The SACRO guide to statistical output checking

This document was developed as an output of the project SACRO (Semi-Automatic checking of Research Outputs). This work is funded by UK research and Innovation [Grant Number MC_PC_23006] as part of Phase 1 of the DARE UK (Data and Analytics Research Environments UK) programme, delivered in partnership with Health Data Research UK (HDR UK) and Administrative Data Research UK (ADR UK).

This document has three sections, plus references:

Part I discusses the operational and statistical theory underlying all aspects of output statistical disclosure control (OSDC). It is intended for those who want a deeper understanding of the concepts underlying practical and efficient OSDC.

Part II is the practical manual, intended to be a reference for users and output checkers. It lists the rules (or rules-of-thumb) to be followed for all possible outputs, organised by class. For each class of statistics, the manual presents a quick summary, an example, a discussion of the key risks, a rationale for any rules applied, and guidelines on how to evaluate the output if an exception is requested. A detailed statistical explanation of the risk assessment is not given, but other works are referenced for those interested. A separate chapter discusses graphs.

Part III provides FAQs on both statistics and outputs, plus some guidance for output checkers on how to respond to queries