Bern Data Talk - Workflows and Challenges in Data Anonymization

Workflows and challenges in data anonymization

Anonymization is a key strategy for the ethical handling of personal data - it helps prevent harm and protects the privacy of study participants. It is also a legal requirement under data protection laws. However, the way anonymization is implemented depends heavily on the discipline, the nature of the research, the type of data involved, available tools, and other factors.

Join us to hear first-hand how fellow researchers tackle these challenges. And don’t miss the Apéro afterwards for engaging conversations with speakers and colleagues!

Speakers and Topics

• Prof. Dr. Florian Busch (UniBE, Institute of Germanic Languages and Literatures): Smartphone Screen Captures as Linguistic Data: Ethical and Technical Considerations of Collection and Processing. --- The SNSF-funded research project Texting in Time (2023–2027) collects authentic screen recordings of smartphone use in Switzerland and Germany. These videos provide insights into the linguistic and social everyday practices of digitally mediated communication in participants’ lives. This contribution outlines the ethical dimensions of data collection and processing as well as the technical challenges, workflows, and objectives that the project both faces and pursues.
• Prof. Dr. Christoph von Dach (BFH, School of Health Professions): Jumping the Hurdles: Anonymization Challenges in an International Erasmus+ Project. --- The talk focuses on the anonymization of data in a multi-method research design of an international Erasmus+ research project in the health sector. Obstacles include the protection of personal data in focus groups and narrative data, as well as in the international exchange of the collected data.
• Luisa Jansen M Sc (UniBE, Institute of Psychology): Putting Privacy to the Test: My Experience with Red vs. Blue Teaming in Research Data Anonymization. --- In this talk, I share my experience anonymizing data from a mixed-methods study evaluating a communication training for professionals. Committed to open data, I initially pseudonymized all datasets prior to publication. To re-assess and improve the privacy of participants, I then applied a red vs. blue team approach: a technique commonly used in security testing, simulating re-identification attacks. I will reflect on the challenges this approach presented and contextualize my process with research findings on data protection practices of other researchers.
  

When? Tuesday, 28 October, 2025, 4:15 – 5:15 pm, followed by an Apéro for in-depth discussions

Where? TBD

More information will follow