Data storage

How data is stored is an important component of data management. It is good practice to think about storage solutions during the planning phase.

Many institutions have guidelines in place. So it would be best if you start by checking with your local institution for relevant systems and procedures. Below are some general points about data storage.

Data Classification

The classification of the data dictates the level of security and access throughout the project’s lifecycle. Correct classification is needed to ensure personal data is protected. Your local institution or ethics board may have their own specific levels and definitions of a universal classification scale ranging from open to strictly confidential.

1. Open – Green

This class is used when it does not cause any harm to the institution if the information becomes known to unauthorized persons. Completely anonymous data falls into this category. So do animations of motion capture data and unidentifiable statistical material. You may also have received written consent to share data openly.

2. Restricted – Yellow

This class is used when it could cause certain damage to the institution if the information becomes known to unauthorized persons. Data sets can contain minor amounts of non-sensitive personal data. This could include optical motion capture data or video material that may be used to identify people.

3. Confidential – Red

This class is used if it could cause harm to public interests, the institution or individuals if the information became known to unauthorized persons. This includes health-related data, special categories of personal data gathered in surveys (like sexual orientation), or even information related to the safety and security of the local building or IT systems.

4. Strictly Confidential – Black

This class is used if it could cause significant harm if the information became known to unauthorized persons. Projects involving protected populations or that may result in great financial or commercial value also fall into this category. Motion capture data does not usually fall into this category.

Choosing a storage solution

Many factors go into choosing the right storage solution for your project. Movement analysis data can be large, personal, use licensed software, and need unique transfer routines. The higher the level of confidentiality, the more secure the storage solution should be. Your local institution will have recommendations for using portable devices, cloud storage, local storage, and network drives you have access to.

Considerations

While the specific storage solution you use for your project will depend on both the study method and your local institution, here are some general recommendations for data storage:

• The volume of movement analysis data can be large; choose a solution capable of storing, accessing, and transferring large files.
• Keep a copy of the original raw data in a protected location to keep it not editable.
• Avoid using low-security devices and time spent in locations without backup.
• Keep working copies of the data in one location to maintain “version control”.
• Use safe storage solutions. Focus on solutions that offer encryption and networked backup through your IT department.
• Make sure to generate good documentation together with the data.
• Software needed for creating and processing movement analysis data can be licensed, so the storage solution should ensure appropriate software access.
• Plan ahead if data sharing is needed across institutions and countries. It may take time to sort through the legal issues of sharing data.

Archiving and Publishing Data

It is common to think about archiving as the last chain of a data management procedure. That is when you are done with your research, and want to “box up” your data for storage, preservation, or publish it for reuse. When you archive your data, you make sure you can read and access them for the long term. Data archiving involves storing your data safely, in a suitable file format, and with adequate documentation. Storing is not the same as archiving!

Publishing your data is the act of publicly disclosing the research data you have collected, making them findable, accessible and reusable. It makes your data reusable for other people.

Here are some things to consider when you are ready to archive and publish your data:

1. Select data for archive: Does the data have the potential for reuse? Is it linked to a published article? Is it of high quality, unique, and/or innovative?
2. Choose an archive: It is becoming increasingly common that journals require depositing data used in papers in a specific archive. This is a good idea since it creates a link between the article and data. However, in many cases, only a subset of a dataset will be deposited as part of a journal publication process in many cases. A better long-term strategy is to archive a complete dataset on its own and according to the FAIR principles. There are a lot of archives to choose from, and they can be general archives like Open Science Framework as well as domain-specific like Physionet. The former is an unsupervised archive where you are responsible for the content yourself. The latter is based on a peer-review process and professional data curation.
3. Prepare for Archive: Organize the files and folders for the archive bundle; you may not want to archive the entire dataset. Make sure the documentation is completed. Ensure privacy and copyright are protected as needed.

When the dataset is archived and published, you can refer to it from articles. It also helps spread the word to your research community, for example, in web forums or on mailing lists.

References

• Data Storage in the RDMkit by Elixer
• Store chapter of the Data Management Expert Guide by CESSDA
• Classification Guide from the University of Oslo