Top Banner

Click here to load reader

of 12

Libraries and Research Data Management – What Works? Lessons Learned from the First LIBER Case Studies

Aug 19, 2014



This presentation by Dr Birgit Schmidt was given at the Scholarly Communication and Research Infrastructures Steering Committee Workshop. The workshop title was Libraries and Research Data Management – What Works?

  • Libraries and Research Data Management What Works? Lessons learned from the first LIBER Case Studies Steering Committee Scholarly Communication and Research Infrastructures LIBER Annual Conference, Riga, 2 July 2014 Dr Birgit Schmidt, Goettingen State and University Library, SC SCRI Secretary
  • Why LIBER case studies on research data management? 10 Recommendations on RDM, filling this with life Several institutions across Europe are building up capacities for RDM, combining infrastructures and support services There is a lot of knowledge to share with the community Engagement starts for different reasons and from various perspectives Sharing of the rationale, how to start and proceed, some challenges faced and the way forward, i.e. lessons learned Where do you feel you just started, where do you make good progress, where do feel more confident? 2
  • 10 Recommendations for libraries to get started with research data management Regrouping the recommendations Areas of engagement Support services #1 Offer research data management support, including data management plans for grant applications, intellectual property rights advice and information materials. Assist faculty with data management plans and the integration of data management into the curriculum. #6 Support the lifecycle for research data by providing services for storage, discovery and permanent access. Infrastructure & standards #2 Engage in the development of metadata and data standards and provide metadata services for research data. #5 Liaise and partner with researchers, research groups, data archives and data centers to foster an interoperable infrastructure for data access, discovery and data sharing. #7 Promote research data citation by applying persistent identifiers to research data. #8 Provide an institutional Data Catalogue or Data Repository, depending on available infrastructure. #10 Offer or mediate secure storage for dynamic and static research data in co-operation with institutional IT units and/or seek exploitation of appropriate cloud services. Policy & disciplinary practices #4 Actively participate in institutional research data policy development, including resource plans. Encourage and adopt open data policies where appropriate in the research data life cycle. #7 ... (with some disciplinary views on data citation) Skills & staffing #3 Create Data Librarian posts and develop professional staff skills for data librarianship. #1 and the integration of data management into the curriculum. Pick your level-of-engagement card now for one area Red You have not started yet but would love to, why? What would you need? Yellow First steps have been taken, what works well, what challenges do you face? Green You feel more confident, what message would you like to share? 3
  • Case Studies What was the starting point? What kind of research data is targeted? What is the organisational framework? Roles and responsibilities Policies What kind of support services are provided to researchers? What kind of infrastructure is provided? What have you learned so far? Whats next? Further information 4
  • The case studies released so far Would you like to see your approach & experiences here too? Institutional perspectives Raw data at the Spanish National Research Council and related services by the institutional repository DIGITAL.CSIC Developing research data management services at Humboldt University Berlin Piloting research data support at the University of Goettingen University of Oxford research data management infrastructure Research data services at University College London Factors for enabling sharing and reuse of research data Library and archive services at the University of Vienna Research data management at Radboud University With emphasis on training the workforce Developing research data management training and support at Helsinki University Library Essentials 4 Data Support National e-infrastructure and services Promoting data citation by DataCite Netherlands, a DOI registration service of TU Delft Library The Lithuanian national open access research data archive (MIDAS) 5
  • What did they learn? Support services Some take-away messages There is still a lack of information on what services we offer them for their raw data (CSIC). People do not trust new services: Pilot users are generally using the service to back up their data. This is slowly starting to change (UCL). In some fields such as physics and geoscience, the practice of somehow managing data seems more developed than in other areas, but in general the researcher community thinks we should preserve and facilitate their access and would therefore value enhanced support (CSIC). The library will concentrate on informing and supporting researchers Services will be mainly provided by other institutions, national and international players. The library plays a part in bridging the gap between several services providers and researchers needs (UHEL). So far, we have been providing advice on a one-to-one basis, responding to questions by single researchers (CSIC). RDM services are best developed by holding in-depth conversations with researchers. They are tailor-made and therefore time-consuming for both researchers and supporting staff. Different types of data require differtn advice and support (U Radboud). A challenge for embedded RDM services is the imbalance of supply and demand. Institutions usually do not have the capacity to initiate the foundations of the infrastructure before the project starts even though basic services are often required early on in the project (UGOE). 6
  • What did they learn? Infrastructure and standards Some take-away messages Learning from a national study and institutional practices, and adding provisions for research data in the operation of the institutional repository. Expansion of knowledge and resources on a nation-wide basis is the next step (e-Infra Austria) (UVIE). Access restrictions depend on the data exchange agreement as developed by the research group and are facilitated through the systems technical setting. The agreement clarifies the terms of data delivery, access and use as well as restrictions and conditions (e.g. metadata must be provided before the beginning of data collection, deposited as soon as possible but not later than one year after the field sample / laboratory experiment has been taken/executed) (UGOE). Closely work with clients to improve functionality. Provide a stable infrastructure and support integration of DOI services in other repositories. Learn from other DataCite registration agencies (DataCite NL). Designing and creating data archives is usually very expensive, time-consuming and demanding work. Sufficient financial and human resources should be gathered and guaranteed in advance. Much of the work of setting up an RDM infrastructure involves gaining a clear understanding of what is required by whom The difficulty to creating a quantitative business case for such an infrastructure is also a challenge (UOX). 7
  • What did they learn? Policies and disciplinary practices Some take-away messages Important actors in the process of keeping data accessible and usable: data creators, repositories and downstream users (UVIE). With many stakeholders involved it can be a challenge to ensure buy-in (UOX). From a survey: the majority of respondents does not has an implicit or explicit policy for managing research data in place. Those with a policy in place include the complexity of data, funder requirements, the structure of the research group, and the need for third-party data access/analysing/annotation (CSIC). University-wide survey to inform the development of a RDM concept and policy, to get an idea of the types and characteristics of research data, gather requirements and undertake a gap analysis. In addition, a roadmap to adopt a coordinated approach when developing demand- oriented RDM services and to serve as a guide for central support and services (HUB). Researchers will be encouraged to share. However, the decision on what restrictions should apply or when the right point in time is to release data will remain their own responsibility. The biggest challenge can be to find a balance in roles and responsibilities between an individual researcher and the institution as a whole (HUB). RDM policies are best made within research institutes as they are aware of the different practices and needs (U Radboud). 8
  • What did they learn? Skills and staffing Some take-away messages There is a large skills variance this often requires more one-to-one consultations which are resource intensive (UCL). Training should be provided and aimed at specific groups such as doctoral students, digital humanities researchers and subject librarians (HUB). Professional skills regarding RDM have to be shared more efficiently, especially with younger researchers (HUB). Embedded librarians that include management of raw data are still a minority, with most cases concentrated on humanities, social sciences and natural resources / agriculture (CSIC). For motivating researchers to actually provide data policies are only of limited use for this even if researchers willingly accept them. This depends on acceptance of data managers within the project (U