From Satisfaction to Digital Competence: Mapping Course Evaluation onto EU DigComp 2.2

The short answer: Most course evaluations ask whether students enjoyed a module. The EU's digital-skills agenda asks whether graduates can do specific things — communicate, handle data, create content, stay safe, solve problems digitally. The European Digital Competence Framework (DigComp 2.2) gives you a shared, policy-relevant vocabulary to connect the two. Mapping evaluation items onto DigComp turns a satisfaction survey into programme-level evidence of competence development — the kind accreditors, employers, and your own curriculum review actually need.

The gap evaluation usually ignores

In 2025, 60% of EU citizens aged 16–74 had at least basic digital skills, up from 56% in 2023 — but the EU's Digital Decade target is 80% by 2030, and on current trajectory Europe is projected to fall short (Eurostat, Skills for the digital age; JRC, How to reach the 80% target, 2025). The disparities are stark: Romania sits at 32% and Bulgaria at 38%, while the Netherlands has already passed 84%. The gaps are widest not in emailing and social media but in digital content creation, basic cybersecurity, and online learning — precisely the higher-order competences universities claim to build.

Higher education is supposed to be part of the answer. Yet the instrument most programmes use to check whether their teaching worked — the end-of-module satisfaction survey — collects almost nothing that maps onto this agenda. "I was satisfied with the module: 4.1/5" tells a curriculum committee nothing about whether graduates can evaluate the reliability of an online source or recognise an AI-generated deepfake. The evaluation and the policy goal speak different languages.

What DigComp gives you: a shared vocabulary

DigComp is the European Commission's reference framework for what it means to be digitally competent. Developed by the Joint Research Centre, its 2022 update (DigComp 2.2) organises digital competence into five competence areas (European Commission, DigComp 2.2, JRC128415, 2022; Digital Skills & Jobs Platform):

1. Information and data literacy — finding, evaluating, and managing digital information.
2. Communication and collaboration — interacting, sharing, and collaborating through digital tools, including netiquette and managing digital identity.
3. Digital content creation — developing content, integrating and re-elaborating it, and understanding copyright and licences.
4. Safety — protecting devices, personal data, health, and the environment.
5. Problem solving — solving technical problems and identifying competence gaps.

The 2.2 update added more than 250 new examples of knowledge, skills and attitudes, including ones that deal explicitly with AI-driven systems, datafication, and emerging technologies — making it unusually current for evaluating courses touched by generative AI. Crucially, DigComp also describes eight proficiency levels, so a competence can be assessed as a developmental trajectory rather than a yes/no.

The value for evaluation is not the taxonomy itself — it is that DigComp is a common language. An item mapped to "Information and data literacy → evaluating data, information and digital content" is legible to an accreditor, an employer, a sister institution, and an AI assistant in a way that "Q7: the module used technology well" never will be.

How to map an evaluation onto DigComp

Mapping does not mean bolting twenty-one competences onto every survey. It means a disciplined three-step alignment:

• Start from the module's intended learning outcomes. Which DigComp competences does this course actually claim to develop? A data-analytics module touches information/data literacy and problem solving; a group-project module touches communication and collaboration. Most modules legitimately map to two or three areas, not all five. This is the same constructive-alignment logic behind evaluating against learning outcomes.
• Ask about perceived competence development, honestly framed. Self-report has real limits (covered below), so phrase items as observable behaviours and confidence in specific tasks — "Can you now assess whether an online source is trustworthy?" — rather than "Did you improve digitally?"
• Probe for evidence, not just a rating. A scale item alone cannot tell a programme committee what developed. A follow-up that asks for a concrete example of using a new digital skill converts a number into auditable evidence and guards against the over-claiming that plagues self-assessment.

Done well, this yields programme-level competence evidence — a map of which DigComp areas your curriculum demonstrably develops and which it claims but does not deliver. That is directly useful for closing the loop with employers and sits alongside parallel mappings onto the ESCO skills taxonomy and the GreenComp sustainability framework.

But doesn't self-reported competence just measure confidence?

Yes — and this is the objection that must be met head-on, or the whole exercise collapses into vanity metrics.

• Self-assessment is systematically miscalibrated. The least competent tend to overestimate themselves most (the Dunning–Kruger pattern), and self-rated skill correlates only modestly with measured skill. We treat this directly in Can Students Tell You How Much They Learned?. A DigComp-mapped evaluation that relies on "rate your data literacy 1–5" inherits all of that error.
• Evaluation is not assessment. A course evaluation can surface perceived competence development and flag gaps; it cannot certify that a graduate is at DigComp proficiency level 6. That requires actual assessment — a task, a portfolio, a test. The honest claim is that evaluation provides triangulating evidence alongside assessment, not a substitute for it (triangulation in evaluation).
• Frameworks can become checkbox compliance. Mapping for a report, without acting on the gaps it reveals, is theatre. The point is curricular improvement, not a tidier accreditation appendix.

The way to blunt the self-report problem is not to abandon student voice but to make it specific and behavioural, and to probe it. "Strongly agree" that you improved is weak evidence. A described instance of evaluating a dubious source, prompted by a good follow-up question, is much stronger.

A worked example: one analytics module, mapped

Take a second-year data-analytics module whose intended outcomes include "evaluate the reliability of a dataset" and "communicate findings to a non-technical audience." Those map cleanly onto two DigComp areas: information and data literacy and communication and collaboration. A DigComp-aligned evaluation would not ask "rate the module 1–5." It would ask whether students can now judge whether a source dataset is trustworthy (information/data literacy), probe for a concrete instance of when they did so, and ask how confident they are presenting a data story to a lay audience (communication). Aggregated across the cohort, the result is a statement a curriculum committee can act on: students report strong growth in evaluating data but weak growth in communicating it — a gap invisible to a satisfaction average, and directly mappable onto the employability evidence accreditors and employers ask for. That is the difference between a number and a competence map.

Where Koji fits

This is precisely the shift Koji for Education is built for: from a number on a satisfaction scale to structured, behavioural evidence of competence development. Koji's six question types let you align a module's evaluation to the two or three DigComp areas it actually targets — a scale item for self-rated confidence, a single- or multiple-choice item to map onto specific competences, and an open-ended probe for the example that turns a rating into evidence.

The decisive difference is the AI-moderated conversational interview. Where a static form takes "I got better at finding reliable information" at face value, Koji probes — what did you do, how did you check it, can you give an example — exactly the move that separates genuine competence from confident self-report. Its automatic thematic analysis then aggregates those examples into a programme-level picture of which DigComp areas are developing, and its programme- and institution-level reporting renders that in language an accreditor or employer can read. Because it is GDPR/AVG-compliant and EU-appropriate, it fits the European policy context the DigComp agenda lives in. Koji does not assess digital competence — it produces the evaluation-side evidence that tells you where to look. (Teams running workforce or product research use the main Koji platform and the same interview engine to map self-reported skills the same way.)

Europe will not close a 20-point digital-skills gap with satisfaction scores. It will close it with curricula that can show, in a shared vocabulary, which competences they develop and which they only promise. Your course evaluation can be part of that evidence base — but only once it stops asking whether students were happy and starts asking what they can now do.

Turn course evaluation into programme-level competence evidence — explore Koji for Education.