Search
Publishing
Research Data Management
Research Data - general, FAIR and Open Data
Research data are all data generated in the scientific research process. They range from quantitative measurement data, survey results, statistical findings, qualitative data such as interview transcriptions, field research notes, audio and video files to software and much more.
Find, use and cite research data
Parallel to the research idea and literature search, at the beginning of a research project there is the search for already existing research data, which can be used in the project. Research data are usually stored in data repositories that can be searched accordingly. In order to find a suitable repository, different search systems can be used. A well-known representative for a registry of research data repositories is e.g. Re3Data. Furthermore, there are also data portals that allow searching for datasets across multiple repositories (e.g. DataCite). If this is not the case, the legal aspects related to the data must first be clarified before any subsequent use. Regardless of the legal aspects, research data must be cited in the same way as literature sources in the sense of good scientific practice. Further information on citing and finding research data can be found at the following link: Forschungsdaten.info: Finden und Nachnutzen
Subject-specific handling of research data
Different research disciplines generate different types of research data, and the way in which the data is handled can also differ again depending on the discipline. The cornerstones of research data management thus form just a basic framework that must be enhanced on a case-by-case basis, taking into account specific project requirements, special research data, and discipline-specific idiosyncrasies. In many scientific fields there are specific RDM projects, subject-specific repositories or data storage or special tools for subject-specific support in RDM. A first introduction to subject-specific RDM can be found at the following link: Forschungsdaten.info: Wissenschaftsbereiche.
What is FAIR data?
Findable
The data and its metadata are easily and unambiguously discoverable by humans and machines, with machine-readable metadata being especially important.
Accessible
Access to the data is clearly regulated via a standardized communication protocol. Furthermore, access to the metadata must be permanent.
Interoperable
The data and its metadata can be combined with other data due to the use of formal standards, allowing automated exchange between computer applications.
The FAIR principles published in 2016 in an article in the journal Scientific Data now form the basis of many initiatives that aim to establish the principles in research. One global initiative, for example, is Go FAIR.Furthermore, more and more funding organizations such as DFG, BMBF or the EU demand that the data in the projects they fund should be FAIR. Whether data comply with the FAIR principles can be checked usingso-called FAIR assessment tools. An overview of well-known tools was compiled by the Thuringian Competence Network for Research Data Management. Further information on the topic FAIR as well as the benefits of FAIR data can be found at the following link: Forschungsdaten.info: FAIRe Daten.
The following link refers to a FAIR article of the TU Wien, where the principles are described in a practical way: Christiane Stork: The FAIR principles for research data (2020)
What is open data?
The Open Data movement is part of the Open Science movement, which has its origins in the free software movement of the 1980s. Since 2004, the Open Knowledge Foundation has been promoting open content, data and knowledge on an international level. For this purpose, it published the Open Definition, which is the basis for the Open Data Definition of the German branch of the Open Knowledge Foundation.
Open data is data that can be freely used, re-used and re-distributed by anyone - at most restricted by the obligation to name sources and “share-alike”.
This does not apply to personal data.
(Open Knowledge Foundation Deutschland: Open Data)
If possible, research data should be open. In addition to the demand from funding organizations to have FAIR data, open data is also increasingly coming into focus. It should be noted here that FAIR data is not open data. While open data must always be freely available and accessible, access to FAIR data may also be restricted. The FAIR principles simply require that access be clearly regulated. For example, if there is only one freely accessible contact where data use can be requested, the data is FAIR but not open.
Making research data publicly available has many advantages. Research results become verifiable and reviewable. Follow-up studies can be conducted on the basis of existing data without having to collect data twice. The publication of different data collections allows comparisons and combinations of data, so that completely new insights can be gained. Thus, data collection takes on a new significance in the context of scientific work. Furthermore, the data can be cited more easily, which is followed by higher citations of associated text publications by a scientist, and the data collection itself can be tracked and rewarded as a scientific achievement if necessary.