Elise Dunham ’13LS: From Graduation to Data Curation

October 12, 2016

Elise Dunham

We caught up with Elisa about how her love of metadata shaped her career

An MSLIS graduate with a concentration in archives, Dunham discovered she was a “metadata person” during her time at Simmons. Read on to find out how this discovery influenced her job search and has shaped her career.

Can you talk about your current position as a Data Curation Specialist at the University of Illinois at Urbana Champaign, and the path that brought you there since completing your MSLIS at Simmons in 2013?

I’m a part of the library-based Research Data Service (RDS) at the University of Illinois at Urbana-Champaign (U of I). My role as a Data Curation Specialist supports the mission of the RDS, which is to provide the U of I research community with the expertise, tools, and infrastructure necessary to manage and steward research data. Operationalized, this means that I meet with researchers to discuss their data management challenges and recommend solutions, provide feedback on Data Management Plans researchers write for grant applications, and work with a team to develop, maintain, and iterate upon the recently-launched Illinois Data Bank, the U of I repository for research data.

As is the case for many LIS professionals, the path that lead to my current position in research data services was not the one I’d imagined when I graduated from Simmons. Throughout my coursework in the archives concentration, I realized that I was a “metadata person.” On the job hunt, I applied for positions with extensive metadata-related responsibilities along with my applications for broader archivist positions. That’s how I became a Metadata Production Specialist at the Roper Center for Public Opinion Research, a data archives. The work I did there pulled me into the data curation field, and here I remain.

You recently contributed a blog post to bloggERS (the blog of SAA’s Electronic Records Sections), as part of their series on processing digital materials, in which you share insight and workflow information on digital research data processing, culled from your work developing the newly-launched Illinois Data Bank at the U of I, and draw comparisons to the work of archivists. What are some of the similarities and differences in the two approaches, and how can one inform the other?

I’ve appreciated how applicable my training as an archivist is to the work we’re doing, and how receptive my colleagues here are to the archives perspective, whether it’s brought to the discussion by me or one of our colleagues in the University Archives. With the Illinois Data Bank, we’re excited about our plans for appraising and selecting research data. Dissimilar to the typical archives workflow, we plan to do this at least five years after ingesting a dataset. As the blog post describes in depth, the reason for the delayed appraisal process is that we serve short-term as well as long-term needs of researchers at the U of I. We have been informed by the work of archivists because we understand that we cannot and should not endeavor to “keep everything” that we receive and that mission-driven collections policies are crucial to our sustainability as a repository. I feel like our work in this area could benefit archivists because we have envisioned a structured approach to making appraisal decisions that could inform the approaches of accessioning archivists and others who make decisions about what to keep and what to discard.

What sort of datasets can the Illinois Data Bank support? Do you find researchers reticent to share their data, or has that climate been altered with the shift towards public access mandates incorporated into grant stipulations? Is outreach a strategic component addressed in Illinois Data Bank planning and implementation?

As outlined in our Accession Policy, “datasets deposited into the Illinois Data Bank must either be generated through the course of a research project or deposited with an expectation that public availability will allow the Dataset to be used for research purposes.” On a more technical level, we support any files that can be transferred to us either via our browser-based self-deposit system or even delivered via the “sneakernet”—physical media walked back and forth across campus—up to 2 terabytes per researcher per year. We don’t restrict deposits to open, plain-text file formats; we leave it to the depositor to provide their data in whatever format makes the most sense for their work, their peers, and their domain. We believe usability is most important, at least in the first five years of a dataset’s life in the repository; preservability of files will be considered as part of the Preservation Review process.

There’s no easy answer to the question of whether researchers are reticent to share their data, as it varies wildly depending on a number of factors, especially the culture of a given discipline. We’ve talked with some researchers who are 100% behind the notions of open science and are eager to share, others who work with human subjects and feel as though the push toward open science isn’t sensitive to their sensitive data needs, and others who have built careers on single datasets and, in a “publish-or-perish” environment, don’t want to lose publication opportunities to someone else accessing the data they’ve invested so much in collecting. These examples only just begin to paint the picture of how diverse researcher perspectives are on data sharing. We do make outreach a priority in the RDS at the U of I, and at the same time we make listening and understanding our priorities as well. We don’t feel it is our place to convince researchers that data sharing is the only and best thing to do; our focus is on providing services and advice around data management and sharing, no matter whether a researcher is absolutely planning to share, on the fence about sharing, or sure s/he will not share.

There is a lot of silo-ing among various LIS and researcher disciplines, in terms of culture, conferences, online interaction—how can we promote a more collaborative and multidisciplinary environment in higher education, like the U of I Library’s partnership between University Archives and Research Data Service?

I was lucky to come into an organization that was already doing a conscientious job of cross-pollinating across the Library system, the Archives and the Research Data Service being prime examples. It’s difficult for me, with the limited perspective I have, to speak to this. Some of my best cross-unit interactions have been serendipitous, perhaps in line at the coffee shop located in the Library or after a talk or committee meeting. I’m sure there a lot of things that can be done at the administration level to promote a collaborative culture, but what comes to mind for those of us just starting out is to make an effort to chat with colleagues when the opportunity arises. Ask them what projects they’re working on and tell them about yours. Even if there’s not a direct opportunity to collaborate, you’ll have learned something and begun a foundation for your next conversation.

Previous to joining U of I, you were a Metadata Production Specialist at the Roper Center for Public Opinion Research, and have lead the Describing Archives: A Content Standard (DACS) workshop revision working group in the Society of American Archivists’ Technical Subcommittee on DACS, in addition to specializing in metadata in your current position. As someone that began as a history scholar, where did this love of metadata develop? Were there specific courses or experiences during your time at Simmons that contributed to this focus?

I took Information Organization with Prof. Kathy Wisser in my first semester and became fascinated with system design without realizing I was doing system design. As an archives concentrator, I learned Encoded Archival Description (EAD) and DACS in an archives metadata class, and I dove deeper into working with EAD on the Leveraging Encoded Archival Description Skills (LEADS) project with Kathy and Jason Wood. I rounded out my coursework with Database Management with Michael Leach and Metadata with Kathy. I’m thankful to Kathy, Kate Bowers, and Michael Leach for nurturing my early interest and helping me shape it into expertise.

The increasingly specialized nature of academic librarian positions can encourage many young professionals to accept employment in new cities and regions. Can you talk about the experience of moving across the country by yourself for a new job?

Moving for a job is definitely scary, and for me my move to Connecticut (after grad school) was the first time I moved somewhere and didn’t have school to help me find a social group. If I could give my past self one piece of advice, it would have been to push myself outside of my comfort zone earlier than I did to seek out friends. What I’ve learned works for me is attending activity-oriented meetings or events to meet people, like a community choir or activist group.

As for the job market—it can get tough, and rejections can be disheartening. Do what you need to do to take care of yourself. What worked for me was reminding myself that the rejections were never about me personally—especially pre-interview when folks hadn’t met me yet. I also strove to maintain a balance in my life; letting the job hunt rule every waking moment would have brought down my mood and ultimately my efforts at finding a job. I mixed things up by going out with friends and reading for pleasure. Making time for yourself outside of the job hunt is critical.

Illinois Data Bank, “Illinois Data Bank Preservation Review Guidelines”, 2016-05-03, https://databank.illinois.edu/policies#preservation_guidelines. Note that while the terms “appraising and selecting” are not present in the text on the Illinois Data Bank website, the “Preservation Review” activity being described is similar in practice.

Photo courtesy of Elise Dunham.