Privacy by design in statistics: Should it become a default/standard?

Article type: Research Article

Authors: Kubo, Baldur^{a; *} | Sahk, Angela^a | Berendsen, Veiko^b | Saluveer, Erki^c

Affiliations: [a] Cybernetica AS, Ülikooli 2, Tartu, Estonia | [b] Statistics Estonia, Tatari 51, Tallinn, Estonia | [c] OÜ Positium, Soola 8, Tartu, Estonia

Correspondence: [*] Corresponding author: Baldur Kubo, Cybernetica AS, Ülikooli 2, Tartu, Estonia. E-mail: [email protected].

Abstract: Privacy by design (PbD) as an approach to systems engineering has been conceptualized for more than a decade. It has inspired the legal norm of managing personal data in EU and incorporated into the GDPR. In practice PbD is far from being the standard in systems engineering in both private and public sector procurement. We illustrate why privacy needs to be engineered into statistical systems and explain risk when privacy is patched into a solution. To show how the statistical system could proceed in adopting PbD we bring an example from a high level self assessment of a statistical organization. We walk through future proofing existing services by redesigning an official tourism statistics application that uses mobile location big data. Finally we envision how statistical organizations can cooperate in supporting each other and learn together how to build trust required in utilising new data sources.

Keywords: Privacy engineering, privacy by design, official statistics

DOI: 10.3233/SJI-190532

Journal: Statistical Journal of the IAOS, vol. 35, no. 4, pp. 623-631, 2019