(Only) Quality Qualifies: Quality Assurance in the Research Data Center Network

This contribution provides a concise overview of how the quality of data and consultation services provided by the accredited Research Data Centres is guaranteed. We focus on three steps: 1) Accreditation and quality assurance by RatSWD, 2) extensive documentation and 3) competent expert consultation. The overview provided here can also serve as a best practice template by describing bundles of quality assurance mechanisms which can be adopted in other contexts by data providers in other disciplines. (Research) Data is provided by numerous institutions in Germany. However, not every data provider adheres to the same quality standards. For this reason, the German Data Forum (RatSWD) awards accreditation to Research Data Centers (RDCs) with well-defined quality standards providing access to sensitive data in the Social Sciences. The management of sensitive research data distinguishes RDCs accredited by RatSWD and provides a basic but fundamental layer of quality assurance as every RDC has to adhere to a well-defined set of quality criteria [1]. Quality assurance for accredited RDCs entails an annual monitoring, and an independent complaints process. By mandating transparent access procedures and comprehensive user support, this ensures that data are readily available and usable for researchers, promoting extensive use and validation. Detailed documentation and proper data preparation guarantee that the data are reliable and can be effectively used for scientific analysis, helping researchers understand the context, methodology, and limitations of the data, which is crucial for producing valid and reproducible results. More uniquely, however, RDCs ensure that the data they provide are of high substantive quality, i.e. help answer substantive research questions. Most importantly, RDC go beyond merely providing an infrastructure for data access, rather they provide consultations either in person or by providing substantial documentation, such as codebooks or data description reports. As RDCs are diverse in the various data types that they offer, ranging from qualitative to quantitative data there is the additional challenge that this data is not generated with research in mind and might not always perfectly fit a research question, making the expertise and consultation of RDC even more valuable. In addition, RDCs go to great lengths to ensure the technical quality of their data and to promote FAIR data. For example metadata are provided and datasets are registered with a DataCite affiliated DOI provider. Especially metadata want to be reused. RDCs facilitate this by increasingly adopting standardized metadata formats, such as the DDI family, and exploring ways to publish metadata in standardized ways e.g. enabling metadata harvesting. RDCs are actively involved in initiatives such as persistent identifiers (PIDs) for variables, FAIR signposting, metadata adoption metrics, and the development of the Open Data Format [4].