Proficiency Levels and Learning Outcomes

The eight-level scale used by DigComp 2.2

DigComp proficiency describes how a person performs a digital task, not how many applications the person knows. DigComp 2.2 uses the eight-level model introduced in DigComp 2.1. The levels combine three signals: the complexity of the task, the autonomy available to the person, and the cognitive demand involved.

The eight levels form four broad bands:

The cognitive progression often moves from remembering, through understanding and applying, to evaluating and creating. These verbs are useful clues, but no single verb proves a level. Read the entire description. “Creates a folder” may still be a basic task; “creates a new organisation-wide records model” may require highly specialised judgment.

Autonomy and task complexity must be read together. Someone acting alone on a familiar, routine task is not automatically advanced. Someone who follows a checklist during an unfamiliar incident may be operating with guidance even if the topic sounds technical.

Worked public-administration scenario

Four officers process the same type of incoming dataset.

• Amir follows on-screen instructions to save it in a named folder. This resembles Level 1.
• Bea independently applies the unit's established naming convention and asks for help only when needed. This resembles Level 2.
• Carlos handles a routine import and corrects a known delimiter problem without help. This resembles Level 3.
• Dana selects among several import methods for a non-routine dataset, adapts the procedure, and documents the solution for colleagues. The evidence points above the foundation levels and may reach an advanced band depending on the complexity and guidance involved.

An EPSO item normally supplies enough behavioural evidence to distinguish the options. Choose the description supported by the scenario, not the most flattering interpretation of the job title.

Common confusion

Years of experience are not a DigComp level. A long-serving employee may still need guidance with an unfamiliar system. A new employee may independently solve a well-defined digital task.

Independence alone does not equal Advanced. Levels 3 and 4 already involve autonomy. Advanced performance adds broader task variety, evaluation, adaptation, or guidance of others.

Finally, do not import the current four-level DigComp 3.0 scale into a question about DigComp 2.2. In 3.0 the names are Basic, Intermediate, Advanced and Highly advanced. Chapter 2 retains the eight-level 2.2 scale because that is the version cited by EPSO/AD/427/26.

Official sources

• DigComp 2.1: eight proficiency levels
• DigComp 2.2
• EPSO/AD/427/26 notice

Reading what a learner is expected to do

A learning outcome states what a person knows, understands or can do after learning. In DigComp, examples help turn broad competence descriptors into observable performance. DigComp 2.2 provides examples of knowledge, skills and attitudes across the 21 competences and proficiency levels; the framework does not claim that every example is compulsory in every context.

Use three questions to classify an outcome:

• Knowledge: What does the person know or understand? Typical verbs include knows, understands, explains, recognises and distinguishes.
• Skill: What can the person perform? Typical verbs include searches, configures, compares, creates, verifies and troubleshoots.
• Attitude: What disposition guides the person's choices? Typical signals include being critical, responsible, open, cautious, curious or supportive.

Attitudes are not vague personality labels. They become relevant when they influence digital behaviour. “Values checking a source before forwarding it” is an attitude; “checks the source against the original publication” is a skill. A person may know the rule yet fail to apply it, so the categories should not be collapsed.

Action verbs also help infer proficiency. “Follows steps” suggests guidance. “Selects an appropriate method” suggests autonomous judgment. “Evaluates alternatives and advises others” suggests advanced performance. “Develops a novel approach for a highly complex problem” points towards the highly specialised band. The object and context still matter: selecting between two routine templates is not equivalent to designing a cross-administration information architecture.

Worked EPSO scenario

A municipality introduces a shared case-management system. The training outcome says:

The officer can compare available sharing settings, choose the least permissive setting that still enables the case team to work, explain the choice to colleagues, and reconsider it when team membership changes.

“Can compare” and “choose” describe skills. Understanding why least privilege matters supplies knowledge. Reconsidering access as circumstances change shows a responsible attitude and autonomous application. Because the employee chooses among alternatives and supports colleagues, the scenario is stronger than merely following a guided Level 1 procedure.

If an exam option says only “knows that permissions exist,” it understates the demonstrated performance. If it says “creates a new access-control model for the entire administration,” it overstates the evidence. Select the narrowest level and outcome that fully fit the facts.

Common confusion

A task result is not automatically a learning outcome. “The report was uploaded” tells you what happened, not whether the employee understood the process or acted independently. The scenario needs observable evidence about the learner.

Difficulty is not the same as complexity. A task can feel difficult because the user lacks practice, while remaining structurally simple. DigComp proficiency asks about the task, autonomy and cognitive demand, not subjective frustration.

Knowledge, skill and attitude are not the same as the five dimensions. They sit within Dimension 4 of DigComp 2.2. Area, competence, proficiency and use case are described by other dimensions.

Official sources

• DigComp 2.2: examples of knowledge, skills and attitudes
• DigComp 2.1: proficiency and examples of use

Identify the edition before interpreting the level

DigComp 2.2 and DigComp 3.0 share a stable five-area, 21-competence architecture, but they do not use the same proficiency scale. Correct interpretation therefore begins with the edition named by the source or question.

DigComp 2.2 uses eight levels, inherited from 2.1:

• Foundation: Levels 1 and 2
• Intermediate: Levels 3 and 4
• Advanced: Levels 5 and 6
• Highly specialised: Levels 7 and 8

DigComp 3.0 uses four levels:

• Basic
• Intermediate
• Advanced
• Highly advanced

The four broad bands provide an orientation, but they are not a licence to replace “Level 6” mechanically with “Advanced.” The JRC revised the descriptors and introduced new learning outcomes in 3.0. A formal crosswalk should use the actual statements for the relevant competence, not just divide eight by two.

Both editions judge progression through task complexity, autonomy and cognitive demand. The 3.0 descriptions make this explicit: Basic covers remembering and implementing simple tasks with guidance as needed; Intermediate covers well-defined tasks and problems autonomously; Advanced covers assessing and applying solutions to varied complex tasks, adapting across contexts and guiding others if required; Highly advanced covers evaluating and resolving highly complex or specialised problems to create or adapt solutions and lead others if required.

For EPSO/AD/427/26, the notice's footnote names the 2022 DigComp 2.2 publication. That makes the eight-level architecture the exam-specific baseline. DigComp 3.0 remains relevant when a question expressly asks about the latest framework or when reading current EU skills policy.

Worked public-administration scenario

A human-resources unit updates an old self-assessment based on DigComp 2.2. An employee is recorded at Level 5 for managing data because she handles varied tasks and guides colleagues. The new form uses DigComp 3.0 labels.

The safe response is not to change “5” to “Advanced” automatically. The assessor should compare the evidence with the current competence statement and learning outcomes, confirm the complexity and autonomy demonstrated, and record the result under the 3.0 rubric. “Advanced” may be the outcome, but it must be supported by the revised descriptor.

In an EPSO item asking only which 2.2 band contains Level 5, the answer is simply Advanced. The level-to-band relationship is exact within 2.2; the warning concerns conversion between editions.

Common confusion

The 3.0 term is Highly advanced, while 2.2 calls its top band Highly specialised. Treating those phrases as interchangeable can reveal that an option mixes editions.

DigComp 3.0 did not add or remove competence areas when it changed proficiency. It also did not make DigComp binding law. The update revised a reference framework.

Do not assume any old resource mentioning eight levels is “wrong.” It may accurately describe DigComp 2.1 or 2.2, and EPSO/AD/427/26 intentionally points to 2.2.

Official sources

• DigComp 2.1 and its eight-level model
• DigComp 2.2 publication record
• DigComp 3.0 proficiency levels

Calibrate from evidence, not status

Proficiency scenarios become manageable when you mark three signals: complexity, autonomy and cognitive demand. Then check whether the person guides others or creates a new solution. Ignore prestige signals such as seniority, job title or expensive software unless they change the actual task.

For the DigComp 2.2 baseline, recall the eight-level ladder:

• Levels 1–2: Foundation
• Levels 3–4: Intermediate
• Levels 5–6: Advanced
• Levels 7–8: Highly specialised

Within each pair, the upper level normally shows greater independence, complexity or cognitive demand. Level 1 needs guidance for simple tasks; Level 2 performs simple tasks autonomously, with guidance where needed. Level 3 handles well-defined routine tasks and straightforward problems; Level 4 deals with well-defined non-routine tasks and problems independently. Levels 5 and 6 broaden toward varied or more appropriate solutions, complex contexts and guidance of others. Levels 7 and 8 address complex problems with limited solutions or highly complex problems with interacting factors, integrating knowledge and proposing new ideas or processes.

Under time pressure, eliminate answers that contradict a stated signal. “Needs step-by-step help” rules out a fully autonomous description. “Creates an original organisation-wide solution to a highly complex problem” rules out Foundation and Intermediate. When the evidence sits between two options, prefer the one that makes no unsupported claim.