How to securely store and manage scientific data

Correct and secure research data management is an essential part of a researcher's daily work. At Tallinn University, we have taken steps to provide better facilities and support for secure data storage during research. The solutions in place enable the project team to store data using cloud storage services so that the stored data can be accessed and processed by all project participants. The university's IT department provides solutions for storing data from ongoing research projects, while we have a cooperation agreement with the Tartu University Library  for storing and where necessary publishing research data from completed projects. Supporting materials on data management are being developed and updated on an ongoing basis and are available on the university's internal website.

It is advisable to think about what research data will be collected/produced, how it will be managed, stored, archived, and to whom and to which research data will be made accessible, before starting a research project. Attention should also be paid to the restrictions on the use of scientific data, the potential growth in the volume of scientific data to be used, etc.

Storage of scientific data during the project

It is strongly recommended that the scientific data generated through research  are stored in a repository that is not located on the local hard disk of the computer used for a researcher’s daily work. The repositories managed by Tallinn University for storing research data are GoogleDrive and OwnCloud. You can read the official user interface functionality descriptions of these systems - GoogleDrive and OwnCloud - at the following links.

In order to decide which repository to use for storing scientific data, the level of security needed to access  the scientific data needs to be defined. More protection and attention is needed for special categories of personal data and other scientific data containing more sensitive personal information/data. For more information on the secure processing of personal data in research, please consult the Data Protection Guide - Processing of personal data in research, available on the internal website.

If the scientific data does not qualify as special scientific data (e.g. key data, non-special personal data, ordinary personal data, other data), a Google Drive repository is recommended, where we already habitually store and share our documents. Google Drive is used by the University as a secure service with data servers located within the borders of the European Union. The latter is generally a sufficient level of security for the storage of normal scientific data. In Google Drive, the project team can store data in a way that a shared drive is created by the IT department, which ensures that data does not need to be transferred from one person's account to another when staff change positions in the organisation or leave. To request a Shared Drive, a request is made by an internal letter through the WD document management system to the Head of the IT Department. The request must include the name of the shared disk (Drive), the associated project, the expected time of use and the administrator.

If the project involves the use/production of sensitive personal data or other personal data of a more sensitive nature, or if part of the data qualifies as sensitive or sensitive personal data, OwnCloud, the storage repository for research data that the University has adopted to date, is strongly recommended. With this service, the data is stored on a server managed by the university itself, and therefore security is guaranteed at a higher level than Google Drive. 

To request access rights, a request is made by internal letter via the WD document management system (request to use the OwnCloud environment). Access to the server will be automatically granted to PhD students (without any request). To request a shared drive for the project team, a request is made by internal letter via the WD document management system to the Head of IT. The request must include the name of the Shared Drive, the associated project, the expected time of use. For more details, please refer to the document distributed on the internal web site entitled 'Computer Network Access Policy' (Chapter 14: Data Storage Information System - OwnCloud). 

Other recommendations

Structure of scientific data storage, file names, versioning, metadata.
When choosing file names, it is advisable to choose systematic and consistent names. The structure of catalogues should be hierarchical, simple, logical and short. 

It is advisable to think about the need for versioning in the project. 
Metadata, i.e. data describing the data (administrative, technical, descriptive), think about how to distinguish/bind them. It is advisable to include a general information file that gives an overview of the content and structure of the project. 

Format

When choosing the format, it is preferable to use standard, machine-readable formats
Access to research data, terms and conditions, sharing (sending links), re-use.
It is advisable to consider early on what restrictions there are on sharing research data; what the roles of different project members are; under what conditions links can be sent. 

Project closure, archiving and publishing the research data 

At the end of the project, if Tallinn University is the lead partner and there is no obligation to archive the data in other repositories, it is advisable to consider archiving, and where possible publishing the research data from the project in the DataDOI repository where a Digital Object Identifier (DOI) will be generated. Thus research data becomes citable for others and it will be registered in DataCite registry. Where it is not possible to publish the research data, it is still advisable to publish the metadata about the research data (if possible) and how access might be available and to who.

 
Question regarding the management of research data can be directed to: Kristjan Alliksoo, kristjan.alliksoo@tlu.ee. Archiving research data after the end of the project in DataDoi repository: Tartu University Library data administrator Evelin Arust, evelin.arust@ut.ee
Advice regarding data management plans can be received from research coordinator of your school or from Xavier Dubois, xavier.dubois@tlu.ee