October.1.2018

Data sharing and the challenges facing educational researchers

By Julie McLeod and Kate O’Connor and Nicole Davis

The way educational researchers share their research is changing. Across the world, interest is growing in encouraging researchers to make their research data openly available for use by other scholars or interested parties.

This is linked to seeing the potential significance research data may have beyond their original use. Governments, keen to maximise return on investment, want to support researchers to make their data available for others to use and move away from storing it in inaccessible repositories.

One prevailing view is that if research has been publicly funded, it should be freely accessible: anyone who could use the data should have access to it because it is then more likely to generate further benefit or knowledge.

So, in the educational research world, funding organisations are beginning to introduce new requirements for data sharing. Institutions are looking at new ways of storing and managing the data produced by their researchers. At the same time, there is much interest in experimenting methodologically with data sharing and in attending to the innovations and challenges associated with digitised data and digital worlds.

These experiments in data sharing take different forms. Data sharing might mean completely open access in one context – accessible to all with no restrictions – or it might mean access to data is mediated by the lead researcher or repository staff. Data sharing repositories might also be used by researchers to store data in a way that would leave open the possibility for access to the data to be made available after a certain period has passed.

As we see it, there are pressing reasons for the educational research community to engage more fully with these developments and to critically consider both the possibilities and problems they raise, particularly but not only in relation to qualitative research.

In qualitative research, the sample size is usually small and the research is typically aimed at understanding motivations or gathering insights into subjective experience, ideas and opinions. In this type of research, context (who is involved and where) is vital.

Educational and social sciences researchers more broadly may have concerns arising from opening up access to and sharing their qualitative research data. These include the potential for re-use or application of the data out of context, or the possible identification of research participants who might not have given consent for their data to be assessible by other researchers.

The pathway to data sharing in Australia

Internationally, new policies point to a changing context in how research data is managed and maintained. Since the OECD first published its OECD Principles and Guidelines for Access to Research Data from Public Funding in 2007, OECD nations have moved to promote increased access to data generated through publicly funded research. Although early efforts to promote data sharing following this report were stymied, there are strong signs that data sharing is back on the global research agenda in recent years.

In 2016, the Higher Education Funding Council for England, Research Councils UK, Universities UK, and the Wellcome Trust jointly authored a Concordat on Open Research Data, which proposed a series of principles for working with research data. The first of these principles defined open access to research data as ‘an enabler of high quality research’ and included the direction that ‘researchers will, wherever possible, make their research data open and usable within a short and well-defined period’.

That same year, the European Commission Directorate-General for Research and Innovation published Guidelines on FAIR Data Management in Horizon 2020 as part of the H2020 Program. These guidelines emphasise that data should be FAIR, meaning Findable, Accessible, Interoperable and Reusable. They require all projects participating in an extended Open Research Data pilot to develop a Data Management Plan and seek to make open access the default setting for research data generated as part of the Horizon 2020 Program.

Similar developments are also evident within Australia. There are signs the Australian Research Council (ARC) may move towards requiring open access of data generated from ARC funded research. The ARC’s Research Data Management Strategy currently states:

The ARC is committed to maximising the benefits from ARC-funded research, including by ensuring greater access to research data. Since 2007, the ARC has encouraged researchers to deposit data arising from research projects in publicly accessible repositories. The ARC’s position reflects an increased focus in Australian and international research policy and practice on open access to data generated through publicly funded research.

This policy is also reflected in the ARC’s current funding rules for its Discovery and Linkage Programs which both include the statement:

The ARC strongly encourages the depositing of data arising from a Project in an appropriate publicly accessible subject and/or institutional repository. Participants must outline briefly in their Proposal how they plan to manage research data arising from a Project.

The recently updated Australian Code for the Responsible Conduct of Research also includes a new section titled ‘Sharing of data or information’, which states:

it is common for researchers to ‘bank’ their data or information for possible use in future research projects or to otherwise share it with other researchers. It is also increasingly common for funding agencies to require the sharing of research data either via open access arrangements or via forms of mediated access controlled by licenses. To this end, data or information may be deposited in an open or mediated access repository or data warehouse, similar to an archive or library, and aggregated over time. Archived data or information can then be made available for later analysis, unless access is constrained by restrictions imposed by the depositor/s, the original data custodian/s or the ethics review body.

The new code includes detailed stipulations around the management and data and data sharing and the ethical practices required to do such work well.

Additionally, while the ARC advise that the current preference is to encourage compliance of institutions and academics, expectation and scrutiny of this in final reports is becoming more stringent and there are signs that open access to research data could become more strictly enforced in the near future. If such changes to requirements do occur, it is unclear whether researchers, university repositories (infrastructure, resources), and research cultures and practices will be equipped to meet them.

Possibilities and challenges of data sharing in qualitative research

Data sharing opens up possibilities for greater transparency in the practices, methods, and outcomes of educational research and has the potential to enhance rigour and impact. However, data sharing and re-use also gives rise to specific ethical and epistemological challenges.

Well-documented challenges for sharing qualitative research data include:

  • managing the re-use of qualitative research materials without compromising the specificity of the context in which they were produced;
  • the creation of appropriate materials to guide re-analysis of archived qualitative datasets;
  • transparency and care in obtaining participant consent for archiving and re-use of research materials at the of data collection as well as subsequently;
  • the ethical and practical protocols governing the management of access to archival repositories;
  • identifying appropriate ways to mitigate perceived and/or actual risk;
  • responding productively and practically to the ethical and methodological dilemmas posed by making more widely available research materials that have been generated by research teams or individuals; and
  • distinguishing between data sharing as simply policy compliance and as an opportunity for creative methodological innovation.

It is timely for educational researchers to actively engage with and address these issues. First, to ensure future policy dictates are not insensitive to the needs and nature of both the field and the subjects of its research. Second, to foster ethical, methodologically sound and creative approaches to archiving and data sharing. Such practices could have the potential to develop new avenues for research and allow it to reach wider audiences, thereby enhancing the impact and reach of such work.

Policies within Australia aimed at encouraging data sharing typically do not adequately attend to the distinctive issues data sharing raises for qualitative research practices. At the same time, more robust engagement among Australian qualitative researchers with these matters will likely open up new possibilities for educational research community to more effectively influence the direction of policy and knowledge-making practices in this area.

Our workshop on data sharing in qualitative research

With this in mind, we recently facilitated a two-day workshop on Open Access, Data Sharing and Archiving of Qualitative Research that aimed to promote critical dialogue on these agendas, specifically addressing the affordances and challenges they present for the sociology of education. This workshop, supported by an AARE competitive grant, sought to canvas ideas and dilemmas across a number of intersecting but also distinct strands of work – spanning policy, research and cultural sectors – that are part of the changing context in which we conduct and communicate our research.

Possibilities for a gateway website

Following this workshop, a smaller group met to consider how we might further develop collaborations, demonstrator projects and a community of research practice in this area. Here, we discussed processes, exemplars and practices for digital archiving of qualitative projects along with possibilities for building a research community to support a web-based gateway or portal that could showcase programs and projects in the sociology of youth and education – tentatively titled Studies of Childhood, Education and Youth (SOCEY).

This gateway website could link to individual project websites, including those that might be designed for archiving of their research materials as well as others which might have a web presence with publications available. The aim would be to enhance the profile, reach and communication of individual projects and to show the scale and scope of the field more broadly. Such a website might facilitate the preparation of research or educational materials for different audiences and communities – schools, participant or advocacy groups.

Over the coming months, work will continue on the site and a core working group, drawn from researchers Australia-wide, to assist in its development.

Keeping an eye on the creative, ethical, practical, methodological and regulatory dimensions of the cluster of activity occurring in data archiving and sharing is part of the challenge.

We welcome any feedback, ideas or expressions of interest in this program of work. Please get in touch directly or leave a comment.

 

Julie McLeod is a Professor (Curriculum, Equity and Social Change) at the Melbourne Graduate School of Education, and Pro Vice Chancellor (Research Capability) at the University of Melbourne. Her research in the history and sociology of education encompasses curriculum, youth, gender and feminist studies. Her recent ARC Future Fellowship project was on ‘Youth identity and educational change in Australia since 1950: digital archiving, re-using qualitative data and histories of the present’ (makingfutures.net; juliemcleod.net). Publications include Uneven Space-Times of Education: Historical Sociologies of Concepts, Methods and Practices (2018); Rethinking Youth Wellbeing: Critical Perspectives (2015); The Promise of the New and Genealogies of Educational Reform (2015); Researching Social Change: Qualitative Approaches (2009); Making Modern Lives: Subjectivity, Schooling and Social Change (2006). Contact Julie at j.mcleod@unimelb.edu.au

 

Kate O’Connor is a postdoctoral Research Fellow in the Social Transformations and Education Research Hub at the Melbourne Graduate School of Education. Her research interests are in curriculum policy and practice in schools and universities, sociology of education, educational technology and digital scholarship. She is currently undertaking research within an ARC-funded program of work on Youth Identity and Educational Change, led by Professor Julie McLeod. Kate can be contacted at koconnor@unimelb.edu.au

 

Nicole J Davis is a Research Assistant at the Melbourne Graduate School of Education and a Humanities & Social Science Informatics Specialist in the Social & Cultural Informatics Platform at the University of Melbourne, as well as a PhD student in the School of Historical & Philosophical Studies. Her PhD research centres on urban history, while her wider research interests and practice span history, the history of education, digital humanities, and the GLAM (Galleries, Libraries, Archives, and Museums) sector. She currently project manages and undertakes research relating to the Studies of Childhood, Education and Youth (SOCEY) initiative alongside Julie McLeod. Nicole can be contacted at davis.nicole@unimelb.edu.au

 

4 thoughts on “Data sharing and the challenges facing educational researchers

  1. Professor Linda Graham says:

    Great article and initiative 🙂

    The same issues apply for quant and mixed-methods projects, although less problematic for the former. The idea of some Johnny Come-Lately mining our massive mixed-methods longitudinal dataset with several thousand variables and no deep knowledge of how we generated them is enough to keep me awake at night!

    I’m also concerned about an important question that proponents of open-access data repositories haven’t engaged with (perhaps deliberately) and that is the intellectual property involved in the creation of data, plus what might be crudely termed ‘return on investment’ to the researchers.

    Australian funding agencies do not fund the full costs of research and the majority of us make up the shortfall using our own funds and time. What recompense is there for those who labour to build the dataset (not to mention go through the pain of winning the funds in the first place), when that dataset is mined by someone who’s invested nothing in developing it? How will the original researcher’s work and IP be recognised in analyses and publications arising from that dataset? And, I wonder, is this ultimately a way of reducing funding for original research through the ARC/NHMRC??

    I don’t think our institutions are asking enough questions or thinking about the long-term implications, and I thank you all for engaging with this important problem. I’d be very happy to contribute in any way I can.

  2. Julie McLeod says:

    Hi Linda- you raise some really important points here and as you say there are implications across the range of research methods. How the ‘original data’ is contextualised is crucial and there are also numerous ways in which access can be regulated; ie not simply ‘open’ or restricted but can degrees of access and they can be time limited too- to summary data, or to full data set, field notes. I think as you suggest that there is much more work to done for educational researchers in leading some of these discussions – rather than reacting after the event, including the important issue of IP and proper acknowledgement of the researchers ( and institutions) involved in generating the original data and study.
    planning to hold a round-table discussion at AARE on this too and would welcome your contributions. thanks Linda!

  3. Catherine Scott says:

    Curating ignorance and rebadging it as virtue. Asking people’s opinions is the research you do when you can’t do research. The deskilling of educational researchers under the guise of maintaining the ‘purity’ of the exercise guarantees the continued production of self indulgent rubbishbof no value to practitioners.

  4. Julie McLeod says:

    Dear Catherine,
    Thanks for your response.
    We were definitely not proposing the curation of poor or ignorant research – from any methodological tradition. I am not sure I fully understand parts of your comment. You mention research that asks people’s opinions. Are you referring to qualitative research? And if so, it’s hard to see how are all such research is ‘self-indulgent rubbish’. I agree that one criterion for good quality educational research could be its potential value for practitioners, but as long-standing debates suggest, there are many other criteria and characteristics to consider when determining what counts as valuable and significant educational research. We are trying to raise issues about the ethics and rationale for data sharing as part of these longer conversations about research purposes and benefits. Regards, Julie

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.