Sixth New England Research Data Management Roundtable Notes 7.27.17
Posted by NER on November 7th, 2017
Posted in: Communities of Interest, Trainings
Community of Interest, data_science, eScience, professional development, Roundtable, workshop
The sixth New England Research Data Management Roundtable was held July 27, 2017 at Worcester Polytechnic Institute in Worcester, MA. Sponsored by the National Network of Libraries of Medicine, New England Region, the NE RDM Roundtables provide opportunities for New England librarians to compare notes, ask questions, share lessons learned, explore new working models, acquire fresh ideas, and develop new partnerships.
The day focused on sharing data: first from the perspective of the researcher, followed by the view from WPI’s academic computing.
Below are the notes captured by all participants at the event during the roundtable discussion. We hope you can use the questions and insights from others to think about your own work or inspire more questions and investigations!
Topic 1: Challenges and successes working with researchers and sharing data.
- Take a moment to reflect on the first talk – perspective of a researcher. What are your key take-aways?
- How many people were involved to make a project work
- How much data was created by one lab
- Importance of creating a culture for Data Management
- Eagerness to share
- Even champions still need help with data management!
- Importance and benefit of having a champion for data management at your organization
- Importance of cross-training students
- Clear that there was a great deal of time spent thinking about workflows and continuous improvement
- Sometimes it takes a big problem for folks to realize they need help
- Scale of work is changing — new interdisciplinary layers add to complexity
- Need for repetition in RDM training
- Do you consider yourself an advocate for data sharing, a facilitator of the process, an educator, something else?
- Trying to get a foot in the door!
- Facilitator — trying to help others be able to do their work
- Mixed advocate and educator role
- Advocacy needs to be done with care
- Helping folks think about data management proactively
- Helping connect folks to the resources they need – break down barriers in communication, departments, centers, organizations
- Helping folks get answers — not scolding into compliance. Librarians as part of the solution, but maybe not the leaders
- Importance of having a champion
- Educating others so that they can become advocates — in their departments, elsewhere
- Advocating can help you hear the pushback/feedback from others, and can itself be extremely informative
- Justifying librarian involvement can be the hardest part.
- Can you talk about some of the challenges you have had communicating with researchers re: sharing data? What about successes?
- Scientists not viewing librarians as reputable sources of guidance because librarians “are not scientists”
- Discouragement from administration, who may advocate for open but also push for protecting information — not necessarily balanced
- Scaling resources — how do we scale up from what is currently provided?
- Language and how we explain topics so all can understand
- Explaining how to use a tool
- Getting information from researchers about their wants and needs
- Tools don’t fix all the problems or make them more organized
- Cultural barricades
- Librarians not very good at sharing their own data — not good at practicing what we teach.
- Librarians need to become part of the research ecosystem — not just an observer
- Researchers don’t know what they don’t know – hard to know where to begin the conversation
- Metadata — sharing is not just putting data out there, it needs to be useful. Researchers often don’t realize they need to submit metadata.
- Time consuming to figure out solutions to unique problems — and librarians can’t always provide the unique solutions to each problem
- Can be intimidating if librarians don’t have training in same discipline — but often researchers just want help
- Perception that no one wants or needs their data.
- Motivated PIs
- Grassroots awareness with students to raise PI awareness — like naming schemes; or — asking PIs “do you know what your graduate students are doing?”
- Working with compliance office
- Good PR from other presentations, faculty
- Reminder that many folks are already doing aspects of data management
- Meeting with researcher after DMP is implemented
- Researchers often just want help — so don’t need to have same background or training as researcher
- What are some of the ways that you talk about benefits to sharing data responsibly? How do you address concerns of sharing data or the additional work involved?
- Important to focus on framing this in a positive way
- Focus on “good” practices — may feel more achievable than “best” practices
- Reinforcing practices related to publishing — you write papers so others can reproduce your work, you cite papers so others can find it — similar principles apply to data!
- Incentivizing data sharing — increasing in practice, and getting credit is of great importance. Talk about citing data to make that an important point.
- Finding the failure points
- Using the term gatekeeper — don’t use overly-restrictive language
- Diving deeper into knee-jerk reactions to sharing data — many claim funding obligations but might be a deeper discussion
- “Stand on the shoulders of giants” — you want to be those shoulders
- “You never know how useful your data may be”
- Making it personal — how to be a good scholar, good colleague, reminding researchers of other work they have built off of, and how challenging that might have been. Avoid finger wagging
- How do you educate researchers and their staff on sharing data responsibly and best practices/sharing?
- Getting into workflows and trainings that already exist
- Creating workshops for different audiences
- Disguise workshops – “loving your research data”
- Reach out to graduate students
- Find a PI advocate
- Recognizing subject specific idiosyncrasies
- Get metadata and other specialists involved early — and know who your resources are
Topic 2: Storing data at our institutions, options for long-term access, and interactions with researchers.
- What storage practices and related software are you able to recommend to researchers, based on the infrastructure and technical expertise available at your organization?
- Box, LabArchives, Open Science Framework
- Open source — avoid things that aren’t backward compatible
- Encourage folks to use local IT solutions
- Sometimes moved to faculty senate decisions
- Sometimes troubles with infrastructure that is dated or unclear
- Troubles with administration
- Larger issues with data security and privacy — education around these areas
- Do you collaborate with other units/people on campus, such as IRB, research integrity, tech transfer office, IT? What are some of the challenges and opportunities partnering with other units around data sharing and long-term access?
- Research Office, Compliance Office, IRB, Tech transfer office
- Challenging to make inroads, know who to talk to
- Establishing relationships takes extensive time
- Present solutions, not problems
- Encouraging faculty and researchers to not keep their hard drive ‘under their desk’
- Communication continues to be a challenge
- Payment — faculty member viewed any service they didn’t have to pay for as not a service
- If there is an increase in cost, transparency will be important — people want to know where their money is going
- Important to have conversations with administration
- Do you have a data repository, IR that accepts data, or other long-term storage solution at your campus?
- DataVerse, FigShare, bePress, Hydra, DSpace, Fedora
- In-house is often not the best option — recommend folks to outside/other solutions
- Creating a data catalog — record in IR that points to data wherever it lives.
- IR is not a good home for all types of data
- Access repository v preservation repository
- No long-term storage option — passed back and forth between stakeholders
- Some slower uptake for IRs early on
- Challenges of terminology between groups
- How do we help researchers make appropriate decisions about third-party data repositories?
- Used to trust government repositories — now not so sure
- Give researchers a checklist for data repository quality
- DataVerse as a model of good policies
- Tables/charts/other decision-making guides — on a libguide, elsewhere
- Use publisher suggestions
- Value the discipline-specific knowledge
- Embargo and privacy issues should be considered
- All projects come to an end. How do we prepare researchers for end-of-project data disposition years after the project ends?
- Start way sooner — need to plan for five years or more after deposit
- Conversations often happen at a time of crisis
- Would be ideal to follow up with researchers after a project — but can’t always be done
- Importance of metadata to cut through the noise, really help make work reusable into the future
- Good guidance from DCC:
- Sunset planning needs to be part of a data management plan
- If I can’t read it in five years, why should I keep it in five years
- File transformation guidance: Brown Dog from NSF; Stata and R
- File naming conventions
- Partner with archivists on preservation, appraisal, and deaccessioning
- Should get our own house in order
For more information about this roundtable or upcoming events, please contact the NER office at NNLM-NER – firstname.lastname@example.org
National Network of Libraries of Medicine, New England Region
View all posts by NER