Research data management

from Wikipedia, the free encyclopedia

Research data management (FDM) describes the set of all methodological, conceptual, organizational and technical measures and procedures for handling research data over their life cycle. The measures and procedures can vary greatly depending on the research area and purpose for which the research data is to be used and therefore include a wide range of methods and topics. Research data management also includes definitions of what should be done with the data after a research project has been completed; z. B. where and how the data is archived and who has access to the archived data.

Research data arise in the course of scientific work. They are subject to a life cycle that ranges from the planning of a research project through collection, evaluation and storage to the sustainable archiving of the data. Research data can vary greatly from subject to subject. This may require subject-specific or even data-specific solutions for managing the data.

Research data management is usually embedded in a research data infrastructure that opens up scientific databases in a standardized form, networks them and makes them sustainably usable or offers services for these purposes.

meaning

Digital research data form an essential foundation of scientific work. Modern research institutions therefore need efficient, high-performance concepts for the management of their data, which enable the systematic handling of the research data with standardized procedures.

Further reasons for systematic research data management are ensuring the verifiability of research data ( good scientific practice ), maintaining the citability and ensuring the reusability of the data for subsequent research projects.

The Rectors' Conference has a recommendation "management of research data - a key strategic challenge for university leaders" in 2014 adopted that emphasizes the importance of research data management in the light of an exponential increase in volume and increasing complexity of the research data and make recommendations for the efficient management of the data.

Other institutions such as the German Research Foundation (DFG) or the Council for Social and Economic Data also give recommendations on how to handle research data .

Research data management tasks

This topic can be approached in different ways. Most often, life cycle models are used, which assign the tasks to be mastered to the individual life cycle phases of the research data. The life cycle approach is therefore described below as an example.

Life cycle dependent tasks

The life cycle of research data can be subdivided into different phases, each of which is accompanied by different processes and places different demands on research data management.

  • Planning: Good planning of the data management of a research project ensures that the data can be recorded and processed in a structured manner. A data management plan (DMP) is often used here.
  • Data collection: The collection or collection of data and the addition of metadata is very different depending on the subject. In a scientific laboratory, these can be measured values that are recorded electronically by a computer-controlled device; in the humanities these can be texts or text analyzes.
  • Documentation: The data can be stored and made available in data repositories , for example . Criteria for the selection of data for storage and access rights must be clarified and the risk of data loss minimized.
  • Analysis: Processing and evaluating the data form the core of the scientific work.
  • Publication: With the publication, the research results obtained are published and shared with others.
  • Archiving: Long-term archiving ensures the permanent availability and traceability of the data and enables its reuse. Additional checks and data enrichment are often necessary here.
  • Re-use: later usage options and access rights must be clarified. In the case of publicly funded research projects, the research data should often also be publicly available (open access). Research results from industrially financed projects, however, usually remain the property of the respective companies.

Overarching tasks

The overarching tasks include cross-cutting issues that are important in every stage of the life cycle. These include:

  • Organization: Selection of an organization that takes responsibility for the secure storage of research data and ensures its availability in the long term.
  • Financing: Development of a financing concept for data storage.
  • Legal aspects take into account issues such as copyright, data protection (e.g. protection of personal data ) and licensing .
  • Identifiers: development of a thoughtful approach to the identification of research data and supplement metadata . This improves the retrievability of the data.

Technical implementation

Electronic storage and management of research data can be done in a variety of ways. Research units have their own IT systems or cooperate with a data center or a provider of corresponding systems. Mostly, data repositories and archive systems are used that are based on databases . Cloud-based technologies bring new possibilities and inexpensive storage and computing capacities.

Scientific research laboratories traditionally use laboratory journals to document the experiments carried out , which today are mostly operated in the form of electronic laboratory journals .

The FAIR Data Principles describe some requirements when storing research data :

  • Findable: The data should be able to be found again.
  • Accessible: The data should be accessible over the long term.
  • Interoperable: The data should be technically reusable and combinable with other data sets
  • Reusable: The data should be analytically and intellectually reusable.

See also

literature

Web links

Individual evidence

  1. ↑ a b c d Jens Ludwig, Harry Enke (ed.): Guide to research data management, handouts from the WissGrid project . Glückstadt 2013 univerlag.uni-goettingen.de (PDF)
  2. ↑ a b c Research data management - a handout. Research data working group of the "Digital Information" priority initiative of the Alliance of German Science Organizations, May 2018, accessed on August 9, 2020 .
  3. ↑ Maxi Kindling, Peter Schirmbacher, Elena Simukovic: Research data management at universities: the example of the Humboldt University in Berlin. In: LIBREAS. Library Ideas 23 (2013). 2013, accessed August 9, 2020 .
  4. ↑ a b c Guidelines for handling research data. German Research Foundation, September 30, 2015, accessed on August 9, 2020 .
  5. ↑ a b Management of research data - a central strategic challenge for university management. Recommendation of the 16th HRK general meeting on May 13, 2014. University Rectors' Conference (HRK), May 13, 2014, accessed on August 9, 2020 .
  6. ↑ a b research data management. University of Kassel, accessed on August 9, 2020 .
  7. ↑ Basic information on research data management. In: Short version of the RatSWD Output Paper 3 (5), 2nd edition (2018). RatSWD - Council for Social and Economic Data, 2018, accessed on August 9, 2020 .