User engagement in research data curation
Download as PPT, PDF1 like938 views
This document discusses several studies on user engagement in research data curation. It finds that institutional repositories for data were developed without input from researchers, leading to systems that did not meet researchers' needs. Barriers to open data sharing included concerns over commercial use and maintaining ownership. Successful data curation requires understanding disciplinary differences and developing trusted relationships with researchers through dialogue early in projects.
![Thank you stuart.macdonald@ed.ac.uk [email_address] All images - creative commons courtesy of Flickr](https://image.slidesharecdn.com/ecdl09notefree-100805045154-phpapp02/85/User-engagement-in-research-data-curation-14-320.jpg)
User engagement in research data curation
- 1. User engagement in research data curation Stuart Macdonald EDINA National Data Centre, University of Edinburgh Luis Martinez-Uribe Oxford e-Research Centre, University of Oxford ECDL Corfu, 30 September 2009
- 2. Data deluge An updated IDC white paper reported that the digital universe in 2007 was 281 exabytes and in 2011 should be 1,800 exabytes (or 10 times that produced in 2006). *“The Diverse and Exploding Digital Universe - an updated forecast of worldwide information growth through 2011- http://www.emc.com/collateral/analyst-reports/diverse-exploding-digital-universe.pdf (Mar. 2008) BBSRC strategic plan (2010-2015) consultation document
- 3. Research data definitions US Office of Management and Budget defines research data as “the recorded factual material commonly accepted in the scientific community as necessary to validate research findings” Words, pictures, numbers, sounds Workflows, methodologies, protocols, standard operating procedures, instrumentation, models, questionaires, code books, set-up files, algorithms, transcripts
- 4. “ it is becoming increasingly clear that effective and efficient management and reuse of research data will be a key component in the UK knowledge economy in years to come, essential for the efficient conduct of research ….” *JISC (2008) “Identifying the benefits of curating and sharing research data” - http://www.jisc.ac.uk/whatwedo/programmes/digitalrepositories2007/databenefits.aspx Research methods experiencing a radical transformation New tools & infrastructures generating research data New ways to use, share and re-use Growing importance of curating research data
- 5. Departmental websites Domain-specific repositories Centralised data repositories (UKDA, NERC, MRC) Libraries and computing/IT services within academic institutions working together to develop and customise institutional repositories to curate research data Data deposition and publication
- 6. Institutional Repositories: open access built for academic publications technology lead No formal requirements analysis procedures User engagement required to develop systems that will meet researchers’ needs Bottom up approach to inform top-down thinking Researchers – key user community overlooked
- 7. DISC-UK DataShare - legal, cultural, technical issues surrounding the sharing of research data in institutional settings Barriers to sharing: time taken to prepare datasets for deposit concerns over making data available before full academic exploitation misuse / misinterpretation (journalists, non-academics) loss of ownership, loss of commercial or competitive advantage repositories will cease to exist unwillingness to change working practices uncertainty about IPR and confidentiality Open data – realism versus altruism
- 8. Charting individual researcher’s information practices across 7 sub-disciplines of the life sciences - http://www.rin.ac.uk/case-studies DCC / ISSTI (University of Edinburgh) Deployed a range of methodologies and tools including short-term ethnographic techniques and semi-structured instruments: Diaries (x55), F-2-F interviews, (x24) Cognitive mapping (1 per case), Focus groups (1 per case) RIN-funded Disciplinary case studies
- 9. Some disciplines lend themselves more than others to ‘openly’ data sharing Research data are varied, specific and complex Data curation and/or sharing only becomes crucial at certain stages of research lifecycle Feeling that only researchers have subject knowledge to curate their own data Keen sense of ‘ownership’ and protectiveness towards data Some findings from RIN Disciplinary case studies project:
- 10. Scoping digital repository services for research data management - http://www.ict.ox.ac.uk/odit/projects/digitalrepository/ Scope requirements for services to manage research data generated by Oxford researchers from a variety of disciplines: Interviews (x37) conducted to learn about data management practices and identify top requirements Workshop (x46) held to compliment findings and to gather examples of good practice regarding use of repository services for research data management Consultation with service units (ORA, data library,NGS, oxford digital library) - identify gaps in service, validate researchers’ requirements
- 11. Scoping digital repository services - top requirements Advice on practical issues related to managing data across their life cycle incl. data management plans, assistance with formatting Secure storage required for large datasets generated by high throughput instruments Sustainable & authenticated infrastructure that allows publication and long-term preservation of research data It is now followed up by the intra-institutional JISC funded Embedding Institutional Data Curation Services in Research (EIDCSR) project - http://eidcsr.oucs.ox.ac.uk/
- 12. Tools – Data Audit Framework http://www.data-audit.eu/ DAF helps to establish relationships with research communities around the issues of data curation Allows institutions to identify, locate, describe and assess how they are managing their research data Provides information specialists who wish to extend support for research data with a vehicle for engaging with researchers e.g. through local research data management training "staff had numerous comments and suggestions for improvement of data management at different levels indicating an awareness of the issues, even where it had not been made a priority to address" - edinburgh data audit implementation project
- 13. Summary Repository development distant from current research needs - due to lack of iterative requirements analysis with researchers Open data ethos detached from disciplinary research needs Trusted relationships - dialogue with researchers early in research process
- 14. Thank you stuart.macdonald@ed.ac.uk [email_address] All images - creative commons courtesy of Flickr