Anonymity Guarantees Research Articles

Protecting respondents identities in microdata release

Today's globally networked society places great demands on the dissemination and sharing of information. While in the past released information was mostly in tabular and statistical form, many situations call for the release of specific data (microdata). In order to protect the anonymity of the entities (called respondents) to which information refers, data holders often remove or encrypt explicit identifiers such as names, addresses, and phone numbers. Deidentifying data, however, provides no guarantee of anonymity. Released information often contains other data, such as race, birth date, sex, and ZIP code, that can be linked to publicly available information to reidentify respondents and inferring information that was not intended for disclosure. In this paper we address the problem of releasing microdata while safeguarding the anonymity of respondents to which the data refer. The approach is based on the definition of k-anonymity. A table provides k-anonymity if attempts to link explicitly identifying information to its content map the information to at least k entities. We illustrate how k-anonymity can be provided without compromising the integrity (or truthfulness) of the information released by using generalization and suppression techniques. We introduce the concept of minimal generalization that captures the property of the release process not distorting the data more than needed to achieve k-anonymity, and present an algorithm for the computation of such a generalization. We also discuss possible preference policies to choose among different minimal generalizations.

Use of the hepatitis-B vaccine and infection with hepatitis B and C among orthopaedic surgeons. The American Academy of Orthopaedic Surgeons Serosurvey Study Committee.

We used a questionnaire, with a guarantee of anonymity to the respondents, and conducted serological testing of 3411 attendees at the 1991 Annual Meeting of The American Academy of Orthopaedic Surgeons to evaluate the prevalences of infection with the hepatitis-B and C viruses and the use of the hepatitis-B vaccine among orthopaedic surgeons. There was evidence of infection with hepatitis B in 410 (13 per cent) of 3239 participants who had reported having no non-occupational risk factors; 2103 (65 per cent) reported that they had been immunized with the hepatitis-B vaccine. Of 3262 participants who reported having no non-occupational risk factors and who were evaluated for infection with hepatitis C, twenty-seven (less than 1 per cent) tested positive for the antibody to the hepatitis-C virus. The prevalence of previous infection with hepatitis B increased with increasing age; four (3 per cent) of 136 surgeons who were twenty to twenty-nine years old had evidence of infection, whereas ninety-six (27 per cent) of 360 surgeons who were sixty years old or more had evidence of infection. The prevalence of infection with hepatitis C also increased with increasing age; none of 135 surgeons who were twenty to twenty-nine years old had evidence of infection, and five (1 per cent) of 360 surgeons who were sixty years old or more had evidence of the virus. The prevalence of vaccination decreased steadily with age: 123 (90 per cent) of 136 surgeons who were twenty to twenty-nine years old reported that they had received the hepatitis-B vaccine, whereas 127 (35 per cent) of 360 surgeons who were sixty years old or more reported that they had received the vaccine. The prevalence of infection with hepatitis B or hepatitis C was not associated with the measured indices of exposure to the blood of patients (the number of cutaneous or mucosal contacts with blood that had occurred within the previous month or the number of percutaneous injuries that had occurred within the previous month or year, as recalled by the participants). In conclusion, the prevalence of immunization with the hepatitis-B vaccine was high among the orthopaedic surgeons studied. Although the prevalence of infection with the hepatitis-C virus was several times greater in the current investigation than has been reported in studies of blood donors in the United States, infection with this virus was not associated with the indices of occupational exposure to blood measured in this study.

Community Response to Crime in a Sun Belt City

This research examines a crime-specific program designed to mobilize citizens to provide information to the police about crimes in exchange for money and a guarantee of anonymity. The Crime Line program is nongovernmental and derives its greatest financial support from the business community. An analysis of the data shows that the program tends to reflect the source of its contributions.

Protecting Research Subjects and Unintended Consequences: The Effect of Guarantees of Confidentiality

There is significant concern among researchers and government administrators that individuals who serve as research subjects be protected from risks associated with empirical inquiry. Survey researchers frequently administer guarantees of confidentiality and anonymity in an effort to minimize subjects' psychological harm and to increase the validity of data. This study tests the effect of guarantees of confidentiality and anonymity on (1) the evaluation apprehension of juvenile offenders who provided self-reports of past behavior soon after arrest, (2) their response rates to individual interview items, and (3) the quality of the subjects' responses. Findings are reported and implications for survey researchers are discussed.

The Scientist as Public Administrator

SACRIFICE of research career for one in research administration is common among scientists in the United States today.' It is paradoxical that so many men desert highly respected, even idealized, way of life, for which their professional training and experience have prepared them, to enter an occupation whose very professionalism is suspect. The research administrator stands in an area of ambiguity and transition in the meaning of work: though he enjoys considerable prestige in society and salary and power advantages over researchers within bureaucratic organizations, the actual conduct of research has been at the core of the image of the good scientist.2 If a man's work is as good clue as any to the course of his life, and to his social being and identity, 3 basic elements in the scientific career and the bureaucratic system may be illumined by understanding the changing of roles by the scientist. Is the scientist who becomes an administrator misled by ambition and then corrupted by power, or is he enabled to realize fuller potentiality and make unique contribution? To answer this, two subsidiary questions must first be examined: Why do men leave research to become administrators? Does the researcher-turned-public administrator remain always the scientist, or are his attitudes toward politics, science, and administration those of the administrator? Where other sources are not indicated, the analysis presented below relies heavily upon series of interviews with Washington area federal research administrators. Two dozen or so each in the National Institutes of Health, the Agricultural Research Service, and the Navy, plus some research scientists in these agencies and some Budget Bureau and Civil Service Commission officials, were interviewed. The administrators, most of whom had been trained as scientists, were representative of variety of ranks and functions, though purposely bunched toward the top. Interviewees, none of whom objected to the taking of full notes with the guarantee of anonymity, spoke in response to statement of the purpose of the research and specific questions appropriate to the person, agency, and level. The writer did not feel that it was impossible to discriminate qualities of openness and perspicacity in those interviewed. Conversational interviewing was no doubt facilitated by the cathartic value to many persons in talking about themselves to detached observer, the ability of men within formal organization to learn good deal about close colleagues, the