Contact Us | Members Only | Site Map

Association of Research Libraries (ARL®)

  Transforming Research Libraries Contact:
Judy Ruttenberg
Guide for Research Libraries: The NSF Data Sharing Policy

Helping Researchers Create a Plan

Share Share   Print

As required by the NSF, the DMP is a supplementary document, intended to describe how the proposed project will adhere to the agency’s policy for disseminating and sharing research results. Researchers preparing grant applications in response to NSF solicitations will need support from librarians, such as subject specialists and data curators/archivists, as well as from information technologists, to address data management planning.

Researchers may not look, at least initially, to their academic libraries for assistance. The NSF requirement thus represents a new opportunity for collaboration between librarians and researchers, one in which – to get started – it may be incumbent on librarians to reach out to research communities, as well as to the administrative unit on campus responsible for oversight of grant-funded research, in order to work together on general guidelines for constructing a plan for managing research data.

How Does the NSF Requirement Fit in with the Grant-Writing Process?

A DMP grows out of understanding how data should be collected, normalized, processed, analyzed, preserved, used, and re-used over their lifetime. The lifecycle management of data is often referred to as data curation. (Digital curation more typically denotes lifecycle management of digital objects, whether these are data or other types of content.)

With the new NSF requirement, the grant-writing process offers a chance for researchers and librarians to plan together for curation of research data before the data have been generated and, therefore, to think carefully about data and metadata standards and policies for access and use, for example. Librarians could consider an initial session with researchers to learn about their data as a “data interview,” not unlike a “reference interview.” In this regard, an article such as “A Subject Librarian’s Guide to Collaborating on e-Science Projects” (Garritano & Carlson, 2009) suggests questions that librarians can ask researchers at the outset about their data:

  • What kinds of data will be collected or created – formats, type, extent?

  • Besides the researcher(s) on the project, who else should be given access to the data?

  • Who owns the data?

  • Will there be restrictions on the data?

  • How might the data be used, reused, and repurposed?

In asking questions like the above, the construction of a data management plan can be instrumental for librarians involved in collection development and management activities – it can be viewed as a prelude to future collection and stewardship of data sets. “In the end, a librarian can help researchers build their collections of data in order to make it as useful as possible to current and future researchers who may be interested in the same or similar research.” (Garritano & Carlson, 2009)

There are tools available to enable librarians to assess researchers’ data management needs, among them:

In addition, Purdue University—in collaboration with the University of Illinois at Urbana-Champaign—has developed a Data Curation Profiles Toolkit (project description).

How Are Libraries Communicating about this with Researchers?

For the most part, libraries are communicating about the NSF’s change to the implementation of its data sharing policy via their websites. Our Resources for Data Management Planning page lists a selection of web pages developed at, or involving, research libraries, mostly in response to the NSF call.

Liaison librarians also have important roles to play in communicating with researchers to help them meet the NSF requirement. At Purdue University Libraries liaison librarians are receiving training in the use of the Data Curation Profile tool; the tool has been adapted for ascertaining particular areas of attention for sharing, distributing, and managing research results in accordance with the NSF’s Grant Proposal Guide. At the University of Wisconsin-Madison (UW-M) Libraries, liaison librarians and the campus-IT strategic-planning task force are working together to plan research data management consulting services. In addition, the liaisons at UW-M Libraries have been invited to participate in a joint library/IT mailing list focused mainly on e-research topics; the mailing list will also be applied for “crowdsourcing” responses to the NSF data sharing policy.

Data Management Planning is a Collaborative Effort

Addressing the DMP requirement from the NSF means bridging the space between librarians, researchers, and their data. A coherent plan will necessitate collaboration across units within and beyond the research library. Many ARL member libraries, such as those listed above, are intently engaged in collaborative efforts. One example is Cornell University Library, which has representation in the recently formed Research Data Management Services Group (RDMSG). The planning group for the RDMSG consisted of faculty, staff, and librarians from a range of disciplines, including the social sciences and astronomy, and range of interdisciplinary groups and centers, such as the DISCOVER Research Services Group and the Center for Academic Computing. The RDMSG reported on its investigation into creating new services for research data management planning, determining existing capacities and analyzing gaps to overcome, which could prove a useful model for other libraries going forward in their own exploration of new service development.

Authors: Patricia Hswe and Ann Holt