Who "owns" research data?

Introduction

The collection or generation of research data is usually associated with great financial, personnel and time expenditure. But who actually "owns" this research data? This question needs to be clarified to "Who has what rights to research data?", as there is no ownership of digital research data. In addition to legal requirements, research ethics principles such as the DFG Code of Good Scientific Practice or the CARE Principles are also relevant for researchers.

When it comes to questions, such as "Am I allowed to publish the data at all?" or taking data with me when I move to another institution ( Section 4 in Wünsche et al., 2022), legal clarifications often become complex and difficult for laypersons. Individual cases have to be examined and different legal positions are often interrelated. Therefore, judgements must be made in the sense of a balanced overall view of the regulations described below.

However, it is not only the relationships between researchers and between researchers and their employers that are important to consider, but also the legal rights and entitlements of the people or groups being researched. This is why we also briefly refer to data protection and the CARE principles.

This text serves to outline various (legal) positions and thus contribute to their fundamental understanding. This understanding is important, for example, in order to create documented agreements on the usage rights to research data on an informed basis.

Documented agreements on the rights of use

In order to avoid certain conflicts over data, it is recommended to discuss responsibilities and expectations at an early stage and to document agreements on rights of use in a written form. In most German research institutions, this is even mandatory in order to comply with good scientific practice, as the requirements of the DFG Code (German Research Foundation 2022) are implemented in a legally binding manner by almost all institutions in Germany (see also Section 3). "Guideline 10: Legal and ethical framework conditions, rights of use" of the Code states that:

“Where possible and practicable, researchers conclude documented agreements on usage rights at the earliest possible point in a research project. Documented agreements are especially useful when multiple academic and/or non-academic institutions are involved in a research project or when it is likely that a researcher will move to a different institution and continue using the data he or she generated for his or her own research purposes.”

Agreements on the rights of use of data should be concluded in the sense of a balance of interests and fairness towards all parties involved.

Protection under copyright law

Unlike books or journal articles, for example, which are generally protected by copyright, research data is only protected by copyright in some cases. This distinction is relevant because only in the case of data that is protected by copyright is the right to publication and attribution reserved for the copyright holder. However, it should be noted that, for example, the right of publication may be restricted by other legal positions, see employment or service relationships, or that the obligation to give attribution through good scientific practice also applies to data in the public domain. Only in the case of data protected by copyright can copyright holders grant permission for use by third parties, e.g. with an open licence. The granting of an open licence for copyright-free (= public domain) data has no legal effect.

In many cases, it is easy to distinguish whether copyright protection applies or not. The next two sections provide information on this.

Research data protected by copyright

The requirements for research data to obtain copyright protection are low. It should be sufficient that the person producing the data has made an intellectual contribution that manifests itself in a concrete creation that shows a minimum degree of own contribution. Linguistic works, pictorial works (including photographic works), cinematographic works, computer programmes, drawings, plans, maps and sketches are generally protected by copyright. However, the prerequisite for this is that these work must be personal intellectual creations - in this context, it is also said that a certain "level of creation" must be reached.

Images and film recordings that do not reach a certain level of creativity may be covered by ancillary copyright, which has similar effects to copyright but different requirements. For example, it is conceivable that images taken by a wildlife camera could enjoy ancillary copyright, even if none of these images contain any intellectual contribution by the photographer, because this person did not even exist. In the case of qualitative research data, such as data from participatory observation in ethnology or an interview, the necessary level of creation is usually reached, so that these are also protected by copyright.

There are also designs, texts (e.g. technical instructions), sound sequences or computer programmes that do not enjoy copyright protection, but these are the exception. Similarly, measurement data is generally not protected by copyright (see next section). In cases of doubt, individual examination is required; in cases of uncertainty, it is safer to assume that protection exists. In any case, the human and/or financial resources used for data collection are not relevant for the assessment.

Research data not protected by copyright

Data itself is free of any rights. Data from experiments or tests are not protected by copyright, nor are measured values of element concentrations in rocks or remote sensing data from satellites. The cost of generating the data or the intended use of the data is irrelevant. It follows that even measurement data from a very complex and expensive specialised device, the collection of which required a large investment, does not enjoy copyright protection. Even data that has been further processed according to aspects based on specialised scientific practices is not regarded as personal intellectual creations. The presentation of measurements of stable isotopes can serve as an example here: these measurements are usually given as δ-values, the calculation of which is based on scientific practice.

Tables/ Databases

According to the definition of copyright law, tables in which data is organised in certain ways belong to databases. They often enjoy copyright protection because the creative achievement of organising the data in a table in a certain way is protected. Exceptions are tables in which this creative achievement is not recognisable (e.g. if data is only arranged according to alphabetical or chronological aspects, with the aim of completeness or according to other simple aspects based on scientific practice).

In addition, there is a special protection of the producer of the database if the "acquisition, verification or presentation" of the collected data requires a "substantial investment by nature or amount".

An example: a table that contains hourly temperature readings for a month in chronological order is not protected by copyright. However, if only a few selected readings from this measurement series were presented in a table due to a specific research question, this individual selection decision could possibly generate copyright protection for this table. The following also applies: the database creator's copyright (see above) prohibits the reproduction or distribution of significant parts of the database.

It is important to distinguish between the copyright protection of the table itself and the possible copyright protection of the data in the table: if a copyright-protected table contains, for example, measured values, the individual measured values are not protected by copyright (as long as they are not copied to such an essential extent that the structure of the database and thus the intellectual achievement behind the database work becomes apparent from the measured values). However, if a table contains, for example, answers from an interview study, the data (in this case textual data) is also protected by copyright.

Codes of good scientific practice

The DFG's Code of Good Scientific Practice (GSP) has been implemented by publicly funded universities and other research institutions in Germany in a legally binding manner, for example through GSP statutes, and therefore applies to most researchers in Germany. However, the institutional GSP statutes may deviate to a certain extent from the DFG regulations, so that in individual cases it is important to always refer to the relevant, legally binding statutes as a basis for individual assessments.

Guidelines 10 and 14 of the DFG Code deal in particular with questions of authorship/authorship of research data, questions of authorisation to decide on publication and modalities of subsequent use by third parties. According to Guideline 14 of the Code, the authorship of research data is subject to the same principles as the authorship of text publications, e.g. articles in scientific journals. According to the explanatory notes to Guideline 10, the use of research data "belongs in particular to the researcher who collects it". This should not refer to those who technically collect the data, but to those who are responsible for the scientific conception of the data collection (Baumann et al. 2021; S 64). This group of people is also entitled to determine whether third parties are granted rights of use, although it remains unclear whether other researchers in working groups or in a division of labour approach to data collection would also have rights of use. The possibility of transferring data in the event of a change of institution is also not specifically defined in the Code.

In order to avoid possible uncertainties in day-to-day research, Guideline 10 recommends individual, documented agreements on the rights of use of research data and research results. This important instrument is discussed in Chapter 1, section Documented agreements on rights of use.

Employment relationships and status groups

In employment relationships where copyright-protected work results are produced, the copyright does not belong to the employer but rather to the individual who created the work. However, depending on the status of a person's affiliation to a university, the employer may have a right of use to the work result (here: research data), so that any existing copyrights may be overlaid. The distinction is made based on whether a person conducts research freely and independently or is subject to the employer's authority to issue instructions.

University professors are considered to conduct research freely and independently, as well as likely employees who are working on a doctoral or habilitation project that is not part of a larger research project.

For other academic staff at universities, the distinction between work that is free from directives and work that is subject to directives is often challenging in practice and can lead to practical problems. In the case of researchers who are bound by instructions, problems can arise, for example, with respect to the right to publish, as both parties, the researcher and the employer, have legal interests. It is also difficult to make a clear distinction between work not subject to instructions and work subject to instructions in the case of doctoral projects that are carried out on the same data basis as research work as part of a third-party funded project. A balance of interests must then be sought in each specific case.

Furthermore, even in work subject to directives, there is often room to methodologically influence the research results. Especially in larger projects, the creative contribution of employees is often crucial for scientifically sound results.

Conflicts can also arise with non-copyright-protected research data, for example, when deciding whether a person leaving the university has the right to continue using the research data they collected. Depending on the specific case, both parties may potentially have rights to use the data.

Although enrolled students are generally not in an employee of the university, they are members of the university. Especially for advanced students, an independent choice of methods and search for knowledge can be assumed, so that the rights of use to research data generated in the context of qualification work can certainly be assigned to students (Baumann 2023, S. 28).

For employees at non-university research institutions (nuR), by contrast, it tends to be assumed that research data generated by them (regardless of whether the data is eligible for copyright protection) are so-called "compulsory works" and that the rights of use therefore lie with the employer.

To prevent conflicts that may arise due to often difficult-to-assess legal situations, the solution suggested at this point explicitly refers to the Code of Good Scientific Practice: "Researchers, where possible and reasonable, make documented agreements on usage rights at the earliest possible stage of the research project" (Guideline 10). These documented agreements should be based on a fair balance of interests.

Rights of researched people or groups: Data protection and CARE principles

Data protection

Data protection refers to the rights of research subjects, i.e. the protection of people's privacy, because their personal data is subject to special protection. If personal data is collected or processed in research projects, the provisions of data protection law must therefore be observed. The applicable regulations are set out in the EU General Data Protection Regulation (GDPR) and in federal and state data protection laws.

Fundamental elements of data protection are, for example, the consent of the data subject(s) to the processing of their personal data, regulations on the further processing of personal data and the rights of the research subjects to information, access, rectification of inaccurate data and erasure of data.

Practical advice on how to proceed in research projects can be found, for example, in the handout: Handout on data protection from RatSWD (Rat für Sozial- und Wirtschaftsdaten (RatSWD) 2020).

CARE Principles

The current movement for open data and open science does not fully recognise the rights and interests of indigenous peoples. Existing principles, such as the FAIR Principles, are primarily designed to facilitate data sharing between institutions, while ignoring power imbalances and historical contexts. In this tension for indigenous groups and peoples, they demand greater control over the application and use of indigenous data and knowledge for collective benefit.

The CARE Principles describe how data should be handled to ensure that indigenous decision-making power over the data and its use is respected. This includes the right to create value from indigenous data that is grounded in the indigenous worldview and the right to utilise the opportunities of the knowledge economy. The CARE principles complement the existing FAIR principles.

CARE is an acronym for: * Collective Benefit * Authority to Control * Responsibility * Ethics

Explanations and notes on the practical application of the CARE principles in ecological and biodiversity research: (Carroll et al. 2020; Jennings et al. 2023)

References:

Baumann, Paul (2023): Rechte an Forschungsdaten. Trier University.

Baumann, Paul; Krahn, Philipp; Lauber-Rönsberg, Anne (2021): Forschungsdatenmanagement und Recht. Datenschutz-, Urheber- und Vertragsrecht. Feldkirch/Düns: Wolfgang Neugebauer Verlag (Arbeitshefte der Arbeitsgemeinschaft für juristisches Bibliotheks- und Dokumentationswesen, Band 28).

Carroll, Stephanie Russo; Garba, Ibrahim; Figueroa-Rodríguez, Oscar L.; Holbrook, Jarita; Lovett, Raymond; Materechera, Simeon et al. (2020): The CARE Principles for Indigenous Data Governance. In: Data Science Journal 19, Artikel 43. DOI: 10.5334/dsj-2020-043.

Deutsche Forschungsgemeinschaft (2022): Guidelines for Safeguarding Good Research Practice. Code of Conduct. DOI: 10.5281/zenodo.6472827.

Jennings, Lydia; Anderson, Talia; Martinez, Andrew; Sterling, Rogena; Chavez, Dominique David; Garba, Ibrahim et al. (2023): Applying the 'CARE Principles for Indigenous Data Governance' to ecology and biodiversity research. In: Nature ecology & evolution 7 (10), S. 1547–1551. DOI: 10.1038/s41559-023-02161-2.

Kuschel, Linda (2018): Wem "gehören" Forschungsdaten? 9, S. 764–766. Online verfügbar unter https://www.forschung-und-lehre.de/forschung/wem-gehoeren-forschungsdaten-1013, last check 10.03.2024.

Rat für Sozial- und Wirtschaftsdaten (RatSWD) (2020): Handreichung Datenschutz, 2. überarbeitete Auflage.

Wünsche, Stephan; Soßna, Volker; Kreitlow, Vanessa; Voigt, Pia (2022): Urheberrechte an Forschungsdaten – Typische Unsicherheiten und wie man sie vermindern könnte. DOI: 10.17192/bfdm.2022.1.8369.