10 11 2020 ws data quality

10 11 2020 ws data quality
November 10, 2020

BonaRes Data Centre hosts virtual workshop on data quality

In its guidelines, the DFG recommends that "for reasons of traceability, connectivity of research and reusability, scientists and scholars should, whenever possible, store the research data and central materials on which the publication is based - following the FAIR principles ("Findable, Accessible, Interoperable, Re-Usable") - in recognised archives and repositories in an accessible manner". The BonaRes Data Centre also aims to increase the quality of the research data in the BonaRes repository for re-use. For this purpose we will gradually integrate standardized quality assurance (QA) tools into our workflow over the next years. In a first workshop with 24 participants on November 3rd, 2020, the wishes and requirements of typical data providers and subsequent users were identified and possible international cooperations were initiated. Seven Module A projects, three of them Rhizo4Bio projects, presented typical data sets and existing methods for quality assurance from their research work in short presentations and formulated requirements for the Data Centre. Five presentations by external experts presented ideas on data preparation, selection criteria, dealing with uncertainties in geodata and quality assurance measures of other repository operators. In close cooperation with data providers and users, the BonaRes Data Centre plans to develop a "DQ Kit" and to make it available to data providers. This kit will help them to perform various tests, e.g. tests for outliers or data plausibility, in a time-saving and clear manner before the upload process. The results should then quickly provide information to the data provider to identify errors in the data set if necessary. When uploading data, the test results should be incorporated into the metadata as automatically as possible. This will provide subsequent users with valuable but value-free information on data sets, more comprehensive than today.