What is a Data Management Plan?
A Data Management Plan, or DMP (sometimes also called a data sharing plan), is a formal document that outlines what you will do with your data during and after a research project. It describes the type of data you use for your research, how are collected, organized, and stored, and what formats you use. When senstive data are used, the DMP must also describe what steps you are taking to make your data secure and compliant with regulations. It details how data will be accessible and documented for sharing and reuse during and after the project is finished. DMP are required by an increasing number of funding entities and research institutions.
A DMP might be informal, to be used internally, guided by policies established by the head of a research lab, collaborators, or IT groups across Harvard. Some funding agencies or projects might require submission of formal DMPs. It is important during the course of a project to make sure that everyone on the research team is aware of where the data reside and how to access them, how they are organized and documented so they are readily usable by the team. DMPs can be updated and revised as needed during the course of the project - a DMP is a living document.
At this point, Harvard does not require a DMP for your research, unless your funding source requires it. However, deliberately planning for the entire lifecycle of your data from the beginning of your project can benefit you, whethe a DMP is required or not, and save you time and money in the long run.
For more information:
- Research Data Management Program - DMPTool by Harvard Library
- Biomedical Data Management Planning by Harvard Medical School
Who requires a Data Management Plan?
An increasing number of federal and private sponsors require DMPs as a condition of providing funding. Usually, researchers will be asked to submit a DMP at the proposal or award stage.
Examples of research sponsors requiring a DMP or similar information:
- U.S. federal agencies (NIH, NSF, NASA, IMLS, DOD, DOJ)
- Philantropic organizations, such as Alfred P. Sloan Foundation, Moore Foundatio, The Gates Foundationhas an open data policy, but DMP is not required
For a full list of organizations requiring a DMP, as well as information about their template, check the DMPTool Public Templates page.
Your Data Management Plan should include:
- Type of data: observation, experimental, simulation, derived/compiled
- Form of data: text, numeric, audiovisual, discipline or instrument specific
- Research community standards preferred (e.g., FITS for astronomy)
- Preservation formats preferred (e.g., tables in CSV, docs in PDF, images in JPG)
- Size of data, stable data: plans for where you will store them
- Sensitive data: plans for secure storage
- Metadata: Data documentation (general, content, technical, access)
- Check whether you need Institutional Review Board approval
- Check whether you need a Data Use Agreement
- Identify a data repository for sharing and archiving the data after your study is finished (find more information at the Data Sharing & Preservation page)
For more information, see: https://dmptool.org/help
What tools can I use to create a Data Management Plan?
- DMPTool is an online tool available to help you create and share your data management plans to meet funder requirements and as a best practice for managing your data. DMPTool help provides step-by-step guidance for creating your own DMP and includes templates and sample plans to help you address requirements specific to Harvard and your funding sources.
- To use the DMPTool as a Harvard affiliate, go to the DMPTool and sign in using the HarvardKey.
If you need additional assistance, visit and contact the Harvard Library Research Data Management Program.