there are several guiding #principles for #openscience – and for scientific responsibility! #OSDMA25

an important part of good research practice is data ethical responsibility, and also for this there are several guidelines.
somethies there are higher ethics risks, e.g. when certain personal data is collected, if children or vulnerable people are concerned, or of data processing techniques invasive or vulnerable to misuse.
#OpenScience #DataEthics #OSDMA25

why do we need #dataprotection – generally refering to the "protection of personal data" – in research? it is important so that data misuse can be avoided. #OSDMA25

do you work with anonymised, with pseudonymised, or with anonymous data? the last one does not fall under the GDPR, but the other two do! #dataprotection #OSDMA25

if #GDPR has to be applied to personal data in research, there are several principles to follow, many of them we already know from the UNESCO recommendations. following the research data lifecycle Vasilka shows how to comply with them. #OSDMA25 #RDM #DataProtection

the aspects regarding the storing of data also applies to non-personal data, as do other aspects as well. and of course the FAIR principles apply. important here: FAIR does not mean "open to all" automatically. but if your data should be open, it should be FAIR!
#GDPR #ResearchData #OSDMA25

for F, A, I, and R there are several #DataProtection measures available. #OpenScience #RDM #FAIRprinciples #OSDMA25

Vasilka has collected the main takeaways on #research #responsibility & #ethics + #GDPR on two slides. but #complience does not end here: what about intellectual property, copyright law, contractual obligations, ...? so think about this before sharing any material! ☝️ #OSDMA25

rounding up her interesting talk Vasilka provides us with a tool named "iVA", an "interactive Virtual Assistant" (https://www.berd-nfdi.de/legal-questions/) and more ressources about #DataProtection et al. 🙏 #OSDMA25
if you would like to contact Vasilka: https://www.bib.uni-mannheim.de/ihre-ub/ansprechpersonen/vasilka-paunova/

from the discussion:
does GDPR also apply to data from outside the european union? that depends e.g. on where the data is collected and processed and the anonymity of the data.
how to handle psychological data from children and adolescents? anonymize them as much as possible.
most AI companies are based in the US., and many of them don't care about GDPR, the AI act et al. what can one do about it when sharing data? raise awareness, check the anonymity of the data, restrict sharing.
#OSDMA25

and one more question: are there self-assessment tools available for this topic of data protection? not as far as Vasilka knows, and it is difficult to pack every aspect of it in a check-list. she recommends attending seminars on research skills or asking for advice at the research data management centre or the open science office of the "local" university. #OSDMA25
[lunch break now. :)]

session 2 of #OSDMA25 is called "Incentives for Open Science".
first up is Christine Emmer from the Mannheim Faculty of Social Sciences, working in Health Psychology. she will talk about "Opening Doors: Incentives and Impact of Open Science in Social Science Research".
I like her slide on "why open science?"! 👌

Christine has been awarded the Bojanovsky price 2024 with a special mention of her "exemplary Open Science practices", including preregistration and open data sharing – a nice incentive!
› https://www.sowi.uni-mannheim.de/mata/news/?tx_news_pi1%5Baction%5D=detail&tx_news_pi1%5Bcontroller%5D=News&tx_news_pi1%5Bnews%5D=26076&cHash=464c64ed7dd812118b4aec93569ad556 [article in German]
the paper on MADOC, the @unimannheim's #openaccess repository: https://madoc.bib.uni-mannheim.de/66299/ #OSDMA25

with this paper as an example she takes us through the process of "doing" #OpenScience. one important lesson learned: "Deviations from preregistration are not a problem!" #OSDMA25

also the "transparency and openness" sections of articles are not rocket science. :)
at Christine's institute they introduced preregistration also for bachelor's and master's theses. this is a bit of work in the beginning, but then very helpful.
registered reports is very helpful as well: it's taking preregistration one step further, with the preregistration submitted to a journal and peer-reviewed. #OSDMA25

for her paper they shared the data and analysis code through OSF, and they published the data as part of a Community Augmented Meta-Analysis (CAMA): https://cama.psychopen.eu/inspection/CAMA_Discrim #OSDMA25

Christine has some valuable tips for if you want to start doing #OpenScience: "there doors are open, and it is worth it!" #OSDMA25

from the discussion: is the preregistration via OSF feasable? yes. should there be more meta analysis? yes. did christine use tools for selecting papers for the systematic review, or was it done manually? as they started in 2019 it was done manually, but now there are [not perfect, but helpful] tools available like e.g. Rayyan.ai. #OSDMA25

next up: Matthew Robertson, Faculty of Social Sciences, working in Social Data Science, on "From Audit to Action – Closing the Trans­parency Gap in Quanti­tative Social Science".
he tells us about the paper "Measuring transparency in the social sciences: political science and international relations", https://doi.org/10.1098/rsos.240313.
they found two types of transparency problems, and they developed two research questions. #OSDMA25

when collecting their data they even ran into trouble with the library! 🙃 #OSDMA25

this is phase 2 of their preparation of the paper and the first results. #OSDMA25

... more results ... #OSDMA25

what can be done so that nobody has to do this again? "just collect the data and send it to Crossref!" #OSDMA25

Matthew thinks that the solution he has in mind is not toooo complicated, but it might need some #incentives to get implemented and used. perhaps professional associations could act as a kind of accelerator? #OpenScience #OSDMA25

in the discussion Matthew answers some detailed questions about the data collection and finding process, and there are some aspects discussed with somebody from a journal. #OSDMA25

there is now a poster session and a coffee break in Mannheim.
the posters by the four Open Science Grant Awardees are also available on the #OSDMA25's website: https://www.bib.uni-mannheim.de/ihre-ub/ausstellungen-und-veranstaltungen/open-science-day-2025/
"The psychology of heat: A global big team science effort" by Jana Berkessel, "SpatialBenchRAG: A Multimodal, Multilingual Benchmark for KG-grounded RAG over Diverse Geographies" by Andreea Iana, "Do Open Large Language Models Conduct Content Analysis like Humans?" by Indira Sen, and …

…"Studying Radicalized Individuals and Their Passive Social Media Use: A Person-Centered and Privacy-Sensitive Approach" by Frieder Rodewald & Nina Osenbrügge. #OSDMA25

session 3 of #OSDMA25 is about "Open Science in Business & Economics Research".
first on the floor is Lennart Reissner, BERD@NFDI, talking about "BERD – Services for Professional Management of Research Data + Algorithms in Business & Economics". the advent of big amounts of unstructured and non-standard data is "disturbing" the traditional model of empirical research, but it also offers potential for discoveries.
[currently I cannot take any screenshots from the slides; I will add some later.]

the BERD@NFDI platform is available here: https://www.berd-nfdi.de/.
Lennart will present some of BERD's core services today.
BERD has 4 key access points: the data portal, analytics, training & education, and services.
starting with data:
your want #OpenBigData for #business, #economics, and related #socialsciences? go here: https://openbigdata.org/! :)
the BERD Research Data Marketplace (in early stage) bridges the gap between researchers & organizational data: https://researchdatamarketplace.com/.
#OSDMA25

in BERD's AI Analytics Portal you get access to machine learning-based publications: https://www.berd-nfdi.de/analytics-portal-new/.
in the workshops, events, and self-paced courses held by the BERD Academy you can learn about data science and data management: https://www.berd-nfdi.de/berd-academy/. #OSDMA25

and there is BERD's Generative AI directory: https://www.gettingstuffdone.ai/,
legal assisctance: https://www.berd-nfdi.de/legal-questions/,
OCR Services: https://www.berd-nfdi.de/optical-character-recognition-ocr/
and Knowledge Graph Infrastructure: https://www.berd-nfdi.de/knowledge-graph-infrastructure/
#researchdata #business #economics #socialsciences #OSDMA25

so, what is BERND@NFDI in a nutshell?
a consortium for bsiness, economics, and related fields with a focus on unstructed text, image, audio, and video, offering a comprehensive selection of services and tools that helps to efficiently find, analyse, and manage a wide range of research data and methods.
» https://www.berd-nfdi.de/about/
#OSDMA25

from the discussion: does BERD contact companies to collect more data? yes, they have approached several companies, and they had good feedback about it. some partnerships are currently in the making. in the future BERD does not need to contact them any longer, but that the companies will offer them their data. #OSDMA25

does Lennart have a preferred BERD service? he thinks that the Research Data Marketplace will be able to enable interesting research. but this is something for the future; at the moment he likes Open (Big) Data portal best, for giving visibility to the data. #OSDMA25

"last in line" is Klaus Tochtermann, Leibniz Information Centre for Economics (ZBW), who will be talking about "The BERD Data Portal – Connected. National. European."
he puts BERD in the NFDI context, being one of 27 consortia: https://www.nfdi.de/, https://de.wikipedia.org/wiki/Nationale_Forschungsdateninfrastruktur (in German).
and speaking of #OpenScience BERD with its services addresses open scientific knowledge, open dialogue with other knowledge systems, open engagement of societal actors, and open science infrastructure, #OSDMA25

Klaus now focusses on the BERD portal – https://berd-platform.de/ – and how it can be connected to other services. being based on InvenioRDM (https://inveniosoftware.org/products/rdm/) the portal uses standards that make it possible to link it e.g. to other NFDI consortia, sections, and committees. this ensures a high degree of consistency with NFDI developments towards "oneNFDI".
#OSDMA25

BERD can also be linked to EOSC, the European Open Science Cloud (https://en.wikipedia.org/wiki/European_Open_Science_Cloud). this is important not only bc the German Science Council recommended integrating NFDI more closely into the European research landscape, participating in the EOSC development, and hence positioning NFDI as a an integral "node" of the EOSC. this could lead to a european "system of systems". #OSDMA25

the BERD datasets are now available in the EOSC Resource Hub (https://open-science-cloud.ec.europa.eu/resources/all), they are connected to resources in Zenodo (https://en.wikipedia.org/wiki/Zenodo), and they are being integrated with Galaxy Europe for data analysis (https://galaxyproject.org/eu/). #OSDMA25

from the discussion: is data in OpenAIRE directly part of the European Science Cloud? yes. #OSDMA25

and with this #OSDMA25 comes to a close.
a big thank you to David Philip (@DPMorgan) & Philipp Zumstein (@zuphilip) and their colleagues for organizing this interesting #OpenScience day at @unimannheim! 🙏