The digital event was organized by the European Commission in collaboration with the European Training Foundation (ETF) and took place in May 2024. It primarily focused on showcasing the evolution of ESCO, whereas examples included big data job vacancy analysis, matching and guidance tools, and the relationship of qualifications, occupations, and skills.

"In order to implement ESCO, it is very important that you understand your labour market, your own needs and possibilities", said Arjen Deij, a senior expert on qualifications systems at ETF, before the start of Petter Nygård’s presentation on Sweden’s experience with mapping skills and occupations to ESCO.

Mapping Swedish Classifications to ESCO

The Swedish standard classification of occupations (SSYK 2012) is owned and managed by Statistics Sweden, following the International Classification of Occupations (ISCO) update cycle.

SSYK 2012 is based on and linked to ISCO-08, adapted to the Swedish labour market. It is distributed and updated by the Swedish Public Employment Service (Arbetsförmedlingen).

Currently, the Swedish classification for occupations consists of:

• 400 occupations groups, equivalent to ISCO- level 4;
• some groups from ISCO are omitted, others are added, and adapted to the Swedish labour market;
• appx 3600 occupations in occupation groups, equivalent to ISCO- level 5;
• nowadays, no further hierarchy in the Swedish model versus the one in ESCO;
• labour market concepts linked to occupations: for example job titles (appr 8000), skills, forecast

Currently, the Swedish classification for skills consists of:

• 55 skill groups with ca 6400 individual skill concepts;
• a few skill collections, such as green skills, digital skills, qualifications and more collections planned;
• skills are linked to one or more occupation groups (ISCO - level 4);
• ongoing work to identify and link occupations’ essential skills (ISCO – level 5);
  
  

The Swedish Taxonomy Database

Swedish skills and occupations are managed in a relational database, where relation types describe correlations between concepts in a broader, narrower and/or related way. Additionally, the relationships are structured such that concepts (1) are more general, (2) are more specific, (3) correspond to, or (4) are similar to each other. To proceed with mapping to ESCO, their predefined relation types for mappings are used.

"The ESCO skills and occupations are imported into the Swedish taxonomy database as their own concepts. They are assigned unique IDs in the Swedish database, along with their ESCO preferred labels and Uniform Resource Identifiers (URI). However, the ESCO hierarchy is not imported. This approach simplifies the maintenance and development of the mapping process", underlines Petter Nygård.

Governance Model and Methodology

Within the taxonomy, classifications and standards related to education, industry index (according to the Swedish Standard Industrial Classification (SNI), forecast data, geography data, among others are owned and updated by external entities. Arbetsförmedlingen is responsible for updating occupations and skills.

The ad hoc update process involves, for example, user suggestions, legislation, search trends, media coverage. The process is primarily manual, relying on stakeholder consultation, but various machine-based tools also provide insights and suggestions.

ESCO Green and Digital Skills Collection

Mappings to ESCO include green and digital skills, forming a collection adapted to the Swedish labour market. These mappings are useful for labour market analysis, helping in identifying national trends and other relevant insights related to the green and digital transition. ESCO is utilized in the EURES* portal to foster cohesion between ESCO and national classifications, contributing to a common European labour market.

Mapping to ESCO: Challenges and Benefits

There are discrepancies between ESCO and national classifications such as ISCO vs. SSYK 2012, stemming from differences in semantic granularity, the number of concepts, occupational qualifications, and sector prevalence. Additionally, the semantic granularity within ESCO may vary, and improvements could be made to the complexity and translation quality of the ESCO skills and competencies hierarchy.

Among the benefits of mapping to ESCO, are the possibility to receive support, feedback, and quick response. The ESCO database is rich and free source of labour market information, featuring an up-to-date classification that even highlights gaps in national classifications.

All versions of ESCO are mapped to the latest version of the Swedish classifications, ensuring alignment. Additionally, new Swedish occupations and skills are consistently mapped to ESCO.

According to Petter Nygård it is preferable to do the mapping process in-house, as establishing quality mappings is time-consuming and impacts significantly national classifications. It is of high importance to keep revisiting the ESCO mapping manual with the latest changes in regulations and guidelines.

Results

As a result of the applied methodology, the most recent ESCO version is incorporated into the ongoing work. For example, most of the skills have a broad match due to the semantic granularity, and the same applies to occupations. Currently, only 22% of occupations have exact matches, indicating room for improvement.

"By thoroughly reviewing ESCO mappings and examining similar occupations, we can enhance the accuracy and precision of these matches, although this requires considerable effort and attention", says Petter Nygård.