Transcript 6 – Research Data Management Workshop

Research data workshop
Research Data Leeds and Prof Bren Neale
On behalf of
Research Data Management teams
Universities of Leeds, Sheffield and York Libraries
Running order
These slides will be
shared
• Scene setting
• Academic perspective on data management
• Exercise – writing your own data management plan
•
•
•
•
What data will you generate?
Practical housekeeping
What challenges do you anticipate?
Who can help?
• Research data repositories, WREO
• Your workshop – what topics do you want to
discuss? “Ask the room”
What is / are data?
• Data is your stuff
Images from http://office.microsoft.com/en-gb/images
What is / are data?
• Not so much what material is but how it’s used
PhD
Publications
Research data
Is physics
data more
complex?
Why?
• Good research practice
• Transparency
• You may be the first
reuser of your data
• Planning saves headaches
• Good skill to have
• Increase impact
• Reach collaborators,
networks
• Compliance
Data lifecycle
creating
data
re-using
data
processing
data
giving
access to
data
analysing
data
preserving
data
Credit: UK Data Archive
Over to Bren..
• Useful links
• The Timescapes Repository:
• http://timescapes.researchdata.leeds.ac.uk/
• The Timescapes website which includes several
methodological guides:
• http://www.timescapes.leeds.ac.uk/
Exercise http://bit.ly/2htlnrO
• Basic data management plan template
Exercise 1: Your data
1. What sorts of data do you generate?
2. Any immediate issues?
3. Do you think a plan would help you?
Make notes in Sections 1 and 4 of the template
Offer any interesting feature of your conversation to the
room.
Ethics, consent, and partnerships
• Consent
• Ensure the wording on any consent form matches what you
plan to do with the data. Make sure consent is informed
consent. (UKDS)
• Industrial partnerships
• Commercially sensitive data may be subject to restriction.
Clarify ownership and release plans. ‘Available’ ≠ ‘open’.
Not all data may be subject to the same constraints
Record keeping and consistency – including decision making.
During your project
1. More planning!
2. Store data
•
•
•
•
•
Filenaming
Folder structure
Formats
Storage and handling
Backup
3. Describe data
•
Metadata and documentation
•
e.g. table values
4. Decide what to keep
What data to keep?
1. What data do I need to keep to validate the
results of my published research?
2. Does my data have value beyond my publication?
3. What’s irreplaceable, very expensive to repeat
Data appraisal
Data Types
Value
Example
Observational data
captured around the time
of the event
Usually irreplaceable
Sensor readings, telemetry,
neuro-images, survey results,
video of performance
Experimental data from lab Often reproducible but can be Gene sequence,
equipment
expensive
chromatograms, toroid
magnetic field readings
Simulation data generated
from test models
Model and metadata more
important than output data
Climate models, economic
(inputs) models.
Large modules can take a lot
of computer time to
reproduce
Derived or compiled data
Reproducible (but very
expensive)
Text and data mining, compiled
databases, 3D models
UoB
Data sharing and how not to do
it..
What
issues are
raised in
the video?
Metadata for discovery and
identification
• Title
• Creator
• Abstract
• Keywords
• Data type
• Geographic coverage
• DOI
• Metadata to enable unambiguous citation
Metadata for reuse
• Field name meanings
• Data guide / structural map
• Data format
• Research design and methodology
• Field notes
• License conditions
• Software
Exercise
Exercise 2: How will your data be organised,
documented and described?
1. Any challenges?
2. Good ideas?
3. Who would it be useful to talk to?
Make notes in Sections 2, 3 and 6 of the template
Offer any interesting feature of your conversation to the
room.
Choosing a data repository
• Does your funder have a preference?
e.g. Natural Environment Research
Council data centres
• Is there a well established subject
repository?
e.g. Oxford Text Archive / CLARIN
Consortium
• Does your publisher have a preference?
• Do you? (Figshare, Zenodo?)
• Each White Rose institution has a locally
supported data repository service
Theses and data
• Hind Abdullah Alsiary
• http://etheses.whiterose.ac.uk/15304/
• Possible to link from thesis to data
• Have the conversation sooner rather than later…
• Permissions and third party materials.
• Record keeping
A word about identifiers..
• What’s a DOI?
• Digital object identifier
• What’s an ORCiD?
• Open Researcher and Contributor ID
• Dataset citation
Exercise
Exercise 3: What are the plans for data sharing
and access in the short and long term?
1. Who needs access to your data?
2. Would you share your data? When?
Make notes in Section 5 of the template
Offer any interesting feature of your conversation to the
room.
Training and Support
• MOOC – Research Data Management and Sharing –
free, Coursera platform, videos, quizzes.
Registration required. (Uni of Edinburgh and Univ of
Carolina at North Chapel Hill)
• MANTRA – free, self paced, online (Uni of
Edinburgh)
• Coursera
• Examples of data management plans
Training and Support
• UK Data Service – practical guidance on all aspects
of data management, including handling sensitive
data
• Digital Curation Centre – online data management
planning tool (DMPOnline), How-To guides
Data management planning tool
• DMPOnline: https://dmponline.dcc.ac.uk/
• Templates for major research funders
Local Research Data Management
Services: York
Contact: [email protected]
Research Data Policy
Local Research Data Management
Services: Sheffield
Contact: [email protected]
Research Data Policy
Local Research Data Management
Services: Leeds
Contact:
[email protected]
Research Data Policy
Music performance: Hugh Davies
project
Deposit “..offered the possibility of rendering these performances
as outputs - entities as concrete, readily identifiable, and as easy to
reference as, say, a journal article would be.”
James Mooney, Lecturer in Music Technology, University of Leeds