Statistical Science in the Courtroom, edited by Joseph L. Gastwirth. New York: Springer, 2000.

The use of scientific and statistical evidence, and related expert testimony, has become commonplace in the American courtroom. While lawyers have increasingly relied upon such evidence and testimony, a major concern for the justice system has been how to ensure the proper interpretation of that evidence. Statistical Science in the Courtroom is a collection of articles and research that addresses an array of legal, statistical, and technical issues involved with the use of science and statistics in the judicial process.

Although the back cover of Statistical Science in the Courtroom describes the contributors as "statisticians and legal scholars," and the book is part of a series titled "Statistics for Social Science and Public Policy," those interested in this subject matter should not be misled: most of the twenty-two chapters in the book are written by statisticians, biostatisticians, and mathematicians. According to the biographies at the beginning of the book, nineteen of the twenty-four contributors hold degrees or professional appointments in technical fields-not in law, public policy, or a social science. This is not to say that the book will not be instructive, informative, or useful for lawyers and social scientists but, rather, caveat emptor-much of this book is not for the mathematical faint of heart. It certainly does address substantive applications of statistics in the courtroom, but it is not merely a book of applied statistics with the courtroom as a setting. Many of the chapters are highly technical and demand of the reader a fair degree of statistical sophistication.

That caveat aside, Statistical Science in the Courtroom offers a fascinating and instructive look into the role statistics play in the modern legal process, and even readers without PhDs in statistics and non-methodologists will find portions of the book interesting and accessible. The collection of articles covers an impressive range of substantive legal issues as well as statistical quandaries. According to the editor, Joseph L. Gastwirth, "Most of the books on statistics in law describe the methods used to analyze data rather than on how involvement in the area of legal statistics affects statisticians and their research programs. The authors of the articles in this book were asked to write about cases in which they were experts or consultants or about the areas of law that rely on quantitative research or how their interest in the use of statistical evidence in legal cases has influenced their research. They were also encouraged to give their views on whether the courts give appropriate weights to statistical testimony" (p. v). While broad, this is a fair description of the book as a whole.

There is no obvious organization to the book; it is not organized by legal area or statistical technique, for instance. The different chapters of the book explore, in no particular order, the use of statistics and scientific evidence in litigation involving DNA evidence, environmental and toxic torts litigation, product liability, discrimination, monetary damages, warranty contracts, tobacco litigation, jury selection methods, the death penalty and deterrence, sentencing issues, pornography, fraud, and drug trafficking. Several chapters address as a general topic the role of statistics in the legal process. The statistical issues raised in the book range from general probability theory to sample size and reliability issues to discussions of specific analytical methods. While all of the subjects warrant some attention, a single book review cannot possibly do each chapter justice. Accordingly, I consider only a selection of the wide range of issues covered in the book here.

The most common subject addressed in this book involves, not surprisingly, DNA evidence. Five of the twenty-two chapters explicitly address the interpretation of DNA evidence. …