Evaluating Computing Programmes for Euro-Inf: From 'Satisfied Students' to Demonstrated Competences

The short answer

The Euro-Inf Quality Label, awarded by EQANIE (the European Quality Assurance Network for Informatics Education), is an outcomes-based accreditation for informatics and computing degrees. It asks whether graduates can demonstrate defined competences across six outcome categories - not whether students enjoyed the module. A course-evaluation programme built to feed Euro-Inf therefore needs to surface evidence of competence development, and a five-point satisfaction average simply does not carry that information. This is the same argument we made for engineering programmes and EUR-ACE and for business schools and AACSB Assurance of Learning; computing has its own framework, and it deserves its own evaluation design.

What Euro-Inf actually measures

EQANIE was founded after the Euro-Inf Project (2006-2008) to award a European quality label to informatics degree programmes that comply with the Euro-Inf Framework Standards and Accreditation Criteria, first adopted in July 2009. The backbone of the framework is a set of programme outcomes, updated by EQANIE's Accreditation Committee in 2015 and reviewed again since, which serve as the reference point for aligning informatics degrees across Europe.

The framework groups those outcomes into six categories. In broad terms they cover: the underlying conceptual basis for informatics; analysis; design and implementation; the economic, legal, social, ethical and environmental context of computing; informatics practice; and transferable (key) skills. The learning-outcome statements are specified separately at Bachelor's and Master's level - a first-cycle graduate is expected to apply established methods, while a second-cycle graduate is expected to work at the frontier of the discipline and handle incomplete or novel problems. The most recent review kept the six-category architecture while sharpening the outcomes and explicitly addressing sustainability, accessibility, privacy and compliance - reflecting how much the professional context of computing has shifted.

Read that list against a typical end-of-module questionnaire and the mismatch is obvious. "The lecturer explained concepts clearly" tells an accreditation panel almost nothing about whether a student can design and implement a non-trivial system, or reason about the legal and ethical context of what they built.

Why satisfaction data fails an outcomes-based label

Three problems compound when a satisfaction instrument is pointed at a competence framework.

First, construct mismatch. Satisfaction and competence are different constructs, and treating one as a proxy for the other is a textbook case of construct-irrelevant variance. A charismatic lecturer can lift satisfaction without lifting the design competence the label cares about.

Second, aggregation hides the signal. Even where students do comment on competence development, averaging a Likert item flattens the distribution. A module where half the cohort mastered the design outcome and half never got there produces the same mean as one where everyone reached the middle - a problem we cover in why averaging Likert scores misleads.

Third, indirect evidence, presented as direct. Accreditation frameworks distinguish direct from indirect measures of learning. Exams, capstone artefacts and portfolios are direct; student perception is indirect. Euro-Inf panels expect direct evidence for outcomes. Course evaluation is legitimately indirect evidence - but only if it captures students' reasoned account of where and how a competence developed, not a global happiness score.

"But students cannot judge their own competence - so why survey them at all?"

This is the serious objection, and it has real support. Self-assessment of ability is famously unreliable; weaker performers systematically over-rate themselves, a pattern related to the Dunning-Kruger problem in self-report. If students cannot accurately rate their own design competence, why collect their views for an outcomes label at all?

The answer is to use student evidence for what it is genuinely valid for. Students are poor judges of their absolute competence, but they are excellent witnesses to the conditions of learning: whether a module gave them the chance to design and build something substantial, whether feedback let them correct real errors, whether the legal-and-ethical context was taught as an afterthought or built into assessment. That is exactly the indirect evidence a Euro-Inf panel can use to triangulate against direct measures - provided you collect reasoned narrative, not a number. Course evaluation should complement direct assessment, never substitute for it; the honest position is triangulation, which is why the triangulation of multiple evidence sources is the load-bearing idea here.

Where Koji fits

Koji for Education is designed to produce exactly the reasoned, outcome-anchored evidence an outcomes-based label needs. Instead of a flat Likert grid, its AI-moderated conversational interview can be structured around the six Euro-Inf categories: a single-choice or scale item establishes a baseline, and an open-ended probe asks the student to point to where in the module a competence - say, design and implementation - was developed and what let them develop it. The follow-up is what a static form cannot do: when a student says "the group project taught me the most", the moderator asks which part, and why.

That open text is then run through automatic thematic analysis, so a programme director sees which outcome categories students consistently connect to concrete learning experiences and which they cannot - a far richer input to an accreditation self-assessment than a category mean. Reporting rolls up to programme level, which is the unit Euro-Inf accredits. And because the moderation is standardised and bias-aware, the evidence is consistent across cohorts and modules rather than varying with whoever wrote the questionnaire that year. Institutions that also run general user research will recognise the same AI interview engine on the main Koji platform.

None of this replaces the capstone exam or the portfolio. It makes your indirect evidence worth reading - so that when a Euro-Inf panel asks how students experienced the development of each competence, you have more than a bar chart. For computing programmes, the move from "satisfied students" to "demonstrated competences" starts with asking a better question.

A practical mapping, category by category

Turning the six categories into evaluation items is less daunting than it looks. For the underlying conceptual basis, ask students to describe a moment a theoretical idea changed how they approached a problem - a probe that reveals whether the concept was understood or merely memorised. For analysis and for design and implementation, ask them to point to the single most demanding artefact they produced and what they would do differently; the specificity of the answer is itself a signal of depth. For the economic, legal, social, ethical and environmental context, ask whether ethics and sustainability were assessed or merely mentioned - the framework's 2017 refresh made this explicit for a reason. For informatics practice, ask which professional habit (version control, testing, documentation) the module actually instilled. And for transferable skills, ask which one they expect to use first in employment. None of these is a satisfaction question; each yields indirect evidence an accreditation panel can weigh against direct measures.

Frequently asked questions

What is the Euro-Inf Quality Label? It is a European quality label for informatics and computing degree programmes, awarded by EQANIE to programmes that comply with the Euro-Inf Framework Standards and Accreditation Criteria. It is outcomes-based rather than input-based.

How many outcome categories does the Euro-Inf framework use? Six. They broadly cover the underlying conceptual basis for informatics, analysis, design and implementation, the economic-legal-social-ethical-environmental context, informatics practice, and transferable skills, specified separately at Bachelor's and Master's level.

Can student course evaluations count as accreditation evidence for Euro-Inf? Yes, but as indirect evidence. Panels expect direct measures (exams, projects, portfolios) for outcomes; course evaluation supplements these by capturing how students experienced competence development, provided it gathers reasoned narrative rather than a satisfaction score.

Why is a satisfaction average inadequate for an outcomes-based label? Because satisfaction and competence are different constructs, and a mean hides the distribution. Two very different competence profiles can produce the same average score, so the number carries almost no information about outcome attainment.

How can course evaluation align to the six Euro-Inf categories? By structuring questions around the categories and probing for concrete evidence - asking students where and how a specific competence developed - then analysing the open text thematically and reporting at programme level.

Does student self-assessment unreliability undermine this? It limits what students can validly judge. They cannot accurately rate their absolute competence, but they are reliable witnesses to the learning conditions a module provided, which is the indirect evidence accreditation can use alongside direct measures.