Research data

Research data

What is meant by research data?

Research data is the foundation of modern scientific work and is generated, collected, processed, discarded, analysed and archived by scientists on a daily basis. These data often take a discipline-specific form and dealing with them, such as archiving them in the sense of good scientific practice or publishing and re-using them, is often a major challenge for researchers. The German Research Foundation (DFG) supports scientists in this area by setting up a national research data infrastructure (NFDI) and the TiHo has also set up a central contact point for its members for all questions relating to research data management.

What does the library offer on this topic?

Sponsoring organisations and scientific journals are increasingly making demands on the availability of research data within the framework of policies or guidelines. The TiHo helps to meet these requirements and implement good data management by offering advice on the topics:

  • Creation of data management plans
  • Researching research data
  • Support with data organisation
  • Data curation and metadata management
  • Publication of research data
  • Use of persistent identifiers
  • Selection of licences

 

We will be happy to answer all your questions and, if you wish, organise workshops and seminars on research data management for all members of the university. Get in touch with us!

What is a data management plan?

A data management plan (DMP) formally summarises all relevant processes related to the creation and management of data already during the planning phase of a research project and is dynamically adapted to current conditions during the course of the project. Some research sponsors (e.g. Horizon 2020 or the BMBF) increasingly require the inclusion of such a DMP when applying for project funding. The structure of a DMP typically includes planning, recording, processing, documentation, archiving, re-use, reproducibility and quality assurance of the collected data. The TiHo library is happy to offer you support in designing the data processes and also for the correct formulations in funding applications.

How do I find research data?

Research data is typically published in the context of a traditional publication and is available as a supplement or external resource in a repository. You can use the customised searches of  DataCiteor Google Datasearchto search specifically for reusable data.

In addition, there is the re3data.org (Registry of Research Data Repositories)which is maintained by volunteers. More than 2000 subject and institutional repositories are listed there according to various criteria.

How do I cite research data correctly?

To ensure that the data producers can also be appropriately credited for subsequent use, data should be made accessible on a trustworthy infrastructure (repository). When research data is submitted, detailed metadata (including content, authors and contact details, etc.) is recorded and a DOI (Digital Object Identifier) is assigned for correct citation.

As the central authority for assigning DOIs, DataCite recommends the following short or detailed citation method for research data:

Authors (publication year), title, publisher, identifier (usually the DOI)
or
Authors (publication year), title, version, publisher, type/type of data, identifier (usually the DOI).

How can I share my data with cooperation partners?

If you would like to share research data with your team or cooperation partners, then as members of the TiHo you can, on the one hand, make use of the diverse services of the IDS, such as the TiHoCloud, or other trustworthy institutions in the German research landscape. These include, above all, the Sync&Share service AcademicCloud and B2DROP at national and European level respectively. Please bear in mind, however, that sync&share services are generally not a suitable backup for your data and are primarily intended for work on different devices.

How can I reuse research data?

If you have successfully found an interesting research dataset, the possibilities for further use in the context of your own research project are determined by observing the licence granted. This is because copyright law also applies to research data, which leaves the sole decision on the type and scope of subsequent use to the intellectual creator.

 

Fortunately, however, many research data are nowadays made available under the very open provisions of the Creative Commons licences. The idea behind granting open licences is to promote the dissemination of one's own research data and thus increase the frequency of citations.

How to publish research data?

Well-established research data management requires an active debate about the value and re-usability of the data generated. The degree of transparency and re-use desired can be regulated by issuing simple and clearly formulated licences.

A selection of subject-specific repositories for data publication can be found on the ZB MEDwebsite, on the website of the journal Scientific Data, which belongs to the Nature group, or in the Registry of Research Data Repositories: re3data.

Some open access journals publish so-called data publications that focus on methodological aspects in the review process. Examples of such journals can be found in the central wiki: forschungsdaten.org.

Under which licence should I publish my data?

As a member of ELAN.eV, TiHo members have access to a tool to help them decide on the right licence. We are happy to provide access to this tool on request. For the best possible reusability, we recommend using the Creative commons Licencesor (often better suited for data and their compilations) the Open Data Commons.

In which repository can I store my data (free of charge)?

We are currently working on expanding TiHo eLib into a repository for all areas of research in veterinary medicine.
Alternatively, you can use the zenodo platform to publish research data free of charge. This generic repository is operated by the CERN Data Centre and enables an integrated peer review process. Another well-known repository for the life sciences and medical fields is Dryad, which is subject to a fee. Unlike zenodo , however, all data published there must be in the context of a peer-reviewed article.

The  re3data platform offers you the opportunity to search for other (subject) repositories under various aspects and possibly come across other interesting data collections.

What are the reasons for publishing research data?

By publishing research data independently, they become visible as your own achievement and can be recognised as a scientific contribution - through citations. By linking data and publication via persistent identifiers (e.g. a DOI), you also underline the transparency and verifiability of your results and enable the reusability of your data for new questions and interdisciplinary approaches. The avoidance of multiple collections of similar or identical data and the fulfilment of the requirements of different research funding bodies also speak for the open accessibility of research data. The decision as to whether and to what extent you publish data is generally left to your own discretion (except, for example, in the case of journal requirements).

How do I structure my data in a meaningful way?

During a research project, a wide variety of data sets are created, which are often kept in different versions and processing stages. For long-term usability, rules and conventions for correct naming and management should be established and well documented according to the size of the data collection. If possible, these rules should be agreed upon for an entire working group or for overarching research projects together with the cooperation partners.

Examples and recommendations

Links to the topic

Reading to get started with research data management

 

Tools and services for research data management

Infrastructure for data storage:

 

Data Management Plans

 

Data portal and repositories

 

Data structuring

 

Licences

 

Requirements of research funders