May was so busy and fun I forgot to post these. In May I attended a research data event called The Evolution of Research Data: Strategies for Curation and Data Management at the University of Virginia. The slides are up, but here are my notes. I had to leave early, so they aren’t complete, but they might give an idea of some of the major topics.
Martha Sites, Deputy University Librarian, University of Virginia
- We should mainstream data across the entire library. It isn’t just one person’s job to be the data expert because this is impossible.
- UVA hired a data miner and scientist for the library, but got pushback from faculty who felt that they should have access to these positions.
- Data means different things to different people so need to be specific with vocabulary and listen for actual needs and not perceived needs.
Panel – Research Data Approaches and Best Practices among Virginia Institutions
- Tyler Walters – Virginia Tech
- They have created the Center for Digital Research and Scholarship, a college level center with Walters as Dean. Center includes data management, consulting, and instruction. They give researchers tools to make “rapid progress” on research. Also provide data literacy and instruction. CDRS has 15 FTE with 4 metadata librarians and 2 data science librarians. Provided a data bootcamp for graduate students that covered data management best practices. Provide curation services and support for digital humanities/ grant funding.
- Yasmeen Shorish – James Madison University
- More similar to UNCG in that have limited personnel.
- Liaisons play a huge role in research support.
- They divided up the research life cycle into the areas they could actually support.
- Research design – Liaison librarians could consult
- Documentation and description of data – Metadata librarians could consult; Created data management libguide.
- Preserve data – They weren’t sure if the library could do this long-term.
- Storing data – Library not have IR so need to find third party or use internal servers.
- Evaluation and assessment of data – In the long term what to keep? Use subject specialists to determine.
- Library can serve as expert in some areas but need to recognize areas for collaboration. Set parameters so it is TRUE collaboration.
- Sherry Lake – University of Virginia
- Three pronged approach
- Assessment (data interviews)
- Planning (help with DMPs)
- Implementation (where to put data)
- They did the data interviews with 30 faculty and created DM vitals tool
- Her center assists with NSF DMP support. Use the DMPTool.
- They used the information collected from the interviews to create their IR for data (file size limit and no cost).
- Library also needs to consider how to support day-to-day data management as the researcher is collecting data.
- Three pronged approach
Anita de Waard, Elsevier, Small Data: Bridging the Gap between Data Repositories, including a Review of Elsevier Research Data Efforts
- Small group at Elsevier is focused on collecting information about research data curation. Their goal is to increase data preservation and improve data use by linking data to research articles.
- A researcher mentioned that we need to emphasize the reuse of secondary data and not just the gathering of primary data but the big issue is finding the secondary data.
- They want to work with existing IRs to support their processes (modernize IRs; clean up back end). They don’t have a product yet, but are doing pilot research in 2013.
- Big focus on digitization of data not already in usable form (digitizing PDFs).
Panel on Research Data Strategies from the Perspective of the Research Communities
- Brian Nosek – University of Virginia
- The main barriers for making data available are perceived norms, motivation or reasoning, no accountability.
- His center is trying to deal by working with existing issues and create incentives for researchers to deposit.
- Solutions are support workflow across data lifecycle; create a low barrier to entry; leverage the existing values; enable openness and other good practices.
- Creating the Open Science Framework: http://openscienceframework.org/ – Allows researchers to store data there without it needing to be final version.