Data Management for Research : Create a Plan

Funding agencies such as the National Institutes of Health and National Science Foundation now require that data management plans (DMPs) or data management and sharing plans (DMSPs) be submitted along with grant proposals.  We expect that other agencies will follow suit.

 

What is a data management plan (DMP) or data management and sharing plan (DMSP)?

A DMP or DMSP documents what a researcher or group will do with the data collected in the course of their project, and upon completion of their research.

 

What should be included in a DMP or DMSP?

Specifics vary by discipline, but some common considerations for drafting a plan include:

• Data characteristics - How much data will be generated, and in what file formats?  Are these formats proprietary?  
• Data documentation - What file naming conventions and data identifiers will be used to organize data?  Is there a standard metadata schema or ontology used in your field for data documentation?
• Longevity and responsibility -  How long should data be retained?  What strategies will be used for storage and backup? Who in the research group, department, or university will be responsible for controlling and managing these activities?
• Sharing, usage and publication -  Who may use the data?  Have you chosen an archive for your data?
• Privacy or security considerations - Is any of the data personally identifiable, or might it pose a security risk?

 

What is the difference between a data management plan (DMP) and a data management and sharing plan (DMSP)?

Different funders may refer to these plans as either DMPs or DMSPs, but the difference is largely semantic. While funders vary a bit in their requirements, the major components of these plans are largely the same. As of summer 2023, funders are beginning to move towards common language, referring to all plans as DMSPs. 

The recorded webinar below describes data management plans in less than 20 minutes. Just click on the image below and log in to open up the webinar in a new tab. For questions, comments, and any accessibility or captioning requests, please email me. 

Note: The webinar is available to Northeastern affiliates only. If you wish to view the webinars and are not a Northeastern affiliate, please get in touch. 

What is this data management plan thing, anyways?

If you’ll be working on a grant proposal, chances are you’ll have to write a data management plan (DMP). This session gives insight into what funders are looking for in DMPs, and walks (well, sprints) through the typical components of a DMP, including some examples from successfully funded proposals.

DMPTool is an easy to use, browser-based resource that helps demystify the process of drafting data management plans. I particularly recommend it for those new to writing data management plans.

DMPTool gives step-by-step instructions and guidance for building a DMP and allows you to:

→ Create ready-to-use DMPs for specific grant-funding agencies including the NIH, National Endowment for the Humanities Office of Digital Humanities, and more than a dozen NSF directorates

→ Satisfy funder requirements for DMPs, including those that require machine readable or machine actionable DMPs

DMPTool also provides a convenient starting point for DMP writers, because resources such as funding agency guidelines, advice for principal investigators, and even sample DMPs are available on the site.

Ready to get started?  Just visit the DMPTool site and create an account with your Northeastern email address to start building your own plan.

This short video highlights features of DMPTool 2.

Data management plans often ask researchers to include information about the system that will be used to store and share data at the end of a project. The Northeastern University Library has provided the following text describing the Digital Repository Service for researchers to use in their proposals:

Suggested text for Northeastern's Digital Repository Service:

Northeastern University (NU) Library’s Digital Repository Service (NU-DRS, https://repository.library.northeastern.edu/) is a long-term digital asset management system developed and maintained by the Northeastern University Library. The NU-DRS provides the following services: 1) deposition of all file types, many of which are natively supported by the system, 2) provisioning and maintenance of a permanent identifier and URL for both the project space and individual data files via the library’s handle server, 3) discovery, access and editorial control using the Shibboleth single sign-on identity management framework, and 4) data storage and backup services, provided jointly by the library and university information systems. The Research Data Management librarian and Digital Production Services staff will work with project staff to determine appropriate metadata models and deposition schedules. Project staff will ensure all relevant files are submitted either to available community resources or to Northeastern's repository system.

Data management standards generally address how data should be collected, measured, recorded, or formatted. Using standards can save time and reduce common errors. Standards also enable interoperability, allowing better data discovery and reuse. You should include information on the standards that you will use in a DMP, as using data standards will strengthen the data's value.

Examples of standards:

• Data file formats that enable reuse, preservation and access
• Common data elements (CDEs), standard survey instruments, or other systematic data collection tools

Standards for metadata:

• Standard terminologies, e.g., from thesauri, taxonomies, or ontologies.
• Persistent Identifiers (eg, ORCIDs, Research Resource Identifiers (RRIDs), accession numbers, DOIs). Learn more about persistent identifiers in this short recording.
• Minimum information standards (e.g., MIAME, MINSEQE)
• Chemical substance identifiers

Registries of standards:

• NCBO BioPortal
• FAIRSharing

If there are no established standards for your area of research, address how will you facilitate access and reuse of the data and metadata.

Resource: Recommended Formats Statement – Datasets | Resources (Preservation, Library of Congress

Lists technical characteristics of and metadata for datasets that best support the preservation of and long-term access to these creative works. Identifies the formats the Library of Congress prefers or finds acceptable.

Data sharing requirements by federal agency

Information on both current and planned future data sharing policies for U.S. federal funding agencies.  From SPARC/Johns Hopkins University Libraries.

Funder requirements and sample plans

Includes templates, sample DMPs and links to specific National Science Foundation directorate requirements.  

Data management plan information from ICPSR

DMP elements, policies & guidance, and a sample plan for depositing data with ICPSR.