Anonymisation glossary

Personal data:

Personal data means any information relating to an identified or identifiable individual. An identifiable individual is one who can be identified, directly or indirectly.

Direct identifiers:

Information that is sufficient on its own to identify an individual (e.g., name, telephone number, Student ID (ERNA ID at EUR), government issued ID, facial image)

Indirect identifiers:

Data that identifies an individual indirectly (e.g., date of birth, gender, ethnicity, location, cookies, IP address, license plate number), especially when several identifiers are combined. This is often also context-dependent, for example notable traits like unusual job titles, very old age, very rare diseases, or specific positions within an organisation can be indirect identifiers, especially in relatively small datasets.

Special categories of personal data:

Data that concerns health, race, ethnic origin, political views, religious beliefs, ideological convictions, trade-union membership, sexual behaviour, sexual orientation, genetic data, or biometric data.

Sensitive data:

Data that can have negative consequences for an individual if made public. The term is not defined by the GDPR. It is used as an umbrella term for special categories of personal data and other types of data such as criminal data, sensitive financial data, social security number (BSN), data with a risk for public safety, or data that can be used to discriminate against other people.

Criminal data:

Data relating to criminal convictions and offences or related security measures.

Anonymisation:

The process by which personal data is irreversibly altered in such a way that an individual can no longer be identified or is identifiable directly or indirectly. Once data is truly anonymous the GDPR does not apply (recital 26 GDPR).

Be aware that whether something is anonymous can change over time, with more data and new technology becoming available. Hence, rather than “anonymity” or “anonymous data”, it is good to talk about data to which anonymisation techniques have been applied.

Pseudonymisation:

Pseudonymisation of data means replacing any identifying characteristics of data with a pseudonym, or, in other words, a value which does not allow the research participant to be directly identified. It is different from anonymisation as, in many cases, it still allows identification using indirect identifiers.

The pseudonymisation key, i.e., the file (or other form) in which the direct identifiers are linked to the pseudonym, needs to be stored separately from the pseudonymised data.

De-identification:

Removal or obscuring of identifying information in a dataset to prevent identification of specific cases. The term can refer to different actions depending on the circumstances and can be confusing. We prefer talking about anonymisation and pseudonymisation, since these terms are defined within the GDPR.

Data minimisation:

Only the minimum amount of personal data necessary to accomplish a specified purpose (e.g., the goal of the research) should be collected. Personal data must not be collected just in case they might become useful in the future. There must be a clear, specified need for collecting the personal data.