Protecting Quality and Confidentiality of Data by MASSC: A Survey-sampling-based Method

• Activity Number: 69
• Type: Topic Contributed
• Date/Time: Monday, August 9, 2004 : 8:30 AM to 10:20 AM
• Sponsor: Section on Survey Research Methods
• Author(s): David H. Wilson*+ and Avinash Singh and Feng Yu
• Companies: RTI International and RTI International and RTI International
• Keywords: disclosure ; risk ; SDL ; MASSC ; privacy ; confidentiality

Abstract

The credibility of a data producer is at stake if the respondents in a database are not assured of the confidentiality of their sensitive information collected by the producer. In a scenario known as "disclosure by response knowledge," the respondent identifies his own record from the database and is concerned about its disclosure by someone who might know enough about the respondent to identify his record. The MASSC method developed at RTI addresses the above problem using a probabilistic framework that allows for random, but controlled, perturbation and suppression. The MASSC method views a database as a population and relies on the analogy between releasing an untreated database and conducting a census of the population. Survey sampling methods are used in MASSC to provide simultaneous control and measurement of disclosure risk and information loss. The methods used in MASSC consist of four steps: micro agglomeration for partitioning the database into risk strata, probabilistic substitution for perturbation, probabilistic subsampling for suppression, and sampling-weight calibration for preserving estimates for key variables.