Drawing on their recent study of South Africa’s evidence ecosystem, Ruth Stewart, Harsha Dayal, Laurenz Langer and Carina van Rooyen, show how the global north has much to learn from evidence ecosystems in the global south. Outlining five lessons that can be learnt from the South African evidence ecosystem, they argue that if notions of … Continued
What should society expect from the humanities? This question has become pressing in the debate around interdisciplinary research in support of public policy that aims to tackle societal issues. To influence that policy effectively, argues Frans Brom, the humanities must transcend individualism. This would mean not only abandoning “outsider” perspectives focusing solely on criticism of … Continued
The University of Melbourne started using DMPonline in 2017 and around the same time began to develop some data management training for graduate researchers. The library had piloted the MANTRA program from EDINA and Data Library, University of Edinbur…
The ways in research shapes and influences the wider world are a key focus of the LSE Impact Blog. This post brings together some of the top posts on the subject of research impact that featured on the Impact Blog in 2019. Invisible impact and insecure academics: structural barriers to engagement and why we should do it anyway Participatory Action […]
The Parliamentary Office of Science and Technology (POST) has played an important role in UK politics, by providing horizon scanning research summaries to parliamentarians on emerging issues. Here, Sarah Foxen and Chris Tyler discuss the challenges and opportunities faced in setting up services that put leading edge research in front of busy politicians and reflect on their work to help set up […]
Party political conferences provide a unique opportunity for academics to engage with politicians and the policymaking process, as well as a variety of different stakeholders in any given policy issue. In this post, Dr Grace Lordan, Professor Tony Travers, Dr Anna Valero and Megan Marsh describe how academics and the public affairs team at LSE have used party political conferences […]
DMPonline case study – University of York by Lindsey Myers
Like many other libraries, internal and external drivers challenged the University of York Library to identify new ways to support its research community. As a result, the Research Support Team emerged from a reorganisation of the Academic Liaison Librarian Team in 2014. Along with open access, responsibility for research data management (RDM) services and support sits with the Library’s Research Support Team.
In our eyes the team is relatively small, consisting of two Research Support Librarians and a Research Support Manager (apologies to solo professionals grappling with all aspects of research support). Of course others are involved and we couldn’t provide a service without the valuable input of our Research Data and Outputs Specialists. This post, a job share, manages process and procedures for the validation of datasets records in Pure and supports our researchers to deposit their research data with the Research Data York service.
With limited resource it is important to seize opportunities when they arise. The team quickly realised the value of DMPonline and we incorporated it into our service provision at an early stage. DMPonline is a tool that we actively promote to our research staff and research students in one-to-one appointments, at the training we deliver and online (e.g. within our RDM online tutorial). We customised the tool, adding our own guidance to it, some time ago. The process of gathering and agreeing on the text for the guidance was a great starting point for communication with colleagues from other services (e.g. IT Services, Archives, Legal) about data management. We then went on to create a simplified data management plan (DMP) template for postgraduate research projects and unfunded research.
In 2018 we cautiously launched a DMP feedback service. I say cautiously as there was concern that I, as the only person reviewing plans, might be inundated. This has not been evidenced, although I’m sure with more robust promotion of the service our fears could be realised. Providing feedback on draft plans has proven to be a useful entry point for conversations with researchers about their research, research data, the services needed to support good RDM practice, and in identifying the skills that researchers have or don’t have for data management. Encouraging researchers to share their DMPs within the institution, as examples to help other researchers write their plans, has proven to be more difficult but I continue with my endeavours.
DMPs are strongly encouraged but not mandated by York’s RDM Policy. Lack of a mandate has guided me to be more cautious in contacting researchers out of the blue as they create new plans in the DMPonline tool. However, our researchers have discovered and are clicking on the ‘Request expert feedback’ button and I’ve noticed more enquiry traffic coming my way as it also promotes the Team as a point of RDM support.
The Digital Curation Centre’s work to develop machine-actionable DMPs is an area of particular interest. The creation of a DMP is seen by some as a one-off administrative burden. I’d like to change this perception and move to a position where DMPs are seen as an integral part of research practice at York. DMPs could provide valuable information to researchers, funders, administrators and to the Research Support Team, to help us successfully manage the long-term storage and access to valuable research data.
We would like to say thank you to Lindsey Myers for sharing this blog post with us. If you would like to get involved in our knowledge exchange and share a story from your institution please do get in touch with us.
The early engagement of ‘stakeholders’ in research is often presented as a simple way to ensure that research is aligned to the needs of research users and therefore impactful. However, who these stakeholders are and what their interests might be is not always obvious. In this post Robert Borst and Annette Boaz reflect on their research on stakeholder engagement as part of a larger European […]
DMPonline case study – Maastricht University Medical Center by Mirjam Kamps
DMPMaastricht is hosted by DataHub Maastricht. DataHub provides data management services to researchers in the Maastricht University Medical Center an…
DMPonline case study – DMPonline@TU Delft by Madeleine de Smaele and Marta Teperek
Madeleine de Smaele
Data management support is provided by the Research Data Services team of TU Delft Library, in close collaboration with the Faculty Data Stewards who are providing discipline-specific support and are the first point of contact for TU Delft researchers. As part of this service, support on writing data management plans has become increasingly important. Indeed, most funding agencies require research data produced as part of a funded project to be made publicly available and are requiring their grant-holders to produce a data management plan (DMP).
In addition, TU Delft Research Data Framework Policy is encouraging this for research projects in general as part of good data management.
The increasing importance of DMPs has also prompted us to critically evaluate the complete workflow and support provided. How can we make the process as efficient as possible and meet expectations and requirements of the university, research funders, and legislation (such as GDPR), and minimise the duplication of effort for researchers as well as support staff? In answering this question we decided that providing DMPonline could be of enormous help achieving that.
When rethinking the data management planning process we got very much inspired by the work of the University of Manchester that designed an integrated workflow in which information is shared between support services across the university. See this blog post for the complete story.
Following the same method, together with representatives from the Library, Data Stewards, ICT Department, Ethics Committee and the Privacy Team, a new data management plan template was created with questions covering all the necessary information. Basically, TU Delft template consists of two sections: General TU Delft data management questions, and TU Delft questions about management of personal research data. The second section is compulsory only for projects working with personal research data.
In parallel, we customized DMPonline in order to make it a recognizable TU Delft tool with its own identity and added guidance and example answers to each template with information on local support, infrastructure and services.
Once a researcher starts writing a DMP, Data Stewards are alerted about this through an API. This allows them to pro-actively approach researchers and offer them support in creating their plans. In addition, researchers are also offered the option of having their full DMPs reviewed, using DMPonline’s ‘request feedback’ button. If researchers take up this service, their plans will be reviewed by the Faculty Data Steward, using the commenting functionality within DMPonline.
When a research project involves personal data, the researcher will be directed from TU Delft DMP template to the Ethics Committee to get ethical approval and to the Privacy Team to get advice on Data Protection Impact Assessment (DPIA), as appropriate. Vice versa, when a researcher submits the ethics application, the Ethics Committee will check whether the DMP has been approved by the Data Steward and if adequate measures will be taken to protect the interests of vulnerable persons (if applicable).
Also other experts (for example, colleagues from the Privacy Team), who are assigned a ‘reviewer’ role in DMPonline, are able to comment on a plan.
On an international level, Science Europe, representing the major funder organisations in Europe, has launched last year the Practical guide to the international alignment of research data management with core requirements for data management plans. Now we see that these requirements are increasingly being taken over by Dutch research funding agencies, such as NWO and ZonMw. We are therefore planning to align our TU Delft data management plan template with these requirements. Once the template is approved by the funding agency, researchers will be allowed to submit a DMP created using our TU Delft template directly to the funder, which makes the process much easier for our researchers.
Regarding the DMPonline tool, we will continue to collaborate with the DCC in the UK to improve functionality and to further streamline the process in a more automated way. For example, one of the features most frequently requested by our researchers are conditional questions. Researchers who do not work with personal research data should not be presented with the whole list of questions regarding personal data processing. In addition, we see clear benefits of using the API for automated notifications. At the moment, the API provides access to the administrative information of every DMP, which allowed us to set up email alerts whenever new plans are created. What we would really like to see is API access to the entire content of DMPs. We envisage that full content API access would allow us to automate more processes within our institutions, such as notifications about DMPs which are likely to require a separate ethics application, or might involve high risk data processing and could need a DPIA. Full content API could also bring us a step closer to making DMPs machine actionable. Therefore, we strongly believe that close collaboration with DCC, Research Data Alliance (RDA), as well as colleagues at other institutions who use DMPonline, will not only benefit the researcher, but also support staff as information will be more efficiently shared and workflows better connected.
“DMPonline allows you to collaborate on a template, so that both the researcher and myself as the Data Steward are always working on the same version, at the same time.” – reflects Esther Plomp, Data Steward at the Faculty of Applied Sciences at TU Delft. “I also like that we can put practical guidelines in a simple interface so that the researcher’s job becomes easier, and then consequently my job is also easier. So it saves time.”
We would like to say thank you to Madeleine de Smaele & Marta Teperek for sharing this blog post with us. We have also a recording from our August 2019 DMPonline drop in session where we were joined by Medeliene and Marta, which you can listen to here.
If you would like to get involved in our knowledge exchange and share a story from your institution please do get in touch with us.