Research Data Policies

What is a research data policy?

A research data policy describes the guidelines for handling research data. There are different kinds of research data policies, for example:

• Journal and publisher policies
• Institutional policies
• Project-specific policies
• Discipline-specific policies
• Policies of research funding organizations

Journal and publisher policies

Since 2016 publishers such as Springer Nature,[„Research Data Policies.“] Elsevier[„Sharing research data.“] and Wiley[„Sharing and Citing your Research Data.”] have set new guidelines for the handling of research data and applied them to their journals. They are guided by the Transparency and Openness Promotion (TOP) Guidelines [Nosek, et al] published in 2015 by the Center for Open Science (COS). Publishers usually distinguish between three to four types of research data policies. The following are examples of the policy types and, in parentheses, examples of journals published at Springer Nature that use these types:

• The sharing and citation of data is encouraged (‘Photosynthesis Research’)
• The sharing of data and proof of data accessibility is encouraged (‘Plant and Soil’)
• Data sharing is encouraged and data availability statements are mandatory (‘Palgrave Communications’)
• It is a prerequisite to share data, to verify this and to allow peer review of the data (‘Scientific Data’). Depending on the journal, it is therefore necessary to check carefully which of the publisher policies apply.

Institutional policies

Institutional research data policies are increasingly being adopted by universities and other educational institutions to regulate the management of research data and to clarify the basic legal aspects. [forschungsdaten.org] In addition to the regulation of open access of research data, they also assess and decide on the allocation of the research institution’s personnel, organisational and technical capacities for RDM, i.e. cost and resource management [Hiemenz and Kuberek]. In half of the German university RD policies, statements on the costs of RDM are considered important in relation to a data management plan.

[…] One example is the research data management policy of Humboldt-Universität zu Berlin [Humboldt-Universität zu Berlin], which was introduced in 2014 […]. In addition to institutional policies, which often have an overarching character and are meant to represent the entire institution, project-specific policies can also be created. These define project-internal regulations and standards and can be adapted specifically to the respective project.

Discipline-specific requirements

For some subject areas specific guidelines for the handling of research data already exist (e.g. psychology, genetics, biodiversity, linguistics, educational research, social and economic sciences). In the social sciences, for example, there is an agreement on the cooperation of European data archives, which was drawn up by CESSDA [CESSDA] (Consortium of European Social Science Data Archives).

In the life sciences, the policy on “Good Clinical Practice (GCP)” and the “Principles of Good Laboratory Practice (GLP)” determine how the data are managed. Both principles are anchored in German legal code [BMJV]. Such discipline-specific requirements are necessary since research data are very heterogeneous and the way they are handled can vary greatly. At the same time, the definition of these standards also serves to establish comparability and interoperability between disciplines.

Thus, the discipline-specific data formats (e.g. in archaeology: 3D data) and the different ways of dealing with research data and their indexing (e.g. in the humanities: data can change permanently in the context of their indexing) need to be considered.

Policies of research funding agencies

Research funders are increasingly creating their own guidelines for handling research data. Examples of this are the European Commission [European Commission] and the German Research Foundation [Deutsche Forschungsgemeinschaft]. Applicants are requested to provide information on RDM, to draw up a data management plan (see in more detail in unit 6, p. 53) or to make research data generated in the project available under open licenses.

Since the existence and content of the data management plan can be included in the peer review, it is advisable for applicants to follow the guidelines closely and provide information as precisely as possible. Potential sanctions for non-compliance with the policy or the data management plan may include withholding the project budget or reduced chances for a follow-up grant proposal. However, audits after the end of the project are currently not being carried out across the board.

This text is an excerpt (p. 46 to 49) from Biernacka, Katarzyna, Bierwirth, Maik, Buchholz, Petra, Dolzycka, Dominika, Helbig, Kerstin, Neumann, Janna, Odebrecht, Carolin, Wiljes, Cord, & Wuttke, Ulrike. (2020). Train-the-Trainer Concept on Research Data Management (3.0). Zenodo..