6. Data management plan (DMP)

All the information dealt with so far should be summarized into a data management plan, typically referred to with the acronym DMP. A DMP is an important part of your research plan, usually as an attachment to it.

With the help of a data management plan, you can perceive the entire life cycle of the research data, i.e., where the data comes from and what happens to the data in the different stages of the research. It guides the researcher to plan the use of the data in every stage of the research project.

Data management of each research project should be planned, and a data management plan created before the project starts. It is important to update the DMP as things become more specific or change as the research progresses.

Why a data management plan (DMP)?

Research data management and its planning are an integral part of good research practices. Planning helps save time and money.

Many research funders (e.g., Academy of Finland, Horizon Europe) require a data management plan. Remember to check the guidelines of your funder and organization first, so you are aware of what kind of requirements they have.

Creating a data management plan before you start to collect and handle the data

reduces the risk of losing research data
helps understanding of complex ownership and user rights issues
supports making data FAIR: Findable, Accessible, Interoperable and Reusable
meets funder’s requirements
reflects principal investigator’s managerial skills as a project leader.

The content and structure of the DMP may vary, for example, according to the requirements and instructions of the funder.

Often it consists of the following sections (DMP 1 – DMP 6)

DMP 1: General description of the data

In the general description of data in the DMP, you need to describe what types of data are produced, collected, processed, and reused during the research project.

Does the research material contain numerical measurements, images, photographs, samples, or codes?
In which file formats will the data be (e.g., .csv, .txt, .docx, .xslx, and .tif.)?
Is the size of the dataset, for example, less than 100 GB, ca. 1 TB, or several petabytes?

An estimation of the size of the data is needed to evaluate the required disk space. In other words, the size of the data in the DMP doesn’t mean number of interviewees or variables. These are described in the research plan.

The general description tells how the coherence and quality of data will be ensured. The quality of data does not mean the quality of research methods, which is addressed in the research plan. Instead, the quality of data management described in the DMP ensures that no data are accidentally changed, and that the accuracy of data is maintained over their entire life cycle, due to the technical handling, such as coping, converting, or transferring data, or during its contextual processing and analysis.

Vinkkejä, tips

You can describe the technical handling of the data in a table format.

DMP 2: Ethical and legal compliance

Ethical and legal issues related to research data management are described in the DMP.

If personal data is being processed for purposes of scientific research, the following issues should be addressed:

what kind of personal data is collected (for personal data, see chapter 2)
is there need for an ethical review
how will data protection be ensured
is the data pseudonymized or anonymized
how will the anonymisation or the pseudonymization be done
what procedures are used to inform research participants about data processing?

In the DMP, copyright and intellectual property right issues should be described as well: who owns the data and what kind of agreements should be drawn. You need to check the guidelines of your organization and the funder of the research and be sure that the permissions for storing and sharing the data are in order.

Agreements regarding ownership and rights of use are made as early as possible in the project’s life cycle.

DMP 3: Documentation and metadata

Data documentation and the description of metadata must be carefully planned so that others can find, use, and cite the datasets properly. This includes humans as well as machines (computers).

README files and metadata standards help to describe where, when, why, and how the data were collected. In this section of the DMP, vocabularies and standard expressions utilized, such as ISO standards for dates or geographical locations, are told as well. A rich and comprehensive description ensures that the data are interpreted correctly using metadata standards, vocabularies, and README files.

Metadata plays a key role in the implementation of FAIR principles. This explains why many DMP models ask to describe metadata practices and the implementation of FAIR principles in the same section.

Vinkkejä, tips

Describe all the types of documentation (README files, metadata) that will help secondary users to understand and reuse your data.
Consider how the data will be organized during the project.
Describe, for example, file-naming conventions, version control and folder structure.

DMP 4: Storage and backup

The DMP specifies

where the data is stored
how to backup files
who will be responsible for controlling access to the data
how the secured access to the data is monitored and controlled during the research.

If there are several research partners involved, plan together secure transfer of data.

Taking care of data security is extremely important if the data is sensitive (e.g., contains (sensitive) personal data or politically sensitive information or trade secrets).

Vinkkejä, tips

The use of a safe and secure storage provided and maintained by your organisation’s IT support or other reliable IT provider such as CSC is preferable.
Do not use external hard drives as the main storing option.
Follow your institution’s data security requirements.

DMP 5: Opening, publishing and archiving data after the research project

In the early stages of the project, it is important to plan further use of the data, i.e., how the data can be made openly available to the scientific community. The DMP describes which part of the data can be made openly available or published.

If the data or part of it cannot be published, this section of the DMP can provide reasons for doing so. Similarly, in this section it can be stated that only metadata is published and the reasons for this are presented.

It is also important to tell where and when the data and/or the metadata are published or made available. In the initial phase of the research, the data journal or data repository may not yet be known, and this can be clearly stated in the DMP. It is good to keep in mind that the DMP is a document to be updated and that changes are likely to occur during the research.

DMP describes also which part of the data will be preserved and which will be disposed of, where the data with long-term value will be archived, and for how long. Long-term preservation is the process of preserving digital information for several decades or even centuries.

Use data repositories with a commitment to long-term curation.
Not all research data need to be archived: long-term storage is only for unique or hard-to-repeat data with long-term value.
Use archives and data repositories specializing in long-term preservation.

DMP 6: Responsibilites, roles and resources of research data management

The DMP describes the responsibilities of data management during the research project and afterwards of the data, i.e., who is responsible (name, institution) for which task. It also assesses the resources required for data management, i.e., what financial, temporal, workload related, or other resources are required to implement the measures described in the DMP.

The guidelines for planning data management instruct to keep in mind the implementation of open science and FAIR principles. In practice, then, this section of the DMP tells what resources are needed to store and open the data according to FAIR principles. The individual measures that support the FAIR principles are therefore not listed in this section.

Estimate your data management costs, for example, with the help of the following guidelines and tools:

Costs of data management (Utrecht University)
Data management costing tool (TU Delft)

Tools, templates and guidelines for writing a DMP

DMPTuuli is a tool created for helping researchers of Finnish research organizations to write data management plans. It helps you to consider all the relevant questions about data management. In order to use DMPTuuli, you need to create a DMPTuuli account or sign in with your institutional credentials (HAKA).

DMPTuuli can be used to work on a DMP with partners, and the completed plan can be attached to the funding application in the required format (e.g., docx, pdf).

DMPTuuli has DMP examples as well as general and funder- or organization-specific guidelines (e.g., Academy of Finland, EU, UEF). You can choose the guidelines you need in DMPTuuli when you start writing the data management plan.

General Finnish DMP Guidance and additional instructions for sensitive and confidential research data are available in DMPTuuli as well.

In addition to DMPTuuli, there are several other internationally used DMP tools as well, such as

DMPOnline (Digital Curation Center, DCC, Iso-Britannia)
DMPTool (University of California Curation Center)
Data Stewardship Wizard
Argos (OpenAIRE, EUDAT).

Note that in addition to this learning material, you can find UEF instructions and guidelines in UEF Data Support and UEF Library websites.
Familiarize yourself with General Finnish DMP Guidance.
If your research has a funder, find out if it has requirements or guidelines for data management.
When writing your DMP, try to give a comprehensive description about how you will manage your research data in the technical point of view.
Answer all questions that are relevant particularly for your research data.
Avoid overlap with the research plan, where you describe the research methods and analysis. In the DMP, you can refer to the research plan and vice versa.
A data management plan should not be thought as a simple administrative task for which standardized text can be pasted in from model templates, with little intention to implement the planned data management measures. So, describe only those measures that are really needed, e.g., to enable data sharing.
If something is still unresolved or only at the planning stage, it is also worth writing it in the DMP. This shows that you are aware of key date management measures and reminds yourself and the entire research group of the questions still open.
Implementing data management measures during the planning and development stages of research project will avoid later panic and frustration.

^(2023-08)

Finish!