What is a Data Management Plan (DMP)?
Research data are recognized as a high-value resource that has the potential to enhance scientific knowledge. Research funders are increasingly requiring that grant applicants provide a detailed data management plan (DMP) as a part of a research proposal. The DMP is peer-reviewed and it is used for evaluation purposes.A data management plan is a formal document that describes how the data will be handled during a project and after its conclusion. The DMP is intended to cover all the steps of the data life cycle, namely how the data is discovered, collected and organized. Moreover it concerns topics on data quality, preservation and dissemination.
How do I write a good data management plan?
A well-documented DMP must comply to the funder requirements that are very different depending on the research community own standards and methods. Knowing the requirements of the organization to which the proposal will be submitted can save time and effort while writing a DMP.
The success of the DMP also depends on having significant knowledge about the data that will be collected during the project. For instance, the volume of data that will be generated has both an impact in financial and time costs, such as preparing the infrastructure to allocate the data and creating the metadata that will ensure data quality and preservation. Therefore, the more accurate is the information about the types, sources, volume and the format of the data more likely the DMP strategy will be successful.
Another aspect is to clearly understand how the data will be documented, since data on its own may not provide the needed information to be exploited, e.g. rows and columns of numbers. Data documentation in order to enable discovery, reuse and proper citation, relies in metadata describing the research context. There are community-based metadata standards that can be used to performed this kind of activity.
Relying solely on personal computers to store the data may expose data to many risk scenarios, compromising its reuse. Thus, DMP´s must include a data storage and preservation strategy. An aspect to consider is the value of data since not all the data calls for long-term preservation, for instance observational data is expensive to collect and, in most cases, cannot be replicated, so there is a need for it to be stored permanently.
It is recommended to store copies in different locations and plan for regular backups. Data repositories are suitable remote locations to ensure data preservation. There is a growing body of general data repositories, in the likes of Figshare and Zenodo, or discipline-specific orientated ones that can be accessed via online catalogs (re3data). Institutions may already have their own data repositories for researchers to look for. Moreover data repositories have specific policies that can influence the DMP, namely costs associated to data curation activities. Researchers may also consider a dissemination strategy regarding data availability (when, how and what?).
Finally, a good DMP practice is to consider data quality issues, project´s data policies, roles and responsibilities and allocate budget to the data management activities.
The following question need to be consider when preparing a DMP:
– What types of data and formats will be captured during the project?
– What is the expected volume of data that will be generated?
– In which way the data will be stored (over the duration of the project) and preserved to ensure its future reuse (when the project ends)?
-Which metadata standards will be used to describe the data?
There are any available tools and guidelines to help me in the development of a DMP?
The DMPtool is free and helps researchers to create DMP. This tool can be used to develop generic plans, yet it provides guidance from specific funders who require DMP.
The Digital Curation Centre DCC has developed the DMPonline that displays different templates that represent the requirements of different funders and institutions, offering support to write a DMP.
The European Commision, under the H2020 Programme, promotes the Guidelines on FAIR Data Management, in order to helps its beneficiaries to make their data findable, accessible, interoperable and reusable. Furthermore, The ‘AGA — Annotated Model Grant Agreement’ is a user guide that aims to explain to applicants and beneficiaries the General Model Grant Agreement (‘General MGA’) and the different specific Model Grant Agreements (‘Specific MGAs’) for the Horizon 2020 Framework Programme for 2014-2020.
