Besides journal publications, an essential component of the research results of the Collaborative Research Center CRC 1456 “Mathematics of Experiment” consists of scientific datasets. These are collected and compiled within the subprojects of the CRC.
The CRC offers many opportunities for using these datasets for collaborative research. This allows exploiting synergies, gaining a wider interdisciplinary understanding of critical issues, and increasing the number and quality of publications. Moreover, the CRC – designed as a long-term project of international visibility and cooperation – requires careful data documentation, quality, storage, and accessibility. These aspects are subject of this policy.
This policy applies to all researchers working in the CRC: principal investigators (PIs), doctoral and postdoctoral researchers (irrespective of their individual funding source), student research assistants, as far as research activities are carried out under the CRC Program or data is used that was generated within the CRC.
§2 Data Management Committee
The Data Management Committee (DMC) approves data standards developed by the Infrastructure Project (INF) and respective subprojects, and serves to resolve potential disputes related to this policy. The DMC is composed of a representative appointed by the CRC board members (along with a substitute) and the PIs of INF.
- Each subproject is responsible for compliance with the agreed data standards.
- Each subproject names a contact person who can be addressed for questions on the subproject’s data such as:
- access rights
- access rights
- additional information, e.g. on how the data has been obtained
- All data is managed in accordance with DFG and University rules.
§4 Storage and documentation of data
- All datasets of the CRC that form the basis of a publication are registered with a unique reference in a central database: The data catalogue.
- INF is responsible for management, maintainance and support of the data catalogue and oversees compliance with this policy.
- All datasets as referred to in (§4a) have to be documented with meta-information (metadata). At least the metadata of each dataset will be stored in the data catalogue. The metadata includes at least the information on who collected the data, version of the data, date of the latest update, as well as any potential use restrictions.
- Each individual dataset requires an association to one responsible subproject and possibly more associated subprojects.
- Workflows for entering appropriate metadata for new records, their documentation and the definition of data standards is established in a joint effort of INF and the individual subprojects.
- Adding new records (i.e. datasets or files) needs to be done by or with the approval of the contact person of the responsible subproject. The decision on whether the metadata provided with a new record are sufficient remains with INF.
- Full data has to be archived according to the best practices of each field. For this purpose, INF administrates storage space for the CRC (CRC1456 dataverse) at GRO.data research data repository hosted by the eResearchAlliance in Göttingen. Full data comprise the elementary (raw) data, metadata, and all details required to understand the elementary data (e.g., description of data structure, methods used, documentation of the data-generating process, experimental design, etc.).
- All data must be provided under a license permitting free use for scientific purposes upon publication.
- All data should be provided in open formats as far as possible. Proprietary file formats should be avoided.
§5 Access to data
- All metadata in the data catalogue can be accessed publicly.
- Each CRC member should usually be granted access on request to all CRC datasets. The contact person from the respective subproject will be asked for consent prior to data sharing with other CRC members. All (non-trivial) data sharing is moderated by project INF.
- CRC members pledge that they will not pass any data or passwords that enable access to the data catalogue to third parties. Unil datasets are made publicly available (see point d and e of this paragraph), release of data to non-members of the CRC will require explicit permission from the responsible subproject or the DMC, which will consider the opinion of the responsible subproject.
- All data will be made publicly available as soon as possible, but at the latest one year after use within a publication.
- If there are legitimate reasons, the PIs of the subproject in which particular data were collected can apply to postpone the data publication procedures. Decisions will be made by the DMC.
§6 Use of data
- All rights on the data remain with the original data owner.
- CRC members making use of data supplied by other CRC researchers for scientific publication must acknowledge use of the data appropriately. Potential co-authorship has to follow the rules of good scientific practice. Before any work using data generated in the CRC is submitted for publication, the PIs of the subprojects that provided the data must be consulted.