Qualitative data anonymisation: theoretical and practical considerations for anonymising interview transcripts

PDF
Publication

FORS Guide Nº 20

DOI
https://doi.org/10.24449/FG-2023-00020
How to cite

Stam, A, Diaz, P. (2023). Qualitative data anonymisation: theoretical and practical considerations for anonymising interview transcripts. FORS Guide No. 20, Version 1.0. Lausanne: Swiss Centre of Expertise in the Social Sciences FORS. doi:10.24449/FG-2023-00020

Abstract

This guide addresses qualitative data anonymisation from a theoretical, legal and practical perspective. After defining qualitative data anonymisation and addressing some common forms of resistance when it comes to anonymising such data, this guide presents the main anonymisation techniques. It argues for a layered approach and encourages the reader to consider anonymisation as an important layer of protection amongst others.

Recommendations

  • Only collect personal data if you really need to. The collection of personal data is subject to constraining legal obligations. The more such data you have, the more complex the anonymisation procedure may be.
  • Get proper consent to handle personal data. It is important to design information and informed consent in a way that takes into account the disclosure (sharing) of data to third parties in compliance with data protection laws.
  • Never promise full anonymisation. Remember that whatever you promise research participants when collecting data is a promise you have to keep, and that depending on what you said you may limit the use of your data or you may even be obliged to destroy them. Think carefully about the wording of your informed consent, be it oral or written, and do not hesitate to consult our FORS Guide on the topic (Kruegel, 2019).
  • Do not limit anonymisation to direct identifiers. It takes more than simply removing direct identifiers, such as participants’ names, for data to be anonymised.
  • Do not confuse anonymisation with pseudonymisation. The consequences are very different, since unlike pseudonymized data, anonymised data are not subject to data protection laws.
  • Do not take anonymisation too lightly. The anonymisation of data is not an exact science, and you must carefully evaluate the risks of re-identification of research participants, as well as of anyone else who was mentioned in the interview, in order to be able to take the appropriate measures.
  • Do not over-anonymise. It is better to add other layers of protection, such as informed consent and access control, than to remove too much information to the point that the data become useless for future reuse.
  • Plan ahead for anonymisation. You need to consider anonymisation as early as possible in the research project, ideally already at the time of the submission of the research proposal and data management plan. Keep in mind that you can request financial resources for anonymisation in grant proposals to the SNSF. Think carefully about when anonymisation should take place during the project, and assign responsibilities to ensure that someone keeps an eye on it and makes sure it is done at the right time.
  • Develop an anonymisation strategy. The anonymisation strategy should be set up at the start of the project and should include a reflection on the data you need to collect, to avoid collecting personal data that are not necessary. Review the strategy as the project evolves, which includes setting up file handling rules and identifying mandatory and possible information to be anonymised.
  • Choose the right anonymisation techniques. Assess potentially identifying information and the corresponding risks. Make the best choices with the aim of both protecting research respondents and retaining as much value as possible in the data.

    • Copyright

    © the author 2023. This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0)

    Publication year

    2023

    Data Citation: How and Why Citing (Your Own) Data
    Data Sharing in the Social Sciences