When research data includes personal data, privacy issues can emerge. Researchers using personal data need to fulfill legal and ethical obligations, eliminating the direct and traceable reference to, or presence of, personal identifiers. Anonymizing such information allows to eliminate the relationship between a person and the data deriving from surveys, tests or studies concerning that person. Thus, the data subject is no longer identifiable when data is anonymized appropriately.
There are different ways to convert personally identifiable information into aggregated data, which can still be re-associated with the initial personal data at a later time and with specific access and use restrictions.
Furthermore, different disciplinary areas (e.g. STEM sciences, HSS sciences) may have domain-specific standards, tools and procedures for the anonymization of personal data.
Here are some guidelines for different disciplines and here, especially, for projects dealing with human data.
A relevant tool which can be used to anonymize research data for University of Milan researchers is Amnesia. Moreover, here is a numeric data anonymisation tool: a practical R package for checking disclosure risk through examining combinations of key variables.
In case of anonymization of personal data by researchers, the procedures and tools which you use should be always documented in both published datasets and DMPs.
Regarding the protection of personal data, please contact UNIMI Privacy and Dpo Support Office.
Contact the unviersity’s technological transfer office if: you research will use other party/non public data/material, you need to share non public data/proprietary material, you think your results will be innovative and have industrial application.
Contact the university’s ethical committee if your research involves experiments on humans/animals.