Analysis of RDM counselling protocols - A standardized model of categories

Various service centers advise and support researchers in research data management (RDM). This includes helpdesks of NFDI consortia, state initiatives, and individual institutions. Despite the large variety between support structures, the topics of consulting are often similar [1]. This allows to standardize the documentation of RDM consultations. Efforts have already been made to standardize the syntax of such documentation [2,3], although most templates are only used internally. However, so far there is no solution to standardize the semantics of the documentation. At the same time, there is considerable potential in using structured information about RDM consultations for the analysis and planning of RDM support services and collaboration between RDM service centers. Thus, we developed a German-language model of categories for the documentation and analysis of RDM consultations [4,5], which consists of two levels of categories – 31 main categories and 94 sub-categories. The model was subsequently evaluated by six experts in RDM counseling services [4]. These experts categorized topics in their own RDM counseling protocols, applying the model by means of a piloting implementation. Afterwards, they took part in interviews (for the interview guide, see [6]) about their experiences with the model and its potential applications. The interviews showed that the category model indeed comprehensively covers RDM topics, although an optional third level with discipline- or institution-specific categories would be valuable. The model was considered sufficiently simple to be used in day-to-day documentation, if implemented in a low-threshold way which does not require to abandon current documentation practices (e.g., in ticketing systems) altogether. Most importantly, the participants saw potential for diverse applications, including the planning of consultation services, the preparation of consultation meetings, and quality-assurance in RDM consulting. Beyond these local applications, the model can also be used for collaboration between NFDI helpdesks. The categories do not contain sensitive information about individual consulting requests and can thus be shared between helpdesks. This offers great potential in finding experts at other helpdesks, as well as sparking in-depth exchange about specific requests. Thus, such a model, if developed further with the community and implemented in a low-threshold way, can create substantial value for support structures by providing a basis for local improvement of consultations as well as for collaboration. This is particularly true for NFDI helpdesks, which typically consist of several colleagues, have a high level of professionalization, and at the same time have a strong disciplinary focus, which requires regular interface work with other helpdesks. Thus, we see the NFDI as ideally suited to implement such a model. RDMO would be a suitable tool for future implementation, and we have already secured interest from multiple stakeholders. We plan to continue the development and implementation of the model, and hope to raise awareness for it, as well as extend the group of stakeholders interested in working on it with us.