Shaping data and software policy in the arts and humanities research community: a study for the AHRC

Taylor, Rebecca, Hettrick, Simon, Walker, Johanna C, Broadbent, Philippa and de Roure, Dave (2022) Shaping data and software policy in the arts and humanities research community: a study for the AHRC Software Sustainability Institute 238pp.

Record type: Monograph (Project Report)

Abstract

Background: digital tools and software are revolutionising the nature of data and research across the arts and humanities community; how data is collected and analysed, and how it is managed, shared and sustained for future generations. > The principles of open research and the necessity of providing effective infrastructure to support the changing shape of research are driving policy and practice across UKRI, higher education and government. > The rich disciplinary diversity that characterises arts and humanities research poses challenges for providing infrastructure and support for research skills that can meet wide ranging needs and priorities. > There is a lack of evidence about current research practices and levels of engagement in and use of digital tools in the arts and humanities community and the implications of this for digital skills gaps and needs. > AHRC commissioned a team at the Universities of Southampton and Oxford to undertake a study into how best to support and build the skills, knowledge and capacity of the research community to utilise digital tools and infrastructure and ensure world class research.

Research methods: our study involved an iterative three stage, mixed methods research design that entailed: 1. A scoping stage which involved a mapping of the arts and humanities community; a review of literature and UKRI data policies and 12 exploratory interviews with researchers and stakeholders. 2. A large-scale survey of researchers from across the community, including those in a wide range of higher education institutions and independent research organisations; in various roles (academic, professional and technical); at various career stages (junior, early, mid and senior); and from a broad array of arts and humanities disciplines. Responses were collected from 335 researchers (281 complete). The survey measured current research practices including data collection and analysis, data management and sustainability, use of software, institutional support and research skills. 3. In-depth semi structured interviews with 30 survey participants who had opted in to further research and were purposively selected to capture disciplinary and career stage diversity. The interviews explored research practices, experiences, cultures and skills including barriers and facilitators to training and development, forms of institutional support and future policy scenarios.

Key results and findings: data and software practices > Arts and humanities researchers were collecting and analysing an array of digital and non-digital data forms, often more than one. Over 80% were working with textual data, 70% with images, 45% with audio, 42% with physical artefacts and 40% with numerical data. > 75% of respondents had used software in their research. Of those over 50% used resources provided by their institutions and around 25% were using open-source software. Just 17% were developing their own software or drawing on expertise to develop bespoke software for their research, which is lower than other communities. > Both survey and interview data suggested a spectrum of engagement with digital tools and software ranging from committed users to steadfastly 'analogue' and theoretical approaches. Most respondents fell somewhere in between. > Routine sharing of data was low, only 27% said they regularly shared it. 22% said they had never shared their data. Digital humanities scholars were more likely than others in the community to be engaged in routine data sharing practices. > Around 38% of researchers mainly stored their data at the end of the project in a shared area provided by their institution (i.e., Google Drive or SharePoint) and 17% used a personal shared 9area. Together only around 12% were using data repositories (either institutional or external). More of a concern was that 25% of respondents said their data was stored on a laptop or computer, suggesting this was not shared and could potentially be lost. Unsurprisingly this was particularly the case for doctoral students who had not completed their research. > The emergence of new digital forms of data collection and analysis was related to more collaborative and cross-disciplinary research, that contrasted with the traditional 'lone scholar' model typical in arts and humanities. Research teams also included various professional technical roles: developers, technicians, librarians, data analysts. The interviews found that in projects with a technical element it was more common to bring in digital expertise rather than develop it within the team. Finding domain experts with digital skills was challenging. Digital skills development > Almost 70% of respondents identified learning data management and storage skills in the past 5 years. Just over half of the respondents had also gained data analysis skills and just over 40% had acquired data collection and data sharing skills. This was the case for all career stages, except junior researchers who were less likely to have done data sharing training. > Asked about where researchers acquired skills and knowledge in data and software, the two categories with the highest response rates were 'learning on the job', either by being self-taught (70%) or through peer teaching of skills by colleagues (50%). Fewer indicated formal training mechanisms such as postgraduate study (35%) or external training courses (15%). The most well used formal training was that provided by institutions, possibly because this was locally available and likely to be free of charge. Around 35% of respondents had accessed this over the past 5 years. > In response to a question about barriers that prevented researchers learning data and software skills, 45% of respondents said they were aware of training opportunities but did not have time to attend, around 27% said they were not aware of training opportunities, and just over 10% felt that their institution did not support the technology or infrastructure necessary to make use of the skill. > The interview discussions shed light on the barriers to formal training, particularly work intensification in the sector combined with institutional cultures and practices which did not directly reward skills development. Early, midcareer and senior researchers all described being overstretched and unable to prioritise training. > There was extensive discussion of more informal mechanisms for skills development, including self-teaching and peer teaching that were free of charge and could be undertaken flexibly. The former involved using online lectures and videos, and reading 'how to' guides. Peer teaching involved learning specific skills from colleagues and team members. > Communities of practice played an important role in supporting peer teaching. Participants had joined online forums and groups focused on a particular methodology or tool that provided virtual communities of practice. Attending and talking to presenters at conferences and seminars highlighted skills gaps, new techniques and potential networks. Others working in teams and on crossdisciplinary projects described how these enabled the transfer of skills and knowledge between team members. Policies and infrastructure > Our review of UKRI data policies found that councils which had modified and adapted UKRI guidance were able to set clear policy agendas, in particular around reuse, deposition in a repository/archive, training and skills. The AHRC’s data policy was the least tailored and specific, suggesting there may be variation in how researchers interpret guidance. 45% of respondents were aware of training opportunities but did not have time to attend 27% were not aware of training opportunities 10% felt their institution did not support the infrastructure necessary to make use of the skill 10> Researchers identified a range of challenges to sustaining data and software, linked to institutional infrastructure. These included: > adequacy of and access to repositories and platforms and effective guidance on their use > long term usability and compatibility of stored data > substantial complexity of legal and technical infrastructure working across multiple institutions > lack of stewardship and oversight. > Many researchers were supportive of sustaining data in spite of the infrastructure challenges.

Conclusions and policy implications: diversity of data and software practice - This is a central feature of the community and should be celebrated. Equally, it requires a tailored and sensitive approach to research policy and guidance that speaks to the groups within the community and provides a clear steer for research practice and skills development. Data policies need to be updated to address all stages of the research process and to include software. > Collaboration and interdisciplinarity - The rise of large team-based studies and interdisciplinary research is shaping and driving digital transformation in the community. There are challenges in providing research infrastructure that can support and promote this. > A spectrum of engagement with digital tools and digital skills - There is considerable digital expertise at one end and low use of digital tools at the other. It is the broad swathe of researchers between the two ends of this spectrum where intervention is likely to have the greatest impact. > The use of digital tools and associated knowledge and skills - Overall, this is low (at least compared to other sectors). Pockets of innovation and engagement exist in and beyond the digital humanities, and among researchers at all career stages. There is scope for a substantial investment in signposting, supporting and promoting training opportunities, providing a creative environment for skills development that mainstreams digital skills alongside others. Promoting digital skills alone may not attract those who do not see the relevance to their research practice. > Communities of practice - These are a rich source of knowledge exchange and informal and f lexible learning for arts and humanities researchers. Supporting and encouraging researchers and technical professionals to develop and use those communities and networks through crossdisciplinary research and knowledge exchange, online and at conferences, will be an important way to ensure the development of the community and the cross-fertilisation of innovation in research practice. > Formal and informal learning interdependencies - Formal training (internal and external) and informal forms of learning via self-teaching and peer teaching exist together, have interdependencies and often play different roles in a researcher's development journey. Both need to be encouraged and supported as part of a wider skills development strategy. > Formal training - This has a key role to play in skills development for junior researchers, as entry level training to ensure a baseline of skills across the sector and to offer more advanced skills development in areas of demand. Understanding the time and workload constraints researchers face and their need for flexible courses with low upfront costs will facilitate take-up of training opportunities. > Data management and sustainability practices - These are not yet fully embedded in research processes and there is a lack of knowledge and understanding of both the broad rationale for open research and the institutional infrastructure that supports it. In addition, this infrastructure may not be functioning effectively. There is a need to ensure clear guidance and widely available and up to date training. > Supporting research skills development - Training grant schemes provide firm foundations for developing broad skill sets. However, job insecurity, intensification of workload, and institutional cultures related to the REF, challenge the skills development of both researchers and research technical professionals. There is a need for inclusive and explicit support for new research roles on projects, new forms of research practice and less traditional skills and knowledge (including digital) in funding opportunities and application processes.

This record has no associated files available for download.