1.
Research DataManagementfor librariansMichael Day and Marieke GuyDigital Curation Centre (DCC)

2.
About this course Short presentations with exercises and discussion Five main sections― Research data and RDM (30 mins)― Data Management Planning (30 mins)― Data sharing (20 mins)― Skills (30 mins)― RDM at Cardiff (30 mins) Coffee break halfway through, after DMP

3.
IntroductionsIntroduce yourself and offer a reflection on the questions: What is your understanding of research? Do you know anything about data management? What do you want to find out today? Do you see a role for librarians in supporting RDM?

4.
Digital Curation Centre (DCC) Consortium comprising units from the Universities of Bath(UKOLN), Edinburgh (DCC Centre) and Glasgow (HATII) Launched 1st March 2004 as a national centre for solvingchallenges in digital curation that could not be tackled by anysingle institution or discipline Funded by JISC with additional HEFCE funding from 2011 fortargeted institutional development Support selection of tools: DAF, CARDIO, DMP Online, toolsand metadata schema catalogues Offer advice and support through ‘How to Guides’, ‘Briefingpapers’ and Web site

5.
AssessNeedsMake the caseDevelopsupportandservicesRDM policydevelopmentDAF & CARDIOassessments Guidance andtrainingWorkflowassessmentDCCsupportteamAdvocacy with seniormanagementInstitutionaldata cataloguesPilot RDMtoolsCustomised DataManagement Plans…and support policy implementationSupport from the DCC

6.
Research data and RDM

7.
Exercise: What are research data? In pairs, list as many types of data as you can, focusing(if appropriate) on the subject areas you support You have 5 minutes

8.
What are research data? http://www.youtube.com/watch?v=2JBQS0qKOBU Video from DCC – first 3.10 minutes

9.
What are research data?All manner of things producedin the course of research

10.
Defining research data Research data are collected, observed or created, forthe purposes of analysis to produce and validateoriginal research results Both analogue and digital materials are data Lab notebooks and software may be classed as data Digital data can be:― created in a digital form (born digital)― converted to a digital form (digitised)

11.
Types of research data Instrument measurements Experimental observations Still images, video and audio Text documents, spreadsheets, databases Quantitative data (e.g. household survey data) Survey results & interview transcripts Simulation data, models & software Slides, artefacts, specimens, samples Sketches, diaries, lab notebooks …

12.
What is data management?“the active management and appraisal of data overthe lifecycle of scholarly and scientific interest”Digital Curation Centre

13.
What is involved in RDM? Data Management Planning Creating data Documenting data Accessing / using data Storage and backup Sharing data Preserving dataCreateDocumentUseStoreSharePreserve

14.
RDM principles and adviceto share with researchersSee in particular:UK Data Archive, Managing and sharing data: best practice for researchershttp://data-archive.ac.uk/media/2894/managingsharing.pdfn.b. Data Management Planning and Data Sharing arecovered in separate sections

15.
Data creation Decide what data will be created and how - this shouldbe communicated to the whole research team Develop procedures for consistency and data quality Choose appropriate software and formats - some arebetter for long-term preservation and reuse Ensure consent forms, licences and partnershipagreements don’t limit options to share data if desired

16.
Documentation Collect together all the information users wouldneed to understand and reuse the data Create metadata at the time - it’s hard to do later Use standards where possible Name, structure and version files clearly

17.
Access and use Restrict access to those who need to read/edit data Consider the data security implications or where youstore data and from which devices you access files Choose appropriate methods to transfer / share data― filestores & encrypted media rather than email & Dropbox

18.
Storage and backup Use managed services where possible e.g. Universityfilestores rather than local or external hard drives Ask the local IT team for advice 3… 2… 1… backup!― at least 3 copies of a file― on at least 2 different media― with at least 1 offsite

19.
Data selection It’s not possible to keep everything. Select based on:― What has to be kept e.g. data underlying publications― What legally must be destroyed― What can’t be recreated e.g. environmental recordings― What is potentially useful to others― The scientific or historical value― ...How to select and appraise research data:www.dcc.ac.uk/resources/how-guides/appraise-select-research-data

20.
Data preservation Be aware of requirements to preserve data Consult and work with experts in this field Use available subject repositories, data centres andstructured databases― http://databib.org

21.
Data Management Planning

22.
Data Management PlanningDMPs are written at the start of a project to define: What data will be collected or created? How the data will be documented and described? Where the data will be stored? Who will be responsible for data security and backup? Which data will be shared and/or preserved? How the data will be shared and with whom?

23.
Why develop a DMP?DMPs are often submitted with grant applications, butare useful whenever researchers are creating data.They can help researchers to: Make informed decisions to anticipate & avoid problems Avoid duplication, data loss and security breaches Develop procedures early on for consistency Ensure data are accurate, complete, reliable and secure

24.
Which funders require a DMP?www.dcc.ac.uk/resources/policy-and-legal/ overview-funders-data-policies

25.
What do research funders want? A brief plan submitted in grant applications, and in thecase of NERC, a more detailed plan once funded 1-3 sides of A4 as attachment or a section in Je-S form Typically a prose statement covering suggested themes Outline data management and sharing plans, justifyingdecisions and any limitations

26.
Five common themes / questions Description of data to be collected / created(i.e. content, type, format, volume...) Standards / methodologies for data collection & management Ethics and Intellectual Property(highlight any restrictions on data sharing e.g. embargoes, confidentiality) Plans for data sharing and access(i.e. how, when, to whom) Strategy for long-term preservation

27.
Exercise: My DMP - a satire Read through the satirical DMP Highlight examples of bad practice Suggest alternative methods / approaches You have 15 minutesMy Data Management Plan – a satire, Dr C. Titus Brownhttp://ivory.idyll.org/blog/data-management.html

28.
A useful framework to get startedThink about whythe questions arebeing askedLook at examplesto get an idea ofwhat to includewww.icpsr.umich.edu/icpsrweb/content/datamanagement/dmp/framework.html

29.
Help from the DCChttps://dmponline.dcc.ac.ukwww.dcc.ac.uk/resources/how-guides/develop-data-plan

30.
How DMPonline worksCreate a planbased onrelevantfunder /institutionaltemplates......and thenanswer thequestionsusing theguidanceprovided

31.
Supporting researchers with DMPsVarious types of support could be provided by libraries: Guidelines and templates on what to include in plans Example answers, guidance and links to local support A library of successful DMPs to reuse Training courses and guidance websites Tailored consultancy services Online tools (e.g. customised DMPonline)

32.
Tips to share: writing DMPs Keep it simple, short and specific Seek advice - consult and collaborate Base plans on available skills and support Make sure implementation is feasible Justify any resources or restrictions neededAlso see: http://www.youtube.com/watch?v=7OJtiA53-Fk

33.
Data sharing

34.
What is data sharing?“… the practice of making data used for scholarlyresearch available to others.” [Wikipedia]Who’s involved? the data sharer the data repository the secondary data user support staff!

35.
Reasons to share dataBENEFITS Avoid duplication Scientific integrity More collaboration Better research More reuse & value Increased citation9-30% increase depending on e.g.discipline (Piwowar et al, 2007, 2013)DRIVERS Public expectations Government agenda Content mining― http://www.jisc.ac.uk/news/stories/2012/03/textmining.aspx RCUK Data Policy― www.rcuk.ac.uk/research/Pages/DataPolicy.aspx Institutional Policy

36.
The expectation of public accessThe RCUK Common Principles state that:“Publicly funded research data are a public good,produced in the public interest, which should bemade openly available with as few restrictions aspossible in a timely and responsible manner thatdoes not harm intellectual property.”

37.
Exercise: barriers to data sharingConstraints on data sharing Possible solutions / approaches Briefly list some reasons why certain data can’t beshared and consider whether any actions could betaken to reduce or overcome these restrictions You have 10 minutes

38.
Managing restrictions on sharingEthicsBalance data protection with data sharing Informed consent – cover current and future use Confidentiality – is anonymisation appropriate? Access control – who, what, when?IPR Clarify copyright before research starts Consider licensing options e.g. Creative Commons

39.
Select formats for data sharingIt’s better to use formats that are: Unencrypted Uncompressed Non-proprietary/patent-encumbered Open, documented standard Standard representation (ASCII, Unicode)Type Recommended Avoid for data sharingTabular data CSV, TSV, SPSS portable ExcelText Plain text, HTML, RTFPDF/A only if layout mattersWordMedia Container: MP4, OggCodec: Theora, Dirac, FLACQuicktimeH264Images TIFF, JPEG2000, PNG GIF, JPGStructured data XML, RDF RDBMSFurther examples: http://www.data-archive.ac.uk/create-manage/format/formats-tableResearch360

40.
How to share research data Use appropriate repositories― http://databib.org or http://www.re3data.org License the data so it is clear how it can be reused― www.dcc.ac.uk/resources/how-guides/license-research-data Make sure it’s clear how to cite the data― http://www.dcc.ac.uk/resources/how-guides/cite-datasets

41.
Skills

42.
How are libraries engaging in RDM?LibraryITResearchOfficeThe library is leading on most DCC institutional engagements.They are involved in: defining the institutional strategy developing RDM policy delivering training courses helping researchers to write DMPs advising on data sharing and citation setting up data repositories ...www.dcc.ac.uk/community/institutional-engagements

43.
Why should libraries support RDM?RDM requires the input of all support services, butlibraries are taking the lead in the UK – why?― existing data and open access leadership roles― often run publication repositories― have good relationships with researchers― proven liaison and negotiation skills― knowledge of information management, metadata etc― highly relevant skill set

44.
Exercise: skills to support RDM Based on the activities we discussed earlier, consider whomay have relevant skills or expertise to share. You have 15 minutesActivity Library and LRC IT Services(OBIS)Research BusinessDevelopment OfficeCopyrightData citationInformationliteracyData storageDigital preservationMetadata...

45.
Possible Library RDM roles Leading on local (institutional) data policy Bringing data into undergraduate research-based learning Teaching data literacy to postgraduate students Developing researcher data awareness Providing advice, e.g. on writing DMPs or advice on RDM within a project Explaining the impact of sharing data, and how to cite data Signposting who in the Uni to consult in relation to a particular question Auditing to identify data sets for archiving or RDM needs Developing and managing access to data collections Documenting what datasets an institution has Developing local data curation capacity Promoting data reuse by making known what is availableRDMRose Lite

46.
An exciting opportunity Leadership Providing tools and support Advocacy and training Developing data informatics capacity & capability“Researchers need help to managetheir data. This is a really excitingopportunity for libraries….”Liz Lyon, VALA 2012

47.
Potential challenges Librarians are already over-taxed!― Other challenges in supporting research (Auckland, 2012)― Getting up-to-speed and keeping up-to-date How deep is our understanding of research, especiallyscientific research and our level of subject knowledge? Translating library practices to research data issues Will researchers look to libraries for this support? Still need to resource and develop infrastructure RDMRose Lite

48.
RDM at Cardiff

49.
Exercise: supporting RDM at Cardiff? In small groups, discuss which activities you thinkshould fall within your role and which shouldn’t. Do you feel confident to support RDM? How would you like to see things develop? You have 15 minutes

50.
Conclusion

51.
Summary In the light of external drivers, researchers at Cardiffneed support for RDM The library has a key role in shaping services forresearchers in this area Library staff have an opportunity to apply their skillsin a new and exciting way 

52.
Feedback Has the event met your expectations?― If not, what would you have liked to see more / less of? Was the content useful? Did you like the mix of exercises?

53.
AcknowledgementIdeas and content have been taken from various courses:― Skills matrix, ADMIRe project, University of Nottinghamhttp://admire.jiscinvolve.org/wp/2012/09/18/rdmnottingham-training-event― DIY Training Kit for Librarians, University of Edinburghhttp://datalib.edina.ac.uk/mantra/libtraining.html― Managing your research data, Research360, University of Bathhttp://opus.bath.ac.uk/32296― RDMRose Lite, University of Sheffieldhttp://rdmrose.group.shef.ac.uk/?page_id=364― RoaDMaP training materials, University of Leedshttp://library.leeds.ac.uk/roadmap-project-outputs― SupportDM modules, University of East Londonhttp://www.uel.ac.uk/trad/outputs/resources