Releasing microdata: disclosure risk estimation, data masking and assessing utility , Southampton, UK Southampton Statistical Sciences Research Institute 15pp.
(S3RI Methodology Working Papers, M09/02).
Statistical Agencies need to make informed decisions when releasing sample microdata from social surveys with respect to the level of protection required in the data and the mode of access. These decisions should be based on objective quantitative measures of disclosure risk and data utility. This paper reviews recent developments in disclosure risk assessment and discusses how these can be integrated with established methods of data masking and utility assessment for releasing microdata. We illustrate the Disclosure risk-Data Utility approach based on samples drawn from a Census where the population is known and can be used to investigate sample-based methods and validate results.
||log-linear models, measurement error, additive noise, micro-aggregation, random rounding, pram, information loss
|11 February 2009||Published|
||12 Feb 2009
||18 Apr 2017 21:59
|Further Information:||Google Scholar|
Actions (login required)