Data stewardship is defined as:
Stewardship for open data science = the formalisation of roles and responsibilities and their application to ensure that research objects are managed for long-term reuse, and in accordance with the FAIR data principles. In short, stewardship applies skills to make and keep data FAIR.
Over the last few years, increasing numbers of gatekeepers such as funding organizations and journals have started to demand data sharing, data management and data stewardship according to the FAIR (Findability, Accessibility, Interoperability and Reusability) principles in order to ensure transparency, reproducibility and reusability of research. By meeting those requirements, implementation of the FAIR principles provides advantages for different stakeholders such as:
- researchers receiving credit for their work and benefitting from shared data by other researchers
- funding agencies aiming for long-term data stewardship
- professional data publishers getting credit for their softwares, tools and services for data handling
- data science community for exploratory analyses.
Making data FAIR requires skills and guidance. Data stewards are regarded as driving forces to safeguard adequate data handling from the planning process in data management plans to publication and the final digital archiving. In order to fulfill those competences and capabilities, a data steward has to bring general skills such as (I) communication skills for relationship management of data stakeholders, facilitation of meetings, reviews, trainings and working sessions and communication of policies, (II) organizational skills to understand, define and document processes, to create data management plans and the ability to develop and apply policies, (III) flexibility for change management in order to improve data quality and (IV) problem solving skills, along with faculty-specific knowledge.
Although researchers are highly specialised in their respective disciplines, there is a lack of resources for appropriate support and training for data management competences. This gap has led to the proposal of a new class of professionals in research, data stewards, who are equipped with the adequate expertise to work effectively with researchers and university services to ensure proper data management. Many universities, particularly in Europe, are currently implementing formal programmes to embed such data stewards within university structures.
TU Graz aims to to hire one data steward per faculty (Architecture, Civil Engineering, Computer Science and Biomedical Engineering, Electrical and Information Engineering, Mathematics, Physics and Geodesy, Mechanical Engineering and Economic Sciences, Technical Chemistry, Chemical and Process Engineering and Biotechnology). Those data stewards will be embedded within the faculty, have knowledge in the respective disciplines and will be coordinated via the library. They will have an education in (one of the) the respective disciplines of their faculty and will receive general centralized training at university level in order to ensure standardized and consistent data management. In their position as data stewards, they will provide services for researchers at TU Graz (e.g. policy development, data management plan creation), but also training (e.g. training on FAIR data, data management tools, metadata creation, data archiving).