Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.502255
Title: A framework for understanding statistical disclosure processes : a case study using the UK's neighbourhood statistics
Author: Mackey, Elaine Catherine
Awarding Body: The University of Manchester
Current Institution: University of Manchester
Date of Award: 2008
Availability of Full Text:
Access from EThOS:
Abstract:
Managing disclosure risk through data restriction does not produce absolutely safe data. Therefore the question addressed in this thesis is: whether the data environment can provide a (complementary) way of managing the disclosure problem for released data. I suggest that it is possible to have at least some control over the data environment but it requires: (i) establishing how a disclosure might actually arise and, (ii) identifying and exploring all components of disclosure. In order to do this the thesis sets out a new framework for assessing disclosure consisting of both theoretical and practical components. The theoretical component of the framework introduces a new way of theorising disclosure termed the environment centric approach. In contrast to more traditional approaches in Statistical Disclosure Control (SDC), which ask 'how safe are the data for release' it places at the heart of its analysis the broader question of what is a disclosure event? This question incorporates not just the notion of 'how might it happen' but also the idea of 'what happens next'. These two elements are explored through an analysis of the actions and interactions of the key agents involved in creating a disclosure. The practical component of the framework employs a game theoretic reasoning to model the interactions between agents. By modelling agents' actions we can move beyond simply describing them to look at how one agent acts in relation to another and further what the outcome of them might be. A focus on the outcomes of disclosure offers us an opportunity to develop strategies for influencing them and in this way further manage the disclosure problem. This is based on the underlying construct that one can manipulate the data environment as well as the data, as a means for dealing with the disclosure problem.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.502255  DOI: Not available
Share: