Runners, repeaters, strangers and aliens: Operationalising efficient output disclosure control

Alves, Kyle; Ritchie, Felix

Abstract

Statistical agencies and other government bodies increasingly use secure remote research facilities to provide access to sensitive data for research and analysis by internal staff and third parties. Such facilities depend on human intervention to ensure that the research outputs do not breach statistical disclosure control (SDC) rules. Output SDC can be principles-based, rules-based, or something in between. Principles-based is often seen as the gold standard statistically, as it improves both confidentiality protection and utility of outputs. However, some agencies are concerned that the operational requirements are too onerous for practical implementation, despite these statistical advantages. This paper argues that the choice of output checking procedure should be seen through an operational lens, rather than a statistical one. We take a popular conceptualisation of customer demand from the operations management literature and apply it to the problem of output checking. We demonstrate that principles-based output SDC addresses user and agency requirements more effectively than other approaches, and in a way which encourages user buy-in to the process. We also demonstrate how the principles-based approach aligns better with the statistical and staffing needs of the agency.

• Journal Article Type: Article
• Acceptance Date: Aug 24, 2020
• Online Publication Date: Sep 10, 2020
• Publication Date: Nov 25, 2020
• Deposit Date: Aug 26, 2020
• Publicly Available Date: Oct 11, 2020
• Journal: Statistical Journal of the IAOS
• Print ISSN: 1874-7655
• Publisher: IOS Press
• Peer Reviewed: Peer Reviewed
• Volume: 36
• Issue: 4
• Pages: 1281-1293
• DOI: https://doi.org/10.3233/SJI-200661
• Keywords: Confidentiality, statistical disclosure control, operations management, output checking
• Public URL: https://uwe-repository.worktribe.com/output/6634925
• Publisher URL: https://officialstatistics.com/