ABOUT US

OUR PROJECTS

Blue sky thinking

Risk utility management tool

A tool to evaluate the sensitivity or disclosure risk of microdata

This project will offer data owners and archives the ability to evaluate the sensitivity or disclosure risk of microdata using an R-based tool working with Statistical Disclosure Control parameters and algorithms. The project will develop and extend the existing R package 'sdcMicroGUI' to create an enhanced software application that will safely deliver anonymised versions of a confidential data source. The tools will be tested with real data from the Office for National Statistics.

Based on a three month demonstrator phase, the next phase involves further development and extension of the R package 'sdcMicroGUI' to create a software application to deliver a usable product to the UK Data Service.  The results of the early work are encouraging, and suggests that a product can be safely used to deliver anonymised versions of a confidential source of data.

The tool's interface will enable easy-to-browse variables and incorporation of key metadata, full log-and syntax-recording for future use. The product will be easy to install and  be fully documented with detailed description of the Statistical Disclosure Control parameters and algorithms. A period of extensive testing of the product will be carried out on 'fake' confidential data with input from key data owners and producers.

It is hoped that the product will be a useful tool for data owners, producers and other data archives to objectify the process by which disclosure risk for data release is undertaken.

Archive contribution

The Archive directs, manages ad provides the intellectual overveiw of the project, with technical input from the Norwegian Data Service in Bergen.

Principal investigator:  Richard Welpton
Funder:  ESRC
Dates:  October 2014 - March 2015
Contact:  Richard Welpton
Links: