Research data management
Research data are details that have been collected or created to serve as a foundation for analyses and validations of research results. The data can be composed of analogue or digital information, such as laboratory reports or software.
Here you get information about what you need to consider when it comes to research data that is created and needs to be handled during different parts of a project.
Save time and streamline the research process by planning your data management well as the project proceeds.
Requirement from research funders
It has become more and more common that research funders, both nationally and globally, require that researchers receiving research funding write a data management plan (DMP) for their research project. The funders’ DMP templates are often based on the so-called FAIR-principles. FAIR is supposed to function as a help and support in the data management of the project.
Creating Data Management Plans in DMPRoadmap
A data management plan (DMP) for a research project increases the chances for correct management and quality assurance of research material. A DMP describes how data is collected, managed, organised, stored, and made available during the research process and how to ensure the data is saved and archived correctly after the project is finished. The plan should be set up already at the beginning of the research project and be updated continuously during the working process.
In February 2020 Lund University launched a new system for data management plans – DMPRoadmap. The system supports researchers who receive, or have already received, project funding foremost from the Swedish Research Council, or when funders require that the university is responsible for setting up a DMP.
Are you planning to carry out research that involves people or the handling of personal details? If you are going to collect data containing sensitive personal data you might need to apply for an ethical approval. Since 2004, it is prohibited to embark on certain research without an approval from the Ethical Review Board.
All costs for a project should be estimated and included when budgeting for a project. It is often possible to apply for more money to cover costs during an ongoing project, but keep in mind to account for potential costs already when writing your application. Examples of costs:
- Time (managing, cleaning and coding the data)
- Software (costs for purchase or licence/year, start-up costs)
- Staff (data manager for the project, research engineer for programming)
- Safe storage (server space, safety cabinets)
- Ethical review application
At UK Data Service you can find information about costs in relation to data management. You can also find a checklist to help calculate costs for research data.
Note that if you want to reuse data it must be cited. Among other things, you need to state the copyright holder, where the data is stored, and possible version referring to a DOI or another persistent identifier.
As the project proceeds it is important to manage data in a structured way. This makes it easier both during the project and when you face the steps of archiving and making your material available to others.
Persistent file formats
Files should be stored in persistent formats, which makes it easy both to find out what you have done to the data and to reuse it. Files created or stored in formats generated by specific software can sometimes only be opened/used by those special programs. This shortens and limits the use of data, and other research material.
- File formats and long-term storage - on the website of the Swedish National Data Service
- How to organise data - on the website of the Swedish National Data Service
- Choice of format when archiving - on the website of the Swedish National Archives (in Swedish)
- File formats and standards - on the website of Digital Preservation Coalition
Think about potential consequences of information security, as it is a complex and important matter. Questions related to information security might concern digital preservation, system security and exposure to external/internal threats. Moreover, it could concern collection security such as protecting contents from loss/change, secure access to the material, storage processes, and legal and regulating aspects such as GDPR issues in relation to personal or confidential information in the digital material.
Be sure to structure your data so that it is easy to find and use, during and after your project. A material that is well organised could save amounts of time and effort as you work. There is no need for advanced solutions to be well organised – more simple ways will be sufficient. Go on reading to find out more about different ways to plan and think when starting on a project.
Folder structure and file names
Name your files according to a structure that is intuitive and/or describe the content – details such as date, place, or other things relevant for the material are good to use. Think in the same way as when you are organising files in folders to know what they contain without having to open them. Also reflect upon an easy way of monitoring different versions of your files.
Describe your data in a structured way. It could be a read me-file, a description connected to documents etc. When you upload files in an archive or a similar service, you use a form to describe what they contain. This generates metadata explaining what the amount of data comprises, and how it has been created and analysed. Metadata is also needed for machine management of data, to rediscover and reuse information and details.
Versioning and back-ups
One way of monitoring changes is to create distinct versions of your data files. All relevant files should also be backed up regularly. This is of extra importance when your files are being constructed to prevent a situation where data is being lost or where you need to enter your data all over again. Find out how the data environment works at your department and what support you can get when it comes to back-ups and the storing of data. Do not forget to keep back-ups in a different place should an emergency occur.
An important reason for versioning is provenience (the documentation of data objects’ origin and history). If you state what you have done to the material and someone at a later occasion questions the conclusions of your project, you can demonstrate the versioning and go back to show how it all happened.
When your project is about to be finished you need to make some important decisions concerning your data and what is going to happen to it onwards, and in what way. Investigate if there are any specific requirements from research funders or others regarding data availability and archiving.
According to Swedish law, research data must be archived.
Availability, reusing and sharing
If you are planning to make your data available for others there are several ways to do this. Note that if your data contains personal details it must be anonymised before it is made public.
Publish and share your data
Open research data is searchable, depending on where it has been uploaded – in a data repositorium, in special data journals, or as a supplement to a journal article.
- Share your data on the Swedish National Data Service (SND)
- On the website of Zenodo you can find a platform for open data within all subject areas
- The website of Figshare is an example of a platform for open data
- Further sharing services on the Re3data website
- More about searching and sharing data - on Lund University Libraries’ website
Feel free to use standardised licenses, for example CC licenses, to make evident how your data can be used.