Research Data Management

This chapter was written in collaboration with research support functions at Stockholm University, Karolinska Institutet, Umeå University Library, Chalmers, SciLifeLab and Swedish National Data Service. This chapter is published under the Creative Commons License CC BY 4.0.

Good research data management (RDM) is essential to good research practice and typically involves five steps. The steps may vary slightly depending on the type of data you handle and the customs of your research field. If your research requires that you handle personal data, your university may have a checklist for what you need to do before starting the project; see these examples from the University of Gothenburg, Umeå University, and Stockholm University.

The first step of RDM is planning. You may need to apply for ethical review, get data access approvals, create data sharing agreements, and other relevant documentation as part of your planning. Typically, personal data, large data volumes and external collaborations complicate your RDM. You facilitate your planning by creating a data management plan (DMP). Many funders require you to set up a DMP to ensure that you have all the necessary aspects of RDM covered. Your university may provide a tool for writing your DMP. 

A DMP should describe 

• the types, formats and volumes of data used in your project
• the systems and solutions you use to manage, analyze and store data, and their security
• potential ethical and legal issues (e.g., with sharing personal information between project parties)
• how you address Open Science requirements and ensure long-term preservation of data
• the costs/resources associated with RDM.

Your DMP should be revised when necessary.

The second step is to create, collect and/or acquire the data you will use. The data should be saved in appropriate formats, organized in a structured, consistent way, and stored in secure systems with backup. Additional security measures (e.g., aggregation, pseudonymization, encryption) will be required if you handle personal or otherwise sensitive data as part of your research. Consult the Data Protection Officer at your university to ensure that the solutions you intend to use for data storing and data processing are secure for your type of data.

Step three will cover any processing and analysis of the data. This may include cleaning the data, merging datasets, applying quality controls, and conducting various analyses. Keeping a logbook is sometimes mandatory and can help document the research process so that others (and you later) can follow every step.

Processing and analysis may include data sharing with collaborators. Often, collaboration that entails sharing data between organizations will require formal data sharing agreements to detail rights and responsibilities with regards to data management and use. Sharing personal data between collaborators must be included in the application for ethical review and the information given to the study participants.

Step four is the publication of the results. Often, your university, funders or journals will have Open Science requirements which require that the data underlying your published results should be made available. You publish data in a data repository to make them open and FAIR. Depending on the nature of your data, you might publish them openly or with restricted access. If necessary, publish only a description of data (i.e., metadata). You are responsible for ensuring that data you publish openly never contain personal or sensitive personal data (GDPR), intellectual property data, commercial or financial data, or otherwise sensitive data (according to Public Access to Information and Secrecy Act).

The fifth step looks toward the future and requirements for the long-term preservation of the research data. Many universities have a records management plan [in Swedish, dokumenthanteringsplan or informationshanteringsplan] that states how long you must preserve various research documents.

All steps should be traceable as part of the research process and documented accordingly in your DMP. Additional information and tips on RDM are available at the Swedish National Data Service (SND). Most universities offer support functions (e.g., grants office, ethics function, Data Protection Officer, research data support, archiving and IT support) that can help you with your RDM. Consult their services when you set up your DMP to ensure that your efforts align with university rules and regulations. Contact them whenever you have questions.