by Donald T. Hawkins (Conference Blogger and Editor) <email@example.com>
Charleston Seminar: Introduction to Data Curation
The 2014 Charleston Conference began with the inauguration of a new event: the Charleston Seminar, which is envisioned to occur annually from now on and become a series of seminars on topics of high current interest. This year’s seminar, “Introduction to Data Curation,” was a 1-1/2 day event conducted by Jonathan Crabtree and Christopher (“Cal”) Lee, both from the University of North Carolina (UNC) at Chapel Hill. Crabtree is Assistant Director for Archives and Information Technology at the Odum Institute for Research in Social Science1 (which hosts the country’s third-largest archive of computer-readable social science data), and Lee is Associate Professor at the School of Information and Library Science. Both are extremely well qualified to teach a seminar on data curation; see the Charleston Conference Website2 for further biographical information.
The seminar was structured as a series of talks by the presenters, interspersed with audience interaction and exercises. The first day consisted mainly of the presentations summarized here.
History of Data Curation
Beginning in the 1950s, as organizations began to amass collections of digital data, the realization grew that those collections had long-term value and needed to be preserved. In many scientific fields, the focus on space exploration in the 1960s provided a significant impetus to data collection efforts, and a broadening awareness of the issues resulted in the development of standards and reference models. These efforts culminated in the late 1990s with a reference model for an open archival information system and involved researchers working in diverse disciplines. (Of course, librarians and archivists have been involved with data curation — though perhaps not with that label — as a routine part of their jobs for many years.) Today, digital curation activities can be found in a wide variety of applications, such as physical media properties, digital forensics and data recovery, social and physical science data archives, digital libraries, and medical information.
What is Digital Data Curation?
According to the Digital Curation Centre,3 “Digital Curation” is the active management and preservation of digital resources for current and future generations of users.
The question this Charleston Seminar set out to answer was “What knowledge and competencies do professionals need in order to do digital curation work?” and the presentations reviewed a number of them.
The DigCCurr Project4 at UNC has developed a course to prepare students to work in digital data curation, and has organized several international conferences, continuing education workshops, and the DigCCurr Professional Institute. A DigCCurr matrix of competencies needed to undertake a digital curation project as well as to organize a digital curation education curriculum is presented on the DigCCurr Website. It lists six major areas:
1. Mandates, values, and principles, including core reasons why the digital curation functions and skills should be carried out,
2. Functions and skills,
3. Professional, disciplinary, institutional, or organizational content,
4. Types of resources,
5. Prerequisite knowledge needed in order to get other things done, and
6. Steps (transition points) in the life of digital objects.
What Makes Data Different From Documents?
You may have noticed that the seminar focused on data curation, not document curation. According to Wikipedia,5 “pieces of data are individual pieces of information” and “data as an abstract concept can be viewed as the lowest level of abstraction, from which information and then knowledge are derived.” Data thus is any information that can be stored in digital form, such as text, numbers, images, video, software, etc., and it therefore includes documents. Curation practices for documents may vary, but once digitized, the issues relating to them are the same as those relating to data.
The National Science Foundation (NSF) has defined four categories of data:
1. Results of laboratory experiments,
2. Records of operations,
3. Observations from sensors or surveys, and
4. Computational simulations and algorithms.
Data can occur in many formats, in contrast to documents which generally contain only text and images.
The Data Curation Process
It is important to understand that digital data curation is an active and ongoing process, and understanding the research data cycle is critical to building relationships with data producers. Some objectives of data curation work are:
• Preserve research data,
• Enable possibility for secondary use,
• Understand the research context where data was created,
• Help next generation researchers discover the data,
• Help researchers understand their appropriate uses, and
• Understand collaboration points with research teams.
Data curators need to focus on quality issues, understand the difference that file formats make, and understand discipline-specific needs. Data curation does not mean just storing the data in a database. Most research outputs include both digital objects and data sets, and often digital objects depend on multiple files having a complex relationship to each other.
Challenges in the data curation process include a wide variation in data citation formats (in contrast to those generally employed for publications, which are fairly universal), missing data, proprietary software used to create and gather the data, and design of the research project. Faced with these sometimes daunting challenges, the data curator may be tempted to simply convert the data to text for storage, but that approach is not good curation practice because much of the representation information may be lost. Treating all formats, metadata representations, and non-textual data in the same way is very dangerous and must be avoided.
Challenges in the Representation of Digital Information
It is well known that the context of information is never captured completely. Information professionals work to bridge the gaps in what is not captured by adding metadata to the information and developing environments for storing and replicating information.
Representations and interpretations of digital objects are complementary (although multiple interpretations are possible). Every digital object has physical, logical, and conceptual characteristics, and preservation makes the information represented by the objects useful. We must realize that there is no such thing as benign neglect of digital objects; they change and degrade over time, so preservation strategies are important. Extensive information on data preservation and archiving is available in the literature, and organizations such as the Internet Archive6 have played a leading role in the development of such operations.
Data Management Plans and Data Curation Profiles
Data management plans are now required in proposals by funding agencies such as NSF. They contain information on how the data generated by a research project will be collected, processed, stored, and preserved, and cover issues such as:
• Access to the data,
• Sharing and re-use policies,
• Data standards and capture,
• Storage and preservation of the data (including backup), and
(For further details, watch for my report of the recent CENDI/NFAIS workshop on data quality in an upcoming issue of ATG.7)
A data curation profile describes the origin and lifecycle of data during a research project and is designed to capture the requirements for data as developed by the researchers. Using the profile, librarians and archivists can make decisions on how to archive and store it, based on scholars’ needs and potential uses of the data. Reasons to develop data curation profiles include:
• To provide a guide for discussing data with researchers,
• To give insight into areas of attention in data management,
• To help assess information needs related to data collections,
• To give insight into differences between data in various disciplines,
• To help identify possible data services, and
• To create a starting point for curating a data set for archiving and preservation.
A data curation profile “toolkit” has been developed jointly by the Purdue University Libraries and the Graduate School of Library and Information Science at the University of Illinois Urbana-Champaign and is available for downloading.8
Metadata, long created and used by information professionals, is commonly defined as “data about data,” but it also can refer to data that facilitates the management and use of other data. It has been described as “the curator’s best friend” and is essential in managing digital resources because it preserves their context, facilitates rights management, controls versions, and supports preservation. The Framework of Guidance for Building Good Digital Collections,9 published by the National Information Standards Organization (NISO), lists six principles applying to good metadata:
1. Ensure that it is appropriate to the materials in the collection,
2. Supports interoperability,
3. Uses standard controlled vocabulary terms,
4. States the conditions for use of the digital object,
5. Is authoritative and verifiable, and
6. Supports the long-term management of the collection.
The Dataverse Network
The Dataverse Network10 is an open-source archiving software application that was developed at the Institute for Quantitative Social Science at Harvard University. According to its Website, its purpose is “to publish, share, reference, extract and analyze research data. It facilitates making data available to others, and allows one to replicate work by others. Researchers, data authors, publishers, data distributors, and affiliated institutions all receive appropriate credit.”
On the second day of the seminar, attendees performed an exercise to become familiar with the Dataverse Network and then each individual developed a workflow and prepared an action plan appropriate to his/her own environment.
Based on the attendee evaluations, this initial Charleston Seminar was a success. Attendees liked the mix of theoretical and practical information, despite the amount of material presented. Over 80% of them said they would attend another Charleston Seminar in the future. One comment summed it up well: “It ran very on-time. And they fit everything in! Very impressive.”
Donald T. Hawkins is an information industry freelance writer based in Pennsylvania. In addition to blogging and writing about conferences for Against the Grain, he blogs the Computers in Libraries and Internet Librarian conferences for Information Today, Inc. (ITI) and maintains the Conference Calendar on the ITI Website (http://www.infotoday.com/calendar.asp). He recently contributed a chapter to the book Special Libraries: A Survival Guide (ABC-Clio, 2013) and is the Editor of Personal Archiving, (Information Today, 2013). He holds a Ph.D. degree from the University of California, Berkeley and has worked in the online information industry for over 40 years.
7. Donald T. Hawkins, “Data Infrastructure: The Importance of Quality and Integrity — A CENDI/NFAIS Workshop,” Against The Grain, upcoming issue, 2015).
Leah was appointed Executive Director of the Charleston Conference in 2017, and has served in various roles with the Charleston Information Group, LLC, since 2004. Prior to working for the conference, she was Assistant Director of Graduate Admissions for the College of Charleston for four years. She lives in a small town near Columbia, SC, with her husband and two kids where they raise a menagerie of farm animals.