The 14th RDA Plenary, held in Helsinki, invited research, industry and policy experts from all corners of the globe to explore diverse ways by which ‘Data Makes the Difference’. As this year’s theme addressed the potential of research data to facilitate decision making, tackle grand societal challenges and engage citizens in knowledge creation for the betterment of society, I’d like to shine a spotlight on the importance of ‘Professionalising data stewardship in the RDA community’ in helping to achieve these aims.

What is ‘data stewardship’? 

It’s well recognised that proper data stewardship is a prerequisite for FAIR and Open Research. According to the Dutch Techcentre for Life Sciences (DTL), Data Stewardship can be defined as the ‘responsible planning and executing of all actions on (digital) data before, during and after a research project, with the aim of optimising usability, reusability, and reproducibility of the resulting data.’ Essentially, data stewardship encompasses all of the various tasks and responsibilities that relate to research data management (RDM) throughout the entire research lifecycle. 

Why do we need data stewardship? 

Whilst the general assumption is that researchers, as the main data producers, are primarily responsible for managing their data, there is a growing appreciation that they need professional support to achieve this. Many researchers lack the necessary time, skills and motivation to fulfil RDM tasks, and the efforts of those who fulfil such tasks often go unrewarded. Data stewardship roles have been implemented in several research institutions to mitigate such problems, and to provide researchers with suitable advice and guidance about proper RDM practice.  

Data stewardship in practice

During the plenary, Mijke Jetten (Radboud University, NL), Marta Teperek (Delft University of Technology, NL) and Iza Witkowska (Utrecht University, NL) organised a Birds of a Feather (BoF) meeting to exchange knowledge and experiences of the international efforts to professionalise data stewardship with the RDA community. 

Invited speakers, Jonathan Petters (Virginia Tech University, USA), Karsten Kryger Hansen (Aalborg University, DK) and Erik Schultes (GO FAIR, NL), presented case studies to demonstrate models of data stewardship in practice. Whilst all case studies shared the common goal of implementing data stewardship to deliver RDM support across diverse research disciplines they each identified challenges of establishing professional, institution-wide data stewardship. 

Towards a community-endorsed Data Stewardship profession

An interactive workshop conducted during the BoF meeting asked participants to form subgroups to discuss and evaluate the challenges of professionalising data stewardship. Three key challenges were associated with… 

#1. Job profile.
What are the roles and responsibilities of data stewards? 

Data stewardship roles are continually evolving to cater for the ever-changing needs of contemporary researchers. Consequently, there is a lack of consensus on the roles and responsibilities of a data steward, meaning that the position is not professionally recognised in many research institutions. 

The requirements of a data steward typically depend upon the institution and department in which they work. For instance, data stewards may be centrally located within the library or IT services, or they may be embedded within specific faculties or departments. They may be employed full-time or part-time. What’s more, their skills and competencies are largely influenced by the domain-specific needs of their local research community, available funding and resources, and size of institution. Due to the diverse nature of the job profile, many question if data stewards should be employed as academic researchers, administrators, IT staff, librarians or archivists. 

BoF participants agreed that devising a clear job profile for data stewards (with useful terminology and definitions) will help to clarify the role when addressing institutional management and will attract academic personnel with relevant disciplinary expertise and experience to apply for data steward vacancies. As a Working Group deliverable, the participants proposed to compile a collection of data stewardship job descriptions, a selection of competencies; and, create a matrix of data stewardship tasks to certify the job specification. 

#2. Career Tracks.
What are the opportunities for professional development for data stewards?

Following on from the aforementioned discussion, it’s not surprising that data stewardship is not recognised as a professional career. Upon certification of the job profile, it will be useful to introduce professional levels of data stewardship (i.e. junior, senior, etc.). By way of example, Delft University of Technology have recently published their data and code roles of the future to define specific job profiles and their relationships, i.e. data stewards, data managers, software engineers, data scientists and data champions.

At the institutional level, understanding the different data stewardship roles will help management to employ professional experts at the correct level and salary with consideration for paid remuneration according to career progression. As there isn’t a clear career trajectory for data stewards at the (inter)national level, BoF participants suggested the involvement of a professional society and support from the RDA Europe National Nodes could help to coordinate, standardise, and increase the visibility of data stewardship as a professional career, as well as achieve uniform professional development for data stewards across countries. 

#3. Training.
What educational resources and curricula are available to data stewards?

The provision of adequate training for data stewards was identified as another major challenge. The BoF workshop revealed that the quality of training differs according to institution. However, training is generally perceived to be lacking with many data stewards ‘learning on the job’. On a positive note, participants believed that once data stewardship becomes an established profession, it will be easier to provide a training curriculum that best satisfies a data steward's roles and responsibilities.

Although most data stewards possess good RDM and domain-specific knowledge, many would benefit from pedagogical training in order to acquire the soft skills to efficiently engage with researchers and meet their needs. BoF participants advanced ideas for the provision of online training programmes that offer generalised and specialised training for data stewards, thereby granting them an opportunity to expand their knowledge and further their professional development by learning new educational material. Perhaps this output could be achieved in collaboration with the Education and Training on Handling Research Data IG and by consulting existing platforms, such as  FOSTER to collect, review and share training materials. 

The efforts and achievements to professionalise data stewardship within the RDA community are in progress. To read more, see the collaborative BoF meeting notes. 

In a world where researchers conduct best RDM practice, and comply with FAIR and Open Research principles, the role of the data steward would be obsolete. However, we are not there yet and we need data professionals. It’s time to professionalise data stewardship.