Suchergebnisse

Intro -- ACKNOWLEDGEMENTS -- 1. THE SCOPE OF THE STUDY -- 1.1 Introduction -- 1.2 Data models -- 1.3 Measurement models -- 1.4 Outline of the study -- PART I. THEORY AND METHOD -- 2. THE DETERMINISTIC MODEL: THE GUTTMAN SCALE -- 2.1 Introduction -- 2.2 The perfect scale -- 2.3 The two sets as populations -- 2.4 The imperfect scale: the problem of "error" -- 2.5 Coefficients of scalability -- 2.6 Procedures of scale analysis -- 2.7 The quasi-scale: a stochastic model -- 3. PROBABILISTIC RESPONSE: LATENT STRUCTURE MODELS -- 3.1 Latent structure analysis -- 3.2 Latent structure models for dichotomous data -- 4. HOMOGENEITY AND HOLOMORPHISM IN SCALING MODELS: SOME SIMPLE PROPERTIES -- 4.1 Holomorphic two-parameter models -- 4.2 Scale statistics: score, reliability, scalability and patterns -- 4.3 Approximate sampling distributions -- 5. A CLASS OF SCALING PROCEDURES -- 5.1 Summary and evaluation of the findings -- 5.2 Description of scaling procedures -- PART II. APPLICATIONS IN POLITICAL RESEARCH -- 6. CROSS-CULTURAL COMPARISON: THE DISCOVERY OF DIMENSIONAL IDENTITY -- 6.1 On cross-cultural comparison -- 6.2 The study -- 6.3 The study -- 6.4 A Dutch study (Amsterdam, 1964) -- 6.5 A second Dutch study (Amsterdam, 1965) -- 6.6 A scale of political interest -- 6.7 Conclusions -- 7. THE CROSS-CULTURAL ROBUSTNESS OF SCALES: POLITICAL EFFICACY -- 7.1 Dutch-American comparisons of the "sense of political efficacy" -- 7.2 Some further possibilities -- 8. AN EXPLORATION OF POLITICAL EFFICACY -- 8.1 A Dutch extension of the efficacy scale -- 8.2 An analysis across sub-groups -- 8.3 A broader efficacy dimension: a combined scale (17 items) -- 8.4 Investigating the double monotony of scales -- 8.5 The reliability of doubly monotone scales -- 9. APPLICATIONS OF MULTIPLE SCALING -- 9.1 Scaling opinion leadership.

In this paper a model is developed for assesing disclosure risks of a microdata set. It is an extension of the one presented in Bethlehem et al. (1988). It is used to calculate (an upper bound of) the risk that an investigator is able to reidentify at least one individual in an anonimyzed data set, and hence discloses some sensitive information about him. This risk is shown to depend on, among other things, two variables which can be controlled by the statistical office which is disseminating such a data set: the 'coarseness' of the key variables and the size of the data set. The model yields guidelines as to the usage of these two instruments to control the disclosure risk.