"Research data management as part of academic teaching" – A forum of NFDI4Culture's Cultural Research Data Academy (CRDA)
Event dates: June 17 and 18, 2021
The entire report is also available for download as a pdf file here.
How important is the topic of research data management within academic education in the culture community? What are the different approaches to RDM education in the community, and what are the best practices in the individual disciplines? What are the challenges and problems in anchoring the topic in teaching and what contribution can NFDI4Culture make to reducing these obstacles? From which established, non-subject-specific offers and formats can the community learn? Does the culture community need a competency framework for RDM education?
This year's forum of the Cultural Research Data Academy, an interdisciplinary and decentralized institution within the NFDI4Culture consortium, was dedicated to this comprehensive set of questions. The academy is tasked with the bundling and development of subject-specific and demand-oriented training and further education opportunities in the field of data and code literacy.
The two-day digital workshop, which took place on June 17 and 18, 2021, was intended as a low-threshold event to initiate the broadest possible exchange within the community, which the CRDA would like to continue with all interested parties in the coming years. Participants from various areas of the culture community were invited to present their expertise in the field of RDM education, highlighting various offers and formats with which the topic of research data management has already found its way into teaching or is still to do so, and to discuss these with the audience.
Divided into four panels and a separate discussion round on the topic of a competence framework for RDM training, the workshop was first devoted to the experience gained so far in generic training programs, which can serve as an important basis for the culture community, before two panels took a look at interdisciplinary teaching programs. The presentations in the last thematic block focused on subject- and media-specific mediation approaches.
The first panel "Building on Experiences and Infrastructures" was opened by Dr. Elisabeth Böker, who described her extensive work in the Communication, Information and Media Center at the University of Konstanz as well as on the regional RDM-Initiative in Baden-Wuerttemberg, called bw2FDM, from which the information platform forschungsdaten.info also emerged, that has been in existence since 2019. With the help of this platform, not only state initiatives and designated sites are networked with one another, but terms and concepts of RDM are communicated in a general and coherent manner via topic pages and glossary. The focus lies especially on training offers: Open Science courses are not only linked on the platform, but likewise own webinars are organized. Böker particularly pointed out to query the level of knowledge of the participants, since even basic information on the rather special topic of RDM could be missing here. In addition, it would be useful to integrate practical examples regarding the legal aspects of RDM into training offers.
The second contribution was made by Patrick Helling from the Data Center for the Humanities (DCH) at the University of Cologne. He also introduced the work of the Cologne Center for eHumanities (CCeH), a teaching and research center founded in 2009 by the Faculty of Arts at the University of Cologne, which is now the coordinating body for Digital Humanities of the North Rhine-Westphalian Academy of Sciences and Arts. Helling spoke of the wide range of courses on offer, as well as the opportunity for students and staff to obtain IT certificates. Thus, there are exercises for the study programs of media informatics, information processing and linguistics, Studium Integrale and the graduate school, in which backup and archiving, publication as well as re-use, but more specifically also DMP, software and data curation are taught. Helling emphasized the subject specificity in the courses and also that more input from sub-disciplines would be desirable, which could be ensured by building up a network of experts.
Katarzyna Biernacka from the HEADT-Centre (Humboldt-Elsevier Advanced Data and Text Centre) at Humboldt University Berlin concluded the inputs of the first panel with an insight into the activities of the Sub-Working Group (UAG) Training/Continuing Education of the German Initiative for Network Information(DINI) and the competence network nestor as well as a discussion of the needs and challenges in continuing education.
Different training models were addressed and the close coordination with the Research Data Alliance Germany (RDA-DE) was emphasized. In the long term, there should be a certificate course on FDM, but its introduction is not yet foreseen. Biernacka emphasized the relation of the UAG Training/Continuing Education to the FDMentor Train-the-Trainer concept and underlined the different curricular integration. Likewise, the requirements for further qualification were emphasized, where the workshop format had proven itself (limited in time, focused on specific topics, interactive, practical and oriented to participants). For this purpose, not only the FDMentor train-the-trainer concept offers a full method case, but the Humboldt University also offers a large collection of materials on its hompeage.
The second panel then took up the topic of "Utilizing Synergies – Interdisciplinary Offers", which was to be continued on Friday.
Prof. Dr. Peter Niedermüller and Akademie-Jun. Prof. Dr. Stefanie Acquavella-Rauch from the Institute of Art History and Musicology at the Johannes Gutenberg University Mainz and the Academy of Sciences and Literature Mainz presented the research-oriented and cross-university program of the master's program "Digital Methodology in the Humanities and Cultural Studies," which started in 2016. "Digital Methodology in the Humanities and Cultural Studies" is organized and coordinated as a consecutive, research-oriented degree program by mainzed, the Mainz Center for Digitization in the Humanities and Cultural Studies, which in turn acts as a regional competence and consulting center in the Rhine-Main region. In addition to a basic teaching of core competencies in practice-oriented exercises and research-related block seminars, students from the cultural sciences and humanities as well as computer science choose focal points in elective modules and specializations: Text and Language, Spatial Reference, Music and Media. These cover, for example, digital music editing and music informatics, digital English linguistics, translation studies and translation technology, digital editing and modeling. A special feature is the obligatory project work of several weeks at a research institution. In the future, data literacy and 'teach the teacher' will also be emphasized in training programs for tutors.
Dr. Dennis Mischke, Research Coordinator of the Network for Digital Humanities at the University of Potsdam, gave in his next contribution an insight into the very broad range of courses offered at the University of Potsdam on the basis of the BMBF project Research | Learning – Digital (FoLD), which aims first and foremost to close gaps in DH education across institutes, to develop concepts for teaching and to implement them in individual courses of study: Introductions and courses in Digital Literary Studies thus introduce text mining and computational text technologies to improve data, digital, and code literacy. The program also includes hackathons, tutorials, and self-study units, which are primarily aimed at computational literacy and currently focus on the philologies, but are expected to spread to other subjects in the near future.
The panel was closed by Julia Röttgermann with an input introducing the work of the Digital Humanities professorship in Trier. Prof. Dr. Christof Schöch joined the panel for the discussion. This professorship coordinates the Master of Science Digital Humanities program at the intersection of the humanities and computer science. Röttgermann gave a rich overview of the offerings of the Trier Center for Digital Humanities. In this, application-oriented content and teaching units (Open Science, XML formats, GitHub, FAIR, Wikiversum...) are integrated and taught in the presentation of methods and practice tools with a focus on Digital Literary Studies. Finally, in the discussion, it was equally pointed out that this teaching toolbox was not a fixed selection, but an evolving curriculum, which likewise opened it up to student interests beyond large collections and texts.
The third panel continued the complex " Utilizing Synergies – Interdisciplinary Offers". Prof. Dr. Vivien Petras spoke about the master's degree program in "Digital Data Management," a joint program offered by Humboldt University Berlin and the Potsdam University of Applied Sciences. The program teaches data literacy, i.e. competencies in (research) data from the fields of science, culture and business. Its modules range from the framework of data management to technologies and agile project management. The program is based on the EDISON Data Science and FAIR4S frameworks. In contrast to offerings that complement courses of study, a complete course of study has been developed here. However, it is not only aimed at regular students, but is also open to interested parties. They can attend individual modules as continuing education courses. Accredited only at the beginning of 2020, the course providers can already draw a positive balance.
Dr. Andreas Hütig and Lennart Linde presented the integration of the course offerings on data literacy into the Studium generale of the Johannes Gutenberg University. The central feature of the Studium generale is interdisciplinarity, which at JGU not only refers to the students participating in the courses, but also the project group is interdisciplinary. Area-specific needs and assessments have been identified for four subject groups (STEM, humanities/cultural studies, social sciences, medicine/life sciences). Core competencies in dealing with data are to be acquired. This is done in four steps: the development of a research question, the selection and evaluation of suitable data sets, their independent analysis and the final documentation. To solve the tasks, students are introduced to common tools, among other things. Looking back at completed courses, the presenters noted that common problems from Studium generale, such as the different entry levels of students, pose some challenges in course design. In general, however, the course is well received and the aim is to make it a permanent part of the Studium generale at JGU.
Following the interdisciplinary offers of the previous panels, the final round of presentations was devoted to discipline-specific offers on RDM.
Prof. Dr. Christoph Seibert is professor of music infomatics at the Karlsruhe University of Music. He talked about data and code literacy in the courses of musicology and music informatics. Since 2004, the two subjects have been thought of more closely together at the HfM and meaningful mutual additions of musicological and musicinformatics study content have been realized. In addition to the basics of programming in Python, research ethics, statistics and data management play an important role against the background of the Digital Humanities. Students can study both subjects equally alongside each other in the Bachelor's degree or choose one of the two as a complementary subject. This possible combination brings both opportunities and challenges; it has already been observed that trained competencies in music informatics rarely find their way into musicology, where they expand a separate field. In the future, the topic of research data management should also be more concretely planned into the training of students.
Prof. Dr. Tobias Matzner then reported on the experiences and approaches of the Institute for Media Studies at the University of Paderborn. Students are largely aware of what the data they deal with in their studies is about and which tools they should use to process it. Nevertheless, there is a great need to teach data literacy, especially the area of data ethics and the recognition and identification of algorithmic biases. Matzner also addressed the social dimension of data literacy and thus, among other things, the aspect that every citizen needs a certain level of data literacy; in the area of social media and their algorithms alone, a basic level of data literacy is required in order to be able to deal correctly with these media.
A theater studies perspective was brought to the event by Dr. Nora Probst, head of the Digital Humanities Department of the Theater Studies Collection (TWS) at the University of Cologne. At TWS, students are taught primarily good research data management in the area of data literacy. Theater Studies collection data are very diverse in their materiality and media; especially in collection acquisition and curation, aspects such as the correct handling of metadata, creating the longest possible reusability and interoperability of datasets are of great importance. However, it is not just a matter of storing the data correctly, but above all of interpreting data sets. In addition, data ethics is also an important element of training in Cologne, and data criticism is a teaching concept.
Dr. Harald Klinke concluded the panel with a presentation on subject- and media-specific offerings in art historical research and in the field of digital art history. In addition to teaching common tools, repositories, and metadata standards, Klinke highlighted creating your own datasets as an important teaching method. Creating, visualizing, and later analyzing datasets is a method that can show students the benefits and problems of digital data analysis in a concrete way. For art history, it is also important to be able to work with concrete and sufficiently large data sets, and so close contact with GLAM facilities and university collections is necessary. In conclusion, Klinke saw the integration of research data management as a digital competence in university education as a future challenge.
Another element of the program was a discussion panel with all participants, which was dedicated to the question "Do we need a competency framework for RDM education in the Culture Community?".
The expectation for the reactions of the forum participants to this pointed question was naturally more than a simple "yes" or "no". Rather, in light of the NFDI's goal of initiating open, user-centered, and science-driven collaboration on research data, the focus here was also to be on the experiences and needs of the community.
In the opinion of NFDI4Culture, the teaching of data and code literacy, and thus also of aspects of research data management, should not just begin at the level of doctoral students, but should already be addressed in BA and MA education. NFDI4Culture sees itself as a networking and communication platform for the culture community and would like to moderate the coordination process for the creation of a competence framework tailored to the community.
In the first of two polls on the title of the panel, quite different positions emerged, clustered around the following areas:
- Orientation (also for own courses)
- Serves to clarify basic questions and to agree on basic competencies
- Avoid duplication of development of content
- Official consolidation/recommendation of content
- Standards are hugely important
- RDM needs to be strengthened in the community
- A competence framework provides guidance for smaller institutions
- Provides transparency for students as well
- A basic RDM education belongs in all degree programs (at the latest at master's level)
- There are already enough competency frameworks, plus specific, very different requirements
- Implementation could be difficult
- (Undesirable) standardization/bureaucracy/too much regulation and standardization
- freedom of teaching
- lack of acceptance
- A competence framework does not do justice to the heterogeneity of the contents and misses the actual needs
- Is it worth the effort?
These points of view were reflected in the constructive discussion. In addition, it was critically noted, among other things, that parallel developments should be avoided, that establishment must be ensured, that creative thinking must not be suppressed by a competence framework, that competence frameworks must be constantly reviewed, that orientation towards the FAIR principles must consider their "half-life" and that (problem) awareness of RDM must still be created among many colleagues.
Issues of data ethics and data law, as well as best practices and sharing of experiences (including nationwide) are seen as important for the CRDA to mediate and "moderate"; without the idea of exclusivity, the planned NFDI4Culture Forums and Workshop formats should be suitable and flexible enough for such moderation. The question of the status of competence frameworks in other NFDI consortia also came up, and was answered with reference to the cross-cutting "Training & Education" team that is being established; among other things, NFDI4Culture could be a pioneer in a cross-cutting area between degree programs on RDM.
The creation of platforms from which peer-reviewed materials can be accessed and through which data publications can take place is seen as an opportunity of competence frameworks. In addition, a competence framework must be oriented to the data specificity and to the different competencies; the prerequisites vary greatly – for example, between colleagues from the DH fields and those from the sometimes still largely analogous humanities. This difference is based, among other things, on the fact that the topic of RDM is hardly perceived outside of DH. Changing this should be a central issue, free from the discussion: RDM is good scientific practice, therefore it cannot only be relevant for DH; one should try to get everyone on board and convince the critics or analog colleagues with different strategies (e.g..: "They are already generating research data, they just need to do it properly ...", "Even the Internet is a big device ...").
Basically, data literacy is well presented on Wikipedia, likewise there are already some competence frameworks at universities. Agreeing on common standards is a challenge, stakeholders should be guided by the data lifecycle (create, provide, evaluate, visualize, interpret data), among others, but also by questions like: Who needs to be able to do what? What do we expect from infrastructure staff and scientists in the future? Who makes which RDM offers, how much competence do scientists need to critically evaluate them?
In the second survey, a word cloud (see above) was used to collect key issues that a future competency framework for RDM education in the community might consider. Some aspects that had already been raised reappear at this point, but have been weighted here along with the supplemented terms.
This year's Cultural Research Data Academy forum offered spotlights on the current status of university education in research data management in the culture community. The inputs and discussion contributions reflected the well-known complexity of the topic, but also presented different approaches to solutions for a stronger anchoring of FDM in the education of students.
The contributions and inputs from a wide range of institutions, initiatives and individuals from the field of higher education showed that there are already numerous ideas, concepts and teaching offers, generic as well as discipline-specific. Especially in the conversations and discussions between the presentations, it became clear how fruitful and profitable a closer networking of the participants and other actors in this field can be. In addition to the exchange about best practices and hurdles in the development of services, the participants expressed their desire for further cooperation and collaboration in the development of services.
The CRDA would like to act as a communicator and link, as a stimulator for networking. We do not want to break off contacts that have been made and conversations that have been started, and we offer to facilitate contacts and to create virtual spaces for joint exchange in the future. We would like to continue to expand the network in future forums, involve actors in the cultural sector and discuss potential for improvement in the area of FDM and training. If you are interested in an exchange with us or the input providers of the forum, please contact the team of the Cultural Research Data Academy via the contact form on our website.