Data Management Plan
Like your Career Development Plan, a Data Management Plan (DMP) is also a living document and is designed to be updated and refined regularly throughout the life cycle of the research project. It includes details on how data is initially going to be collected, how it will be analysed safely, how it will be disseminated, and how it will be stored long-term with eventual destruction.
Most research grants are going to ask you for a DMP. Some might ask for a fully formed document (e.g., two pages), while others might ask for just a short paragraph on what sort of open access practices you are going to implement. Either way, it’s good to think through this section early in the research planning as it will likely inform a lot of your decisions you make about the project e.g., what types of data are you going to collect, where you are going to share it, who is going to have access to your data, how long you are going to store it for etc.
This blog is going to give you some specific areas to think about, and an example way you can structure this information for a research/postdoc grant application.
What is Open Access?
Open Access is the ‘practice of providing online access to scientific information that is free of charge to the user and that is re-useable' (The European Commission). For example, you publish an article open access so that everyone can access it without having to have a membership to the journal. You publish your dataset in an online free repository so that people can use the raw data for other studies. You publish a preprint of your work on ResearchGate so that people can see and comment on what you are working on before it is “really” published in a peer-reviewed journal.
In general, open access practices in research improve transparency, increase collaboration, and ultimately aim to having higher quality research that will have maximum societal and economic benefit. It is generally thought that research that is publicly funded should be as open as possible and as closed as necessary.
What does FAIR research mean?
Findable: Ensure the data is easily findable and identifiable using metadata and standard identification mechanisms e.g., Digitial Object Identifiers (DOI). E.g., having a standard file labelling system. Ensuring you include the version number on your documents, when it was last edited and by who. Think about the keywords you will provide so that your data and research is more easily found.
Accessible: The aim is to make your data and outputs as open as possible. If you cannot share a dataset, give an explanation as to why, and consider putting it under a timed embargo. If data is deposited in a free online repository, make sure you link the DOI so that people can find the files. If your dataset is available for download, think about if you need to prepare a document explaining how the data can be used.
Interoperable: This principle is concerned with your research data being easily used by others. E.g., if you have a dataset, can someone replicate your study using the dataset? Did you use open access software, or do they need specialised software to replicate it? Is the data structured in a standard format? As above, think about if you need to explain or give instructions on how others can use your research resources to repeat the study.
Reuseable: Think about the types of licensing you will use as this will decide how widely the data can be reused. If you embargoed the data, give details on when it will be available.
Science Europe DMP Template
The below template has been taken and edited from the Science Europe’s Practical guide to the International Alignment of Research Data Management.
There are seven sections, each with a number of subheadings. Although in your actual DMP you will have lots and lots of information in each field, for your grant you are likely going to be restricted to a couple of pages. So, keep that in mind while you are preparing it and remember to always read the criteria in the grant call document.
1. General information
Here is the place to give information on when the DMP will be created, who it will be reviewed by, and how frequently it will be reviewed.
2. Data description and collection
2a. Data collection process.
You are likely going to have a table or section in the grant where you talk about specific data collection methods so you can refer to that section here. Think about FAIR principles here, so what metadata are you using, how are you labelling files, are the methods you are using standard methods in your field, what software are you planning to use.
2b. Types of data collected.
Will the data be quantitative, video or audio files, transcripts etc. How big will the files be? What format will they be collected (e.g., MP3, csv)? Ideally, that data will be open to support widespread usage.
3. Documentation and data quality
3a. Metadata and documentation.
Here give details about the use of DOIs and ORCID, file labelling conventions. As mentioned above, think about having a cover page within your documents that outlines the version number, when it was last edited and by whom. State here where the metadata will be stored (again, ideally somewhere open access). If appropriate to your field, include the metadata standards you’re using.
3b. Data quality control.
Will you develop Standardised Operating Procedure documents? Will equipment be calibrated? Will data be taken in multiples? Will someone review a random sample of the data to ensure high quality? Will surveys be double scored? Will researchers undergo training?
4. Storage and backup during the research process
Your university ethics application forms can be useful for this section to copy and paste information that you already provided previously!
4a. Data storage and backup.
Is data stored online4? If so, how is the computer protected? Is it on a cloud storage? Is it backed up and if so, where is it backed up to and how frequently? Ideally data is stored on an intuitional cloud, with automatic backup to at least one other separate location. Avoid USB sticks and external plug-in hard drives.
4b. Data security and protection.
How can you access data if there is an incident? Do you have IT support from host institution? Who from the researcher team has access to the data? Is your DMP in-line with the institution’s data protection policy?
5. Legal and ethical requirements, and Codes of conduct
5a. Compliance with personal data legislation and security.
Mention your informed consent procedure if your work involves human participants. How will you anonymise participants (e.g., participant ID)? Will you remove any identifiable information so that they remain anonymous? Will you adhere to GDPR? Can you link Participant ID and de-anonymised data should someone want to withdraw from the study? Who is authorised to view participant personal data?
5b. Other legal issues.
Things like who owns the data and IP should be considered here. What is the IP policy of your host institution and is there an office where you can go for support (e.g., Technology Transfer Office or Office of Research etc.). If you have partners or a management board for your project, it might be a good idea to draw up a partnership agreement that details things like IP. You can also talk about things like an IP register here.
5c. Ethical issues and codes of conduct.
State where you will be seeking ethical approval (e.g., institutional and/or NHS etc.). Will there be a data protection assessment as part of the ethics application?
6. Data sharing and long-term preservation
6a. Data sharing (process).
Mention here that you will adhere to the institution’s policy on data sharing, how long data will be stored for. If you intend to deposit data artefacts on an open access repository, state what repository, what will be deposited, and by what licence here.
6b. Data preservation.
Indicate when data will be destroyed and how. List any future uses of the data. Mention here if repository uploads will be part of the publishing workflow of the project. Identify any institution or other support you will seek. You can again reiterate that no identifiable information will be published.
6c. Methods for access and use.
Will your sharing methods be in line with FAIR principles? E.g., will people need specific software or tools to access data, will it be free, is there an embargo etc.
6d. Persistent identifiers.
E.g., research ORCID IDs, Unique Repository Identifier (URI), Digital Object Identifiers (DOIs).
7. Data management responsibilities and resources
7a. Data stewardship.
Who will be responsible for what? E.g., postdoc, mentor, PhD student, research assistant etc. Roles might include data capture, data quality assurance, data back-up. Indicate who is responsible for implementing the DMP, how frequently you will meet and update the document.
7b. FAIR principles.
Mention here if data management time has been allocated within each work package, or the general time commitment of managing the project’s data. List any other resources (time or financial) required to prepare and store the data, including the long-term costs involved in maintaining the data.
Helpful resources
