Maria Ilvanidou
Community Engagement Officer
Digital Curation Unit, Institute for the Management of Information Systems, Athena Research Centre
What digital skills do Arts and Humanities researchers need most today to use digital research infrastructure services effectively? And how can research infrastructures help build them? These questions drive the ATRIUM Skillset Assessment Survey, conducted within the ATRIUM project to understand the skills researchers already have, the ones they’re missing, and the training support they need to work confidently with digital tools and services in the ATRIUM Catalogue. The findings were analyzed and presented in the ATRIUM Skillset Assessment and Gap Analysis Report published this August.
ATRIUM brings together four European research infrastructures –DARIAH-EU, ARIADNE, CLARIN, and OPERAS– to strengthen digitally-enabled data-driven research, improve access to digital services and advance frontier knowledge in the Arts, Humanities, and Social Sciences (AHSS). It relies on strong community engagement, with researcher participation playing a central role for shaping services and training. Its expanding Service & Software Catalogue lists digital tools/services that support all stages of the research lifecycle, from creating and processing data to publishing and reusing it. To ensure these services can be used effectively, ATRIUM is developing a training programme that will be articulated through a pedagogical narrative in the ATRIUM Curriculum and will be hosted in DARIAH Campus. The Skillset Survey and Gap Analysis provide the foundation for this curriculum.
The survey ran from October to December 2024 and gathered 334 responses from researchers based in 31 countries, with the vast majority located in Europe: the majority of answers came from France (22.8%), Greece (16.5%) and Portugal (7.5%). Respondents are highly multilingual (60.5% use two or more working languages) and work primarily in the Humanities (71.3%). Most identify as senior researchers (33.5%), PhD students (21.9%), or early-career researchers (17.4%), giving a broad view across career stages. Their research practices are equally diverse: while textual data dominate (88%), many also work with images (48.2%), tabular data (38%), geospatial data (24.9%), audio (21.3%), video (20.4%) and, increasingly, 3D models (9.6%). Nearly 60% work with both analogue and born-digital material and 25.4% exclusively with analogue. Respondents engage with multiple research activities when working with data, of which data analysis (65.9%), data collection (62.6%) and data publication (49.1%) are the most common.
The survey paints a picture of researchers who are active digital users but who navigate a fragmented landscape of digital tools. Across 276 open-ended responses, which have been cleaned, normalized and grouped in broader categories, researchers mentioned 366 unique tools and services (an average 3.6 tools/services per respondent). These range from spreadsheets and GIS software to NLP environments, AI platforms, and specialised research tools. The diversity is striking: most tools (62.6%) were mentioned only once, pointing to a varied ecosystem in which many niche or specialised solutions are used by individual researchers. Some types of tools are widely used, especially those linked to everyday digital scholarship: working with office and productivity software, which understandably is the top category of digital tools or services. The rest of the top-five categories point to a shift toward more computational, data-intensive research. While not all text analysis, programming, or quantitative work is AI-focused, many of these tools feed into data science pipelines used in AI and Machine Learning, reflecting their growing influence in AHSS research.
When asked about their familiarity or experience with skills identified by service providers as essential for the use of some of the ATRIUM Catalogue services, several gaps stand out clearly. While many ATRIUM services can be used through simple web interfaces, others require specialised knowledge that many researchers do not (yet) have. Major skill gaps include:
Programming and development: with around half of respondents not familiar with coding, APIs, or command-line interfaces.
Automatic Text Recognition: where demand for HTR/OCR workflows is high but experience using them remains relatively low.
Metadata standards: thesauri and semantic technologies, where roughly a third of the respondents report lack of familiarity.
Container-based tools and backend technologies: including Docker and REST APIs, which remain unfamiliar to the majority of researchers.
A core principle of the ATRIUM Curriculum is learner-centred design developing resources around the needs and experiences of learners and providing clear objectives and outcomes (Tasovac and Garnett, 2024: 18). For this reason, the survey examined researchers’ training needs, preferred formats, and the challenges they face when using online learning resources.
When asked about the skills or knowledge they would like to gain (open-ended question), data analysis (39.22%) and data visualisation (35.33%) emerged as the most frequently mentioned categories.
Data Visualisation – Advanced skills in data visualisation, particularly with platforms like Tableau, Power BI, and GIS software, would enable me to create more insightful and impactful visual representations of complex data. Data Analysis and Machine Learning – I’d like to deepen my knowledge in data analysis, especially in statistical analysis and machine learning techniques using Python or R, to apply advanced analytics in my research projects.
Beyond topics, researchers described challenges that prevent them from using online learning resources effectively. The five most common barriers were: difficulty finding relevant resources, concerns about quality and credibility, time constraints, lack of certification or recognition, and low engagement or motivation, often linked to limited interaction or feedback.
The findings from the researchers’ survey and the service providers’ input feed into the next phase of ATRIUM: shaping a training curriculum that is useful, navigable, and supportive of researchers’ daily work as they advance their digital skills to make use of digital services. The recommendations outline a pathway on how this can be done in practice.
The first set of suggestions concerns the training platform itself. Researchers asked for more structured personalized learning paths, ways to follow progress, and better filtering options to help them find what they need. The report also suggests introducing a light recognition system (such as digital badges or European Digital Credentials) and carrying out regular quality assessments and updates to keep materials reliable and up to date.
Format matters too. The report recommends shorter modules with clear learning outcomes, as well as adding concise chapter abstracts in written manuals/guidelines. Video tutorials (with subtitles, transcripts, and simple exercises) are also encouraged. Respondents favoured blended learning options, combining self-paced learning materials with live, interactive sessions, and they stressed the importance of offering training at different levels, from introductory to more advanced.
The recommendations for prioritization of topics closely match the gaps identified in the survey. These include Open Science and FAIR principles (supported by practical exercises and even train-the-trainer material), as well as the skills needed for certain ATRIUM services, such as command-line basics, container tools like Docker, Git/GitHub, and programming in Python, Java, JavaScript, and R. The report also highlights core research skills (data collection, preparation, analysis and visualisation; workflows; data publication; and knowledge representation) which researchers identified as areas of interest.
Finally, the report notes that training should not focus only on end-users. Service providers also benefit from support on accessibility, multilingual documentation, and reducing technical barriers, all of which help make their services easier to use and more inclusive.
Strengthening digital skills ensures more equitable access to research infrastructures and better data and research practices across languages, media and disciplines. For ATRIUM, training is not just an optional extra but a central strategy to enable researchers to use the services available to them. The Skillset Assessment and Gap Analysis report offers a detailed, evidence-based picture of the skills researchers already have and the areas they need more support. Its insights are relevant beyond ATRIUM, contributing to ongoing discussions and efforts in digital capacity building across the wider AHSS community.
Read the full Skillset Assessment and Gap Analysis: DOI 10.5281/zenodo.16918111
Explore ATRIUM Service and Software Catalogue: https://atrium-research.eu/services
Discover training resources on DARIAH-Campus: https://campus.dariah.eu/
Tasovac, T., & Garnett, V. (2024). The Guidelines for Producing the ATRIUM Curriculum. Zenodo. https://doi.org/10.5281/zenodo.13867113
—————
Funded by the European Union under Grant Agreement n. 101132163. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union. Neither the European Union nor the granting authority can be held responsible for them.
Community Engagement Officer
Digital Curation Unit, Institute for the Management of Information Systems, Athena Research Centre