RDM: VU FSS guidelines for data management
Introduction
As a faculty committed to excellence in the quality of the research our staff and students undertake, it is essential to have guidelines on good practice in Research Data Management (RDM) as part of our framework to support the integrity of our research. These FSS guidelines serve both as a means of developing and supporting a culture of good practice in data management and demonstrating that we are committed to a culture and environment where high standards are encouraged and expected.
The purpose of these guidelines is to reduce work pressure by condensing the various, and often conflicting, regulations into one cohesive set of procedures that ensure maximum compliance. Details on the various underlying policies, and how they’re reflected in these guidelines, can be found here.
Stepping stones for good data management
Below are the guidelines for research data management in various stages of the research life cycle. In these guidelines, the following verbal forms are used:
- “shall”, “are required to” and “must” indicate a requirement;
- “should” indicates a recommendation;
- “may” indicates a permission;
- “can” indicates a possibility or a capability
Planning & Design Phase
- Researchers must follow the ethics review procedure of the Research Ethics Review Committee (RERC).
- Researchers must write a Data Management plan (DMP) using DMP Online, so that they can easily provide an up-to-date version to their department head at any moment.
- Researchers shall ensure that all planned activities with personal data comply with GDPR. In particular:
- They must plan to take appropriate technical and organization measures to secure data. Because of the wide variety of data used in the faculty, there is not one answer as to what measures are appropriate. Researchers should discuss the measures they take with colleagues, department heads, with the faculty data steward, privacy champion and/or the RDM support desk.
- They should ensure that all personal data is processed with full consent of all data subjects. If consent cannot be obtained, the researcher must ensure that there is another legal ground for processing the data. A privacy champion can assist with this.
- They shall ensure that if personal data is handled by third parties, the proper agreements are in place to do this securely, for example Data Processing Agreements. A privacy champion can assist with this.
- They shall ensure that all data processing activities (collection, analysis, publishing, archiving, etc.) are entered in the VU’s central data processing registry. Currently, DMPs created using the VU template in DMP Online are linked automatically to this registry, meaning the researcher does not need to take additional action for this.
- Contracts and agreements relating to the commissioning, funding and conduct of research, including data sharing, intellectual property rights, collaboration and non-disclosure agreements must all be processed through IXA-GO to ensure the safeguarding of (the autonomy of) your research. Such contracts must be signed by those with the appropriate delegated authority to do so on behalf of the University. The signature process is a chain of responsibility that starts with the submission from the Researcher for approval of the Head of Department, before the final signature from the Managing Director of the faculty or a member of the Executive Committee of the University. The Data Steward can advise researchers on how to manage this process.
During ongoing projects
- Researchers shall keep their DMP up to date.
- Researchers should ensure that their data is stored in such a way that it can later be archived in accordance these guidelines (see section ‘after’, below) without excessive effort. This includes:
- Ensuring data is well-organized (for more information, see the page on data organization);
- Data is stored in the same place as vital documentation. Depending on the discipline of the researcher, this can include interviewer guides, questionnaires, topic lists, sampling information, power calculations, etc.
- Making sure the data is accompanied by a basic “Readme File” containing basic metadata such as an explanation of the purpose of the data, who is responsible for collecting it, and how the folder is organized, etc.
- Researchers must ensure that data is reliably, traceably and securely stored throughout the research life cycle. The VU offers storage infrastructure that meets these requirements (see the storage finder). If project data needs to be stored elsewhere (for example with project partners), researchers must ensure that the storage solution chosen meets these requirements. The data steward will help in this assessment.
- Researchers must take appropriate technical and organizational measures to secure any personal data.
- They can store directly identifiable data (see definitions, below) separately from other data, either by storing it on a different server or device, or through encryption.
- They must not store directly identifiable data longer than needed. Note that it may be impossible to remove directly identifying data without editing the raw data, which would compromise data integrity and provenance. In such cases, the directly identifiable data may be stored as long as long as the rest of the raw data. Researchers are expected to decide what data to destroy what data to keep. Researchers should discuss the choices they make with colleagues, department heads, with the faculty data steward, privacy champion and/or the RDM support desk, and record these choices in their DMP.
After publication of results
Archiving and Registration
Researchers must ensure that the underlying data for each published empirical study (article, volume, book chapter, PhD thesis chapter, Research Master’s thesis, consultable internal report, etc.) is archived according to the following:
- What: all data that can be reasonably deemed necessary to verify the findings of the research. This includes the raw data (or a link to it, if secondary data was used), the data that was analysed and a description of all modifications to obtain the analysed data from the raw data (or the computer code used to perform these modifications) and full documentation of all steps involved in acquiring, processing and analysing the data. Detailed advice on what to archive can be found on the page on Data Archiving.
- When: the data must be archived no later than one month after the publication date, and be available until at least 10 years after the publication date. If this is not possible, a justification for deviating from this should be provided in the DMP.
- Where: a secure and reliable location that is accessible for verification (see the section on verification below), and that provides a persistent identifier. The archiving options provided by the VU satisfy these criteria. If data needs to be archived elsewhere (for example with project partners), researchers must ensure that the storage solution chosen meets these requirements. FSS follows the ERC’s approach “as open as possible, as closed as necessary”. In practice, this means that public data is preferable, but that personal data does not need to be published1. Should researchers want to publish such data, they should ensure that they meet all legal and ethical requirements to do so, consulting with the faculty data steward if needed. Public data must always be accompanied by a license and, in case of personal data, information about the informed consent procedure. The decision to publish data or not should be explained in the DMP.
- Who: the first author of the publication is responsible for archiving the data. Second or later authors must know that the data have been carefully stored and how this has been arranged. This is particularly important if the first author does not work at FSS. For PhD candidates and research master’s students, the primary supervisor or the day-today supervisor respectively are responsible for archiving, but can delegate the work to the PhD candidate.
Furthermore:
- Researchers should include in their published empirical studies a data statement containing the repository where the data is archived, the persistent identifier of the data, and instructions on how this data can be accessed and for what reasons. For sensitive data that is not published and that can only be accessed for verification purposes, a persistent email address may be provided where questions regarding the data can be directed.
- Researchers should ensure that all datasets that they produce are registered on the VU’s Research Portal, including sufficiently descriptive metadata, and the persistent identifier of the data set.
Data verification
In case of doubts about the research integrity of FSS research, the faculty board can decide that verification of archived (non-public data) is needed. In making this decision, the board shall balance the need for confidentiality and security with the interests of transparency. If it is decided that the data needs to be reviewed, the Faculty Board will then decide who will access the data while ensuring confidentiality of the data and work with IT, Legal and the Data Steward to ensure that this access is possible.
Administrative procedures
End of employment
If a researcher leaves the VU, the department head should work with the researcher to ensure:
- That the data for any ongoing projects is properly stored according to these guidelines;
- That the data for any pending publications is properly archived according to these guidelines;
- That responsibility for any data sets archived by the researcher on VU infrastructure is transferred to an FSS colleague; and,
- That the researcher doesn’t lose access to data they need for their further career, if such access can be reasonably organized; for example through the signing of a data transfer agreement between the VU and the researcher’s new institution.
To whom do these guidelines apply?
These guidelines apply to all faculty staff members who conduct research in the context of a temporary or permanent employment contract, all PhD candidates who conduct research under the supervision of a professor, and all research master’s students. The guidelines do not apply to bachelor’s and one-year master’s students, unless their research results in an academic publication. Research conducted by bachelor’s and one-year master’s students falls under the formal responsibility of their supervisors.