Skip to Main Content

Creating Data Management & Sharing Plans (DMSPs)

This guide is designed to get you started on writing Data Management (& Sharing) Plans for grant proposals.

What is a Data Management Plan (DMP)?

A Data Management Plan (DMP) is a document that outlines the storage, archiving, and sharing policies for your published research data. 

Different grants will require different sections, but usually they contain information on: 

  • What type of data will be used or created during research
  • How the data will be handled, described, and stored
  • What standards of documentation will be used
  • How data will be handled during and after the project

The purpose of writing a DMP is to think preemptively about how to apply the many steps of data management and curation before the project starts because it is easier than do so ad-hoc or retroactively. 

Additionally, writing a data management plan offers many personal, institutional, economic, and research benefits, most importantly ensuring the integrity of your research. 

What is Considered Research Data?

Not all data you create during your research is considered research data. The White House defines research data as "the recorded factual material commonly accepted in the scientific community as necessary to validate research findings" (OMB Circular 110). 

Usually a DMP considers research data as the final datasets used and produced for analysis, and any accompanying metadata or documentation. Sometimes graphs, figures, derivative datasets, any custom software or code, and your publications may be considered research data under your funder's guidelines, it is best to double check. Additionally, only digitized data (not hand-written lab notes or collected specimen) should be considered "research data" for the purpose of DMPs. 

 

 

Data Management Plan FAQs

Why do I have to make a DMP now? 

To comply with federal research data policy, all research data created with federal funding must be made available to the public. Since the US is funded by taxpayers' money, then allowing them access to the data is seen as fair. Additionally, data management and sharing ensures that research dollars can be used as efficiently as possible and allow for reuse of information in the future. 

Aside from requiring submission of your data at the end of your grant, by requiring a DMSP at the beginning of your project the hope is that you will have thought through the process fully and will therefore be more likely to follow through. Ideally, treating your data as it will be made public throughout your project will make the final submission easier than retroactively trying to format and clean it correctly. 

Does my grant application need a DMP?

Almost certainly if it is a federally funded research project. Most exceptions are grant funded activities that do not produce research or data (e.g., infrastructure upgrades, travel, outreach activates, etc.). 

Even if your proposed research involves data that cannot be shared due to privacy concerns (such as human genomic data) a DMP is still needed to explain why the data cannot be shared, and what measures you have taken to share as much data as possible. Most sensitive datasets can still be shared through summaries or anonymization, which means it would still fall under data management policies. 

What needs to be covered in my DMP? 

Thankfully every agency that requires a DMP submitted also has either a template or list of items to be covered. 

For more information on your grant, see the Funder Guidelines tab. 

How complete does my DMP need to be? I don't know all the details for my project right now. 

Most agencies expect your project to evolve over the course of funding and will allow revisions to your data management plan. Although it is important to complete your plan with as much foresight as possible, it is understandable when repositories change, formats shift, and additional datasets are created. Just be sure to update your DMP. 

What if I don't want or have time to share my data for free?

Unfortunately, the conditions to receiving US Federal Government Funding is that any of the research created using it will be available to the public. If you do not want to share your data, you may consider looking at private grants instead.

Can I share it with an embargo period?

Again, unfortunately as a condition of receiving US Federal Government funding, research created using it must be immediately available at the time of publication or grant close.

For now, occasionally your funding agency (usually NASA, DOD, or DOE for security reasons) will grant you a short embargo period (usually about 1 year) for your research paper. However, with the OSTP August 2022 memo, after 2025 even your research papers must be published Open Access at the time of publication or end of grant. 

I'm afraid to share my data because it might be misused or misunderstood.

Understandably, it can feel very vulnerable to share your research data to a large crowd. However, with careful planning you can ensure your research is understood and used correctly in the future more than if you don't share your data. 

Sharing your data openly allows you to include important contextual information in your metadata and link your data to your publications resulting in greater understanding. Sometimes researchers can feel empowered after sharing their data because they are allowed to explain in greater detail how the data was collected, formatted, and checked for quality. It isn't uncommon for a researcher to also include an "appropriate reuse" statement in their metadata explaining what limitations reuse of your dataset might entail. 

For example in Geography, it is common to state whether a dataset is precise enough for detailed analysis or is best used only for cartographic visualization. 

Will my DMP impact my proposal scoring and opportunity for funding? 

Although each funding agency handles it a little differently, usually it is the completion and inclusion of a DMP that is considered, not the content. Upon grant award most agencies will work with you to fix the respective concerns in your DMP, but will not block funding because of them. 

HOWEVER, if your project specifically relates to data sharing or your data management plan is integral to understanding your proposal, it may be included as a part of your scoring.

DOE applications are the exception and may be denied immediately if your DMP does not met the requirements at submission.

How is noncompliance handled? 

Again, it varies by grant. Most require proof that you are following your DMP annually during your project check-in. Noncompliance may result in termination of funds or considered in future award decisions.

How will I fund the cost of sharing my data?

Most agencies allow award funding to be used towards the cost of sharing, preserving, and archiving your data - as long as it is included in the proposed budget. See our funding tab for more information.

Can I host my own data instead of using a repository, e.g., my own website, Google Drive, Box, etc.?

Data repositories offer additional security, infrastructure, findability, curation, long-term funding, and other preservation services that self-created or cloud storage options do not have. Most agencies have pre-approved domain-specific repositories that can increase the impact and longevity of your research offering searchable metadata, search engine indexing, and long-term access.  

In theory you can, but it would be difficult and is likely not worth the time and monetary investment. Hosting your own data for the sake of meeting federal grant requirements is only recommended in instances where there is not a repository that is capable of hosting your data due to file size, format, or other restrictions. It is recommended to talk to a librarian or someone at OIT before making your own repository.

How do I get started?