S. James Press

Judith M. Tanur

Subjectivity of Scientists and the Bayesian Approach

Alicia L. Carriquiry

William F. Eddy

This edited volume surveys a variety of topics in statistics and the social sciences in memory of the late Stephen Fienberg. The book collects submissions from a wide range of contemporary authors to explore the fields in which Fienberg made significant contributions, including contingency tables and log-linear models, privacy and confidentiality, forensics and the law, the decennial census and other surveys, the National Academies, Bayesian theory and methods, causal inference and causes of effects, mixed membership models, and computing and machine learning. Each section begins with an overview of Fienberg’s contributions and continues with chapters by Fienberg’s students, colleagues, and collaborators exploring recent advances and the current state of research on the topic. In addition, this volume includes a biographical introduction as well as a memorial concluding chapter comprised of entries from Stephen and Joyce Fienberg’s close friends, former students, colleagues, and otherloved ones, as well as a photographic tribute.

In Memory of Stephen E. Fienberg

Statistics in the Public Interest

The social survey has become an essential tool in modern society, providing crucial measurements of social change, describing social life, and guiding government policy. But the validity of surveys is fragile and depends ultimately upon the accuracy of answers to survey questions. As our dependence on surveys grows, so too have questions about the accuracy of survey responses. Authored by a group of experts in cognitive psychology, linguistics, and survey research, Questions About Questions provides a broad review of the survey response problem. Examining the cognitive and social processes that influence the answers to questions, the book first takes up the problem of meaning and demonstrates that a respondent must share the survey researcher’s intended meaning of a question if the response is to be revealing and informative. The book then turns to an examination of memory. It provides a framework for understanding the processes that can introduce errors into retrospective reports, useful guidance on when those reports are more or less trustworthy, and investigates techniques for the improvement of such reports. Questions about the rigid standardization imposed on the survey interview receive a thorough airing as the authors show how traditional survey formats violate the usual norms of conversational behavior and potentially endanger the validity of the data collected. Synthesizing the work of the Social Science Research Council’s Committee on Cognition and Survey Research, Questions About Questions emphasizes the reciprocal gains to be achieved when insights and techniques from the cognitive sciences and survey research are exchanged. "these chapters provide a good sense of the range of survey problems investigated by the cognitive movement, the methods and ideas it draws upon, and the results it has yielded." —American Journal of Sociology

Inquiries into the Cognitive Bases of Surveys

Questions About Questions

Comparing and contrasting the reality of subjectivity in the workof history's great scientists and the modern Bayesian approach tostatistical analysis Scientists and researchers are taught to analyze their data from anobjective point of view, allowing the data to speak for themselvesrather than assigning them meaning based on expectations oropinions. But scientists have never behaved fully objectively.Throughout history, some of our greatest scientific minds haverelied on intuition, hunches, and personal beliefs to make sense ofempirical data-and these subjective influences have often aided inhumanity's greatest scientific achievements. The authors argue thatsubjectivity has not only played a significant role in theadvancement of science, but that science will advance more rapidlyif the modern methods of Bayesian statistical analysis replace someof the classical twentieth-century methods that have traditionallybeen taught. To accomplish this goal, the authors examine the lives and work ofhistory's great scientists and show that even the most successfulhave sometimes misrepresented findings or been influenced by theirown preconceived notions of religion, metaphysics, and the occult,or the personal beliefs of their mentors. Contrary to popularbelief, our greatest scientific thinkers approached their data witha combination of subjectivity and empiricism, and thus informallyachieved what is more formally accomplished by the modern Bayesianapproach to data analysis. Yet we are still taught that science is purely objective. Thisinnovative book dispels that myth using historical accounts andbiographical sketches of more than a dozen great scientists,including Aristotle, Galileo Galilei, Johannes Kepler, WilliamHarvey, Sir Isaac Newton, Antoine Levoisier, Alexander vonHumboldt, Michael Faraday, Charles Darwin, Louis Pasteur, GregorMendel, Sigmund Freud, Marie Curie, Robert Millikan, AlbertEinstein, Sir Cyril Burt, and Margaret Mead. Also included is adetailed treatment of the modern Bayesian approach to dataanalysis. Up-to-date references to the Bayesian theoretical andapplied literature, as well as reference lists of the primarysources of the principal works of all the scientists discussed,round out this comprehensive treatment of the subject. Readers will benefit from this cogent and enlightening view of thehistory of subjectivity in science and the authors' alternativevision of how the Bayesian approach should be used to further thecause of science and learning well into the twenty-first century.

The Subjectivity of Scientists and the Bayesian Approach

Intriguing examination of works by Aristotle, Galileo, Newton, Pasteur, Einstein, Margaret Mead, and other scientists in terms of subjectivity and the Bayesian approach to statistical analysis. "An insightful work." — Choice. 2001 edition.

Judith M. Tanur

Books

Your next favorite book is out there