As part of the Data Stewardship Programme, the University of Basel offers train-the-trainer courses for its prospective data stewards, to train them in research data management (RDM).
Matthias Templ gave a workshop on Introduction to data anonymization of quantitative data.
This workshop addressed the basics of data anonymization, data synthetization, disclosure risk and data utility after anonymization.
To be clear: this course was about anonymisation and not pseudo-anonymisation using encryption methods. The latter does not solve the problem of data protection. The course gave an overview of data anonymisation methods but does not leave out how to measure disclosure risk - the second most important task before and after applying anonymisation methods. We also discussed data utility, the most important part of data anonymisation, as anonymisation is about providing anonymized data with high data utility. Participants gained an awareness of data attack scenarios, the sensitivity of data and the exploratory nature of data anonymisation. Participants have been made aware that the choice of anonymisation methods depends on many parameters and is case-dependent. After this workshop, data stewards will know when data anonymisation support is needed, what disclosure control methods are available and where to find additional information and services to sufficiently ensure a low disclosure risk of personal data in a research project while maximising data utility.
The workshop dealt with the introductory basics of data anonymisation of quantitative data, but also shows alternative data anonymization based on data synthesis. In addition, the problem for anonymization of qualitative (text) data was outlined.
The course was primarily aimed at data stewards and other people working in RDM support.