Book Reviews

Prove It With Figures: Empirical Methods in Law and Litigation, by Hans Zeisel & David Kaye, New York, NY: Springer-Verlag, 1997, ISBN 0-387-94892-9, 353 pp., $64.95
International Statistical Institute D.J. Hand 1988
Journal of the American Statistical Association Bruce S. Wier 1998
Jurimetrics staff 1998
Public Opinion Quarterly David L. Sills 1998
Zentralblatt MATH R.V. Lenth 1999

International Statistical Institute

D.J. Hand, Book Review, International Statistical Institute Vol. 18, No. 1, Apr. 1998

The objective of this book is to explain statistical concepts which are useful in legal situations without a formal statistical background. It is devoid of mathematics (although not numbers), which is entirely appropriate for its intended audience. It includes over twenty case histories, spanning a wide range of areas and issues, and including analyses of the death penalty, jury selection, mass torts and vaccine evaluations. Such a book is very timely. There has been a recent growth of recognition of the relevance and applicability of statistical arguments in legal fields -- and also of the unfortunate ignorance about this amongst those in the legal profession. The book deserves, and even needs, to be brought to the attention of law students and the judiciary.

Apart from the evident value this book will have for its intended audience, it makes stimulating light reading for the statistician keen to see statistical ideas in relatively novel situations. Indeed, the appearance of this book makes one wonder if legal applications of statistics might someday have an impact on the development of future statistical methodology, in the same way that agricultural and medical applications have had in the past.

Journal of the American Statistical Association

Bruce S. Wier, Book Review, Journal of the American Statistical Association, Vol. 93, No. 9, 1998, p. 1250

Although this book is aimed at presenting basic statistical ideas to lawyers, and does so very well, the statistician who is called upon to present expert testimony will the legal arguments and case citations of great value. Law Professor Kaye is unusually well-versed in statistical thinking and has contributed substantially to the field of parentage testing and the use of DNA evidence. (Zeisel is now deceased.) As expected from such a background, the book stresses the role of statisticians in determining the probabilities of evidence under alternative hypotheses (guilt or innocence) rather than determining the probabilities of the hypotheses being true. The book covers several other areas as well, including discrimination cases and product liability based on epidemiological studies. The authors discuss the case involving Bendectin, leading to the 1993 Supreme Court decision in Daubert v. Merrell Dow Pharmaceuticals. Statisticians need to be aware that the Court agreed that trial judges should exclude scientifically unreliable testimony by experts, and should inquire into the rate at which the reasoning used by an expert produces erroneous conclusions. It seems, therefore, that the role for statisticians can only increase and that this book will prove a valuable resource.


Jurimetrics: The Journal of Law, Science, and Technology,
Winter 1998, Vol. 38, No. 2, pp. 239-240

Prove It with Figures is an excellent primer for the legal professional who has little previous acquaintance with statistics and empirical methods. The book begins with brief but understandable explanations of various types of studies used to analyze questions a researcher might have. It then explains some statistical methods for addressing those questions. The book acquaints readers with basic principles of statistical analysis without overwhelming them with detail.

Yet, Prove It with Figures does not leave readers with only a dry skeleton of statistical methods. Instead, the work uses lively and often contemporary issues, cases, and studies to bring its exploration of these topics to life. Some of the contemporary examples include silicone breast litigation, Agent Orange litigation, domestic violence studies, and studies on the death penalty's effects on violence. The book even demonstrates how one can analyze the authorship of essays in the Federalist Papers using statistical methods.

As the book progresses, it focuses on the use of statistics in trademark litigation, jury selection, and litigation involving DNA. It also provides a handy glossary of terms for the topics discussed. This book offers a compact and intense exploration of this field. But taking the time to read Prove It with Figures is well worthwhile for those who wish to develop a familiarity with empirical and statistical methods.

Public Opinion Quarterly

David L. Sills (Social Science Research Council), Book Review, Public Opinion Quarterly, Summer 1998, pp. 287-289

Hans Zeisel's first research publications were written in prewar Vienna, when he was a somewhat younger colleague of the also young Paul F. Lazarsfeld. Die Arbeitslosen von Marienthal (1932; English edition 1971), by Marie Jahoda, Lazarsfeld, and Zeisel, is the most enduring product of these years.

When the 1938 Anschluss forced his emigration from Austria, Zeisel became a leading practitioner of scientific market research in New York. His methods book, Say It with Figures (1947), based on research conducted in both Vienna and New York, became a bible for students and researchers alike, and was eventually published in six editions and in seven languages.

Zeisel had been trained in the law in Vienna, and in 1953 he joined the law faculty of the University of Chicago. Here he began a distinguished career in legal research, focusing on the jury as a central legal institution and on the use of data from statistics and surveys as evidence in the courts.

Prove It with Figures is intended primarily for law teachers and law students; its didactic nature is demonstrated by its frequent summaries and its lists of "Critical Questions" at the end of chapters. But it is more than a textbook and more than an exposition of research methods appropriate to the law, and for these reasons it should command the attention of all public opinion researchers interested in new methods of study design. Its title both recalls the earlier Say It with Figures and sets the stage for much of the new volume's content. That is, the book is both an extension of the methods described in Say It with Figures and a compilation of examples of these and other methods largely as applied to research into law and litigation.

Zeisel had planned this book for many years, but his teaching, research, and consulting activities prevented his completing it before he died in 1992. Fortunately, his colleague David Kaye, a professor on the law faculty of Arizona State University with a strong background in statistics, has brought the project to a very satisfactory completion. Most of the book consists of drafts left by Zeisel at the time of his death; three chapters, on "Epidemiologic Surveys," "Coincidence and Significance," and "DNA Profiling," were written largely by Kaye.

The organizing thesis of the book is the problem of determining causation—causation defined not in a philosophical sense but rather intended for the practical needs of dispute settlement. By combing the research literature for examples of how social research has been used by the authors and by others to discern causation, Zeisel and Kaye have come close to writing a handbook for general social science research.

Much of the text of Prove It with Figures consists of summaries of research undertaken to convince prosecutors, judges, and juries that such-and-such actually happened, that some sequence of cause and effect "proves" (or "disproves") the position taken by one side in a legal dispute. Examples are studies of the health impacts of taking an antinausea medication (pp. 62-64), of receiving a breast implant (pp. 53-57), of smoking (pp. 57-60), of exposure to asbestos (pp. 60-62), and of exposure to low-frequency-electricity transmission lines (pp. 64-68). Other examples are studies of alleged trademark violations, in which the central problem is to learn whether or not the public confuses the name of a new product with a similar trademarked name (pp. 147-74). These are all problems of interest to public opinion researchers, and the study designs described in this book have wide application in the field.

Still other research summarized in the book concerns the efficacy of different aspects of the legal process itself. Examples are the outcomes of jury trials versus those of trials in which verdicts are handed down by judges (pp. 14-15); the frequency of "not guilty by reason of insanity" verdicts when differing instructions are given to jurors (pp, 16-21); the deterrent effect of capital punishment (pp. 39-41); and the impact on ultimate verdicts of moving the locale of trials to communities where there has been less pretrial publicity (pp. 138-44).

Many of these complex issues cannot be studied by the use of either opinion polls or randomized controlled experiments alone. Accordingly, various forms of simulation and what has come to be called "triangulation" are used. (Triangulation is introduced as a term borrowed by the philosopher Herbert Feigl from the techniques of surveying and navigation to designate any scientific effort to approach the truth of a proposition through more than one research channel.) This may take the form of comparing the results of studies based on hypothetical cases with those of studies based on limited data from actual cases, or of comparing survey data with observational data. (It is perhaps relevant that in the 1930s study of Marienthal cited above, Zeisel and the others learned how unemployment led to apathy by comparing interview results with data showing a decline in both the circulation of library books and the sale of political newspapers.)

A number of examples of the use of content analysis are given. One is a case in which the Ford Motor Company was charged with negligence after a rear-end collision between a Ford Pinto and a van caused the Pinto gas tank to explode, killing the car's three occupants. Through a content analysis of the local anti-Ford publicity, the judge became convinced that a fair trial was impossible, and a change of venue was accordingly obtained (pp. 128-32). And content analysis was used by Frederick Mosteller and David L. Wallace to provide evidence in a dispute over the authorship of 12 of the pseudonymous Federalist Papers of 1787 and 1788. By examining the frequency of use of various marker words, the researchers determined that in all probability James Madison wrote 11 of the 12 disputed papers (pp. 132-34).

The instructive value to law students of a methods book largely based on empirical research into law and litigation is obvious. Using this book will also enhance their exposure to the important but still-struggling law-and-society movement in law schools. And Prove It with Figures serves as an important reminder to all public opinion researchers that research on an issue of public concern need not be limited to obtaining interview or polling data. Simulated experiments, observation, triangulation, and content analysis are all useful components of an imaginative and comprehensive research plan.

Zentralblatt MATH

R.V. Lenth, Zentralblatt MATH, Sept. 2, 1999

This is an introductory statistics text for lawyers and law students, patterned to some degree after the late Zeisel's 1947 classic, Say It With Figures. There are about 220 pages of exposition followede by extensive notes, a glossary, and indices.

The book begins with a chapter discussing types of studies -- controlled randomized experiments, observational studies, and epidemiologic studies -- then moves on to discussions of significance testing, sampling, and content analysis. So far, these are fairly standard topics, but they are interwoven with numerous interesting examples concerning hung juries, product-liability suits, and so forth.

The subsequent chapters are more specialized toward legal topics. These include "Surveys and Change of Venue," "Trademark Surveys: Genericness," "Trademark Surveys: Confusion," "The Jury: Composition and Selection," and "DNA Profiling: Probabilities and Proof." Each chapter is provided with great case studies. Example: How do we design a survey to answer the question of whether a proposed hotel chain named "McSleep Inn" is too likely to be confused with a well-known hamburger chain? (The motel chain lost the suit and was enjoined to pick another name.)

This book is terrific reading for anyone interested in the use of statistical methods in society -- lawyer or not.