Search PubMed⌕ Search

Biomedical subjects

S E Fienberg

Publications and source records attributed to S E Fienberg.

16 recordsLinked to original sources
PubMed2000

Bounds for cell entries in contingency tables given marginal totals and decomposable graphs.

Upper and lower bounds on cell counts in cross-classifications of nonnegative counts play important roles in a number of practical problems, including statistical disclosure limitation, computer tomography, mass transportation, cell suppression, and data swapping. Some features of the Frechet bounds are well known, intuitive, and regularly used by those working on disclosure limitation methods, especially those for two-dimensional tables. We previously have described a series of results relating these bounds to theory on loglinear models for cross-classified counts. This paper provides the actual theory and proofs for the special case of decomposable loglinear models and their related independence graphs. It also includes an extension linked to the structure of reducible graphs and a discussion of the relevance of other results linked to nongraphical loglinear models.

Journal Article↗
PubMed1994

Statistical models for the analysis of ordered categorical data in public health and medical research.

In the late 1970s statisticians extended the methods for analysing loglinear and logit models for cross-classified categorical data to incorporate information about the ordinal structure of the categories corresponding to some of the classification variables. In this paper we review one class of such extensions known as association models. We consider association models with and without order restrictions on the parameters and we use these models to answer research questions about several medical examples involving ordered categorical data. We emphasize the interpretation of parameters in the association models and how this relates to the research questions of interest.

Adult↗
PubMed1993

A three-sample multiple-recapture approach to census population estimation with heterogeneous catchability.

"A central assumption in the standard capture-recapture approach to the estimation of the size of a closed population is the homogeneity of the 'capture' probabilities. In this article we develop an approach that allows for varying susceptibility to capture through individual parameters using a variant of the Rasch model from psychological measurement situations. Our approach requires an additional recapture. In the context of census undercount estimation, this requirement amounts to the use of a second independent sample or alternative data source to be matched with census and Post-Enumeration Survey (PES) data.... We illustrate [our] models and their estimation using data from a 1988 dress-rehearsal study for the 1990 census conducted by the U.S. Bureau of the Census, which explored the use of administrative data as a supplement to the PES. The article includes a discussion of extensions and related models."

Americas↗
PubMed1985

Longitudinal analysis of categorical epidemiological data: a study of Three Mile Island.

The accident at the Three Mile Island nuclear power plant in 1979 led to an unprecedented set of events with potentially life threatening implications. This paper focusses on the analysis of a longitudinal study of the psychological well-being of the mothers of young children living within 10 miles of the plant. The initial analyses of the data utilize loglinear/logit model techniques from the contingency table literature, and involve the fitting of a sequence of logit models. The inadequancies of these analyses are noted, and a new class of mixture models for logistic response structures is introduced to overcome the noted shortcomings. The paper includes a brief outline of the methodology relevant for the fitting of these models using the method of maximum likelihood, and then the model is applied to the TMI data. The paper concludes with a discussion of some of the substantive implications of the mixture model analysis.

Accidents↗
PubMed1985

Cognitive aspects of health survey methodology: an overview.

The past 25 years have seen the development of a wide variety of sample surveys dealing with the nature and distribution of illness and disability, and with the utilization of health care services. The sample survey is currently the most widespread and influential instrument for judging the health status of the nation and for guiding health policy. The knowledge, beliefs, and attitudes of survey respondents "subjectively" affect what the survey seeks to "objectively" measure. Even as statistical sampling has been refined, so is it important to reexamine what the cognitive sciences have to offer for survey interview structure and content.

Adult↗
PubMed1985

Recalling pain and other symptoms.

Questions relating to symptoms are an important ingredient in many surveys of health status. Yet, the understanding of mechanisms for the recall of pain, and the cognitive aspects of memory for pain and other symptoms, have eluded investigators. Even within limits imposed by current imperfect knowledge of the physiology of pain, more collaborative research on recall would improve the completeness and accuracy of clinical diagnostic interviews, insurance adjudications, morbidity statistics, and health survey interviews.

Female↗
PubMed1985

Cognitive aspects of health surveys for public information and policy.

Health survey data are an important and efficient source of information for policy makers and administrators. But caution is warranted: surveys do not show cause-and-effect relations, and they are no substitute for randomized controlled experimentation in predicting behavior. The variety of surveys--governmental and private--is increasing, and both methodology employed and interpretation of results can be improved in suggested ways.

Adult↗
PubMed1978

Effect of fetal monitoring on neonatal death rates.

We analyzed data from 15,846 live-born infants to assess the effect of electronic fetal monitoring on neonatal death rates. The crude neonatal death rate was 1.7 times higher in unmonitored infants than in those monitored. Adjusting for inherent risk and changes in mortality rates and monitoring rates during the years of the study lowered the relative risk to 1.4 (95 per cent confidence interval, 0.85 to 2.45). The estimated yield from monitoring decreased as the inherent risk of the baby declined. Thus, in the highest-risk group 109 lives might be saved for every thousand babies monitored. In the lowest risk group (babies at term with no risk factors) the neonatal death rate is around one per thousand. The absolute benefit for this large group could therefore not exceed one life saved for every thousand babies monitored.

Female↗
PubMed1975

A differential viability model for twin-pair blood group data.

Much interest in human genetics studies of blood systems is focussed on the possible incompatibility of discordant twin pairs. Here we discuss a differential viability model which accounts for deviations from the standard multinomial model for twin pair data based on the Hardy-Weinberg frequencies. We discuss related estimation and testing problems, illustrating the techniques with data from the Louisville Twin Study.

Alleles↗
PubMedDate not supplied

Statistical perspectives on confidentiality and data access in public health.

Confidentiality and disclosure limitation are topics that are inherently statistical but, until recently, they have received limited attention from statistical methodologists. That situation has changed considerably in the present decade. In this paper, we provide an introduction and overview of some statistical disclosure limitation issues that are of special relevance to public health studies and surveys, and the linkages to current research on bounds for multi-dimensional contingency table entries and 'simulated' categorical data. We also describe how these research methods relate to a new data access query system being developed for use by NCHS and other statistical agencies.

Confidentiality↗