the ESG 2015. That ENQA/EQAR status means ARACIS evaluations are expected to be consistent with European practice, so the evidence you present is read against the same expectations applied across the European Higher Education Area.

ARACIS carries out three broad types of activity: provisional authorisation to operate (for new study programmes, institutions and doctoral domains); accreditation; and periodic (cyclical) evaluation. As a rule, programmes, fields of study and institutions are re-evaluated on a roughly five-year cycle. The process is explicitly two-tier: an internal evaluation conducted within the institution, followed by an external evaluation carried out by ARACIS. Students are also involved directly — student representatives sit on ARACIS evaluation panels.

Where student course evaluation fits

In the Romanian model, student feedback on teaching is primarily part of the internal quality-assurance layer. Each institution operates an internal evaluation and quality-assurance commission (CEAC — Comisia pentru Evaluarea și Asigurarea Calității), which is responsible for collecting evidence — including student feedback on courses and teaching — and demonstrating that the institution has a functioning, self-improving quality system. When the ARACIS external panel arrives, it does not just want the survey numbers; it wants to see that the internal system works: that feedback is collected systematically, analysed, fed into decisions, and that the resulting actions are documented and communicated back to students. This is the closing-the-loop expectation that the ESG embeds, and it is exactly where thin, once-a-year Likert surveys tend to fail.

It is worth noting the direction of travel: the draft revision of the ESG (expected to be adopted as ESG 2027) introduces more explicit references to collecting data on student satisfaction. Institutions that already run rich, systematic student feedback will be ahead of that curve rather than scrambling to meet it.

Honesty note on scope. Exact indicator codes, standards and required documents are defined in ARACIS's current published methodology and the relevant standards for your programme type, and these are periodically updated and approved by Government decision. Always validate the precise requirements against the latest ARACIS methodology for your evaluation. This guide maps categories of requirement to evidence; it is not a substitute for the official standards.

Mapping ARACIS requirements to Koji outputs

The table below maps the kinds of evidence an ARACIS internal/external evaluation looks for to the concrete outputs an AI-native evaluation platform produces. Koji uses the same AI interview engine as the main Koji research platform (koji.so), applied to the course-evaluation context.

ARACIS-relevant requirement (category)	What evaluators look for	Concrete Koji output
Systematic collection of student feedback	Evidence that feedback is gathered regularly, not ad hoc	Scheduled AI-moderated evaluations per module/cohort with response-rate tracking
Internal quality management (CEAC evidence)	A functioning, documented internal QA process	Standardised instruments, audit trail, and exportable institution-level reports
Educational effectiveness	Insight into teaching quality and student learning experience	Thematic analysis of probing interviews, not just rating averages
Closing the loop	Proof that feedback leads to action and is reported back	Built-in action tracking linking findings to documented improvements
Periodic (5-yearly) evaluation	Trend evidence across the review period	Longitudinal cohort reporting across years and programmes
Comparability and fairness	Consistent, unbiased instruments across courses	Bias-aware, standardised moderation applied to every interview
Student voice	Genuine, qualitatively rich student input	Conversational follow-ups that capture reasons, examples and suggestions
ESG alignment / data protection	EU-compliant handling of student data	GDPR-focused data handling and EU data practices

Why AI-moderated interviews strengthen the evidence

A traditional student-evaluation-of-teaching (SET) survey produces a column of averages and a pile of unread free-text comments. For an ARACIS panel that wants to see educational effectiveness and a working improvement cycle, averages alone are weak evidence. Koji's AI moderator asks each student a standardised question and then probes — "what specifically helped you learn?", "can you give an example?" — so the resulting evidence explains why a course works or does not. Because every interview follows the same bias-aware protocol, the evidence is comparable across courses and cohorts, which matters when a panel weighs one programme against the institution's own standards over a five-year cycle.

Turning feedback into closing-the-loop documentation

The hardest part of an ARACIS file for many institutions is not collecting feedback — it is proving the loop closes. Koji's automatic thematic analysis converts hundreds of interviews into a small set of clear themes; the CEAC can then assign actions to those themes, track them to completion, and show students what changed. That action record is precisely the documentary evidence an external panel expects to see, and it is far more convincing than a claim that "the survey results were discussed".

When a different approach may fit better

To be fair: if your institution only needs to satisfy a minimal box-ticking requirement, a free or in-house survey may be cheaper, and tools like Google Forms or Microsoft Forms can capture basic ratings at no licence cost. If your QA office already has dedicated qualitative researchers with capacity to hand-code every comment, the marginal gain from automated thematic analysis is smaller. And for live, in-class engagement during teaching, a polling tool is the right instrument, not an evaluation platform. Koji's advantage is specifically in producing systematic, qualitatively rich, longitudinal evidence at scale with the loop closed — which is exactly what a rigorous ARACIS periodic evaluation rewards.

A practical preparation checklist

1. Confirm the standards. Pull the current ARACIS methodology for your evaluation type (authorisation, accreditation or periodic) and identify where student feedback is required.
2. Standardise the instrument. Move from inconsistent, programme-by-programme surveys to one bias-aware, comparable evaluation across modules.
3. Cover the cycle. Ensure feedback is collected every term so you have trend evidence across the five-year period, not a single snapshot.
4. Analyse, do not archive. Use thematic analysis so qualitative comments become findings the CEAC can act on.
5. Document the loop. Record the action taken on each theme and how it was communicated back to students.
6. Export the evidence. Prepare standardised, institution-level reports the external panel can read without translation effort.

Common pitfalls Romanian institutions hit

Three patterns repeatedly weaken an ARACIS file, and all three are about evidence quality rather than effort:

• Numbers without narrative. Presenting only Likert averages tells a panel what students rated, not why. An evaluation built on probing interviews gives the CEAC the explanatory evidence of educational effectiveness that a five-yearly review actually weighs.
• A loop that never visibly closes. Many institutions collect feedback diligently but cannot show what changed as a result, or how students were told. ARACIS expects a demonstrable improvement cycle; undocumented "we discussed it" is the single most common gap. Action tracking that links each theme to a recorded change is what turns feedback into accreditation evidence.
• Inconsistent instruments across the institution. When every faculty runs its own survey, the data is not comparable and the internal QA system looks fragmented. A single standardised, bias-aware instrument across all modules produces the institution-level comparability a panel can read at a glance.

Avoiding these three is less about buying software and more about treating student feedback as a continuous evidence stream across the whole evaluation cycle rather than a once-a-year compliance survey. The institutions that do best are those whose internal system already produces, on any given day, the evidence an external panel would ask for.

Related Resources

• Course evaluation evidence for the ESG (European Standards and Guidelines)
• Course evaluation evidence for NVAO accreditation
• Closing the feedback loop: course evaluation evidence
• Course evaluation evidence for Polish (PKA) accreditation
• Course evaluation evidence for Italian (ANVUR/AVA) accreditation

Ready to make student feedback ARACIS-ready? See how Koji turns conversational student interviews into standardised, longitudinal quality-assurance evidence — and explore the shared AI interview engine on the main Koji platform.