A Data Management Plan (DMP) is a two page document required by funding agencies that describes what you are going to do with your data during and after your research project. According to the 2013 White House Office of Science and Technology Policy memo, data is defined as "the digital recorded factual material commonly accepted in the scientific community as necessary to validate research findings including data sets used to support scholarly publications, but does not include laboratory notebooks, preliminary analyses, drafts of scientific papers, plans for future research, peer review reports, communications with colleagues, or physical objects, such as laboratory specimens.”
The Data Management Planning guide provides common sections required in granting agencies DMPs, although format might vary based on funding announcement. Each section contains questions to help you identify your data needs, and supplies examples for how this can be approached.
WMU Intellectual Property Policy describes the obligations and rights of the researchers at WMU
DMPTool is a website providing guidance on writing DMPs that meet funder requirements
Example DMS Plans is a directory created by the NIH DMSP Guidance Working Group that compiles published and template DMPs from various disciplines and funding agencies.
Request a review of your draft DMP by submitting a webform or contacting your data librarian directly.
| DMP Section | Sample Language |
|---|---|
|
State data types and file formats Consider: size and accessibility What are the types of data produced and their file formats? |
The proposed research will include 5 MRI images from 30 participants, for a total of 150 DICOM image files around 600MB in size. Self-reported demographics and health surveys will also be collected on paper and compiled into an Excel spreadsheet. |
|
Determine documentation format and standards Consider: standard terminology and file-naming practices What are the scientific standards and structured metadata used by your discipline [metadata schema]? |
A data-dictionary will be used to define the variables, provide metrics, and explain coding decisions. A README file will also be used to explain project purpose and state connections between project files. |
|
Define roles and responsibilities Consider: who does what and with what frequency Where is data backed up in the short term? |
The Principle Investigator (PI) is responsible for the deposit, maintenance and management of the data. Graduate student lab members will work with the PI on providing data documentation. Active data storage will occur on the departmental server. The lab manager will be responsible for regularly scheduled data backups to an external hard drive on a weekly basis to prevent data loss in the case of system failures. |
|
Establish dissemination and sharing policies Consider: obligation and audience Are you under any obligation to share your data (funder or journal)? |
Any datasets and accompanying documentation generated under this project will be deposited into Zenodo (https://zenodo.org) for long-term preservation. Zenodo is an open access repository that specializes in preserving software and issues DOIs, which will be included in each resulting publication. Code used for data processing and analysis will be made publicly available through GitHub (https://github.com), a web-based platform. Any reports, presentations, manuscripts, and other documents that record research outputs generated under this project will be deposited into ScholarWorks (https://scholarworks.wmich.edu/), Western Michigan's institutional repository. |
|
Plan for preservation and archiving Consider: accessibility in the long term How stable is your long-term storage choice and what guarantees does it provide? |
After study completion, proprietary files will be converted to open formats to maximize data reuse. Excel files will be converted to .csv and field notes will be scanned into .pdfs. Data resulting from this research will be shared via the generalist repository Dryad, which provides metadata, persistent identifiers (i.e., DOIs), and long-term access. Data will be made available as soon as possible or at the time of associated publication under the CC0 license. Dryad datasets are backed up to Merritt, the UC’s CoreTrustSeal-certified digital repository, for long-term storage and accessibility. Procedures in place to ensure dataset preservation include storage of data files in multiple geographic locations, regular audits for fixity and authenticity, and succession plans in the event of repository closure. Consistent with WMU policy, data will be retained for three years after study closure. |
|
Bonus: Funding Consider: additional costs for inclusion in grant proposal Will you be purchasing secondary data? |