Applying Statistics in the Courtroom, Robert D. Koons, April 2003
April 2003 - Volume 5 - Number 2
Applying Statistics in the Courtroom
Phillip I. Good
Chapman & Hall/CRC, Boca Raton, Florida, 2001
Robert D. Koons
Counterterrorism and Forensic Science Research Unit
Federal Bureau of Investigation
The author's stated purpose in writing this book is "to ensure attorneys and statisticians will work together successfully on the application of statistics in the law." To this end, he has aimed this book squarely at these two groups. The book is written at a level appropriate for those statisticians and lawyers who know very little about the other's discipline. This text is largely successful at providing an introductory overview to the attorney and the statistician which will prepare them for entering the unknown territory of the other group. The approach in this book is to use selected citations from court opinions to illustrate various points concerning acceptance (and rejection) of statistics in the courtroom. For the statistician who is unfamiliar with legal proceedings, the book provides specific and sound advice on methods for presenting statistical concepts in terms that a judge would understand. Similarly, attorneys who are seeking to use supporting statistics in a variety of legal settings will find many of the ideas presented useful. However, statisticians who are looking for technical details concerning legal use of statistics or attorneys who need an exhaustive review of pertinent cases for preparation of briefs will likely be disappointed by this book. The strength of this book is that it contains practical and useful advice on the use of statistics and probability in such areas as jury selection, discrimination, trademark disputes, criminal law, civil law, and product liability.
The book is divided into four parts and fifteen chapters. The first part, entitled Samples and Populations, consists of four chapters concerning methods of sample selection and the use of descriptive statistics. Chapter 1 discusses the definition of samples and methods of selecting representative samples from the point of view of the court. Chapter 2 provides a legal definition of the representative random sample. Chapter 3 compares sampling methodologies and contains useful advice concerning the proper use of surveys. Chapter 4 discusses the typical methods for presenting easily understood descriptive statistics. The statistician might find Chapter 4 oversimplified until he or she realizes the limited level of detail that can be understood by most jurors and some attorneys or judges.
The next four chapters form a section entitled Probability. Chapter 5 provides a brief, non-mathematical introduction to probability. Chapters 6 through 8 describe the various approaches to the use of probabilities that have been accepted in civil, criminal, and environmental cases, respectively. There is a major emphasis throughout these chapters on the difference between fact-based probabilities and estimates. The discussion in Chapter 6, Criminal Law, of the use of probabilities and particularly the legal requirements for multiplication of probabilities in the well-known People v. Collins case and the Wayne Williams trial, is particularly interesting. Those statisticians unfamiliar with courtroom procedures will either be fascinated or confounded by the variations in what has been accepted or rejected by various courts, and they would do well to heed the warning that "the law is what judges say it is."
The third group of four chapters is entitled Hypothesis Testing and Estimation. Chapters 9 and 10 discuss how large a sample must be from the perspectives of the court and the statistician, respectively. Chapter 10 contains a brief overview of simple statistical procedures for testing hypotheses. Chapter 11 describes correlation and regression of two variables. Chapter 12 provides a similar discussion of multiple variables. The emphasis in this chapter is on the legal acceptance of multiple regression methods in cases of alleged discrimination.
Many readers will find the final section, Applying Statistics in the Courtroom, to be the most practical and useful portion of this book. Chapter 13 describes preventative actions that can be taken to reduce the need to go to the courtroom and challenges that can be presented when questionable statistics are used by the opposing expert. Chapter 14 provides the statistician with a blow-by-blow description of the trial process, should the procedures in Chapter 13 fail to keep him or her out of the courtroom. Chapter 15, Making Effective Use of Statistics and Statisticians, is written for the attorney to assist in making proper and effective use of statisticians and statistical evaluation of evidence. This chapter is particularly useful in that it contains a list of questions that lawyers can use during discovery and to guide the testimony of a statistician.
This reviewer is a forensic chemist, and because many of the readers of Forensic Science Communications are forensic scientists, a review of this book from that perspective seems in order. I opened a book with the title Applying Statistics in the Courtroom hoping to find practical advice on technical methods for the statistical interpretation of physical evidence. Dr. Good is a recognized expert with excellent credentials as both a statistician and a scientist. However, he chose to target this book at an audience that does not include the forensic scientist. Forensic scientists might find this book interesting in that it applies statistical concepts that we frequently use to situations that are likely to be unfamiliar to us, such as the application of sampling concepts to jury selection. However, this book will disappoint those hoping to find enlightenment as to how to apply statistics to thorny probability problems concerning evidence interpretation.
Applying Statistics in the Courtroom is probably most useful to attorneys who can use the lists of questions and advice on the application of statistics to make effective use of statisticians in writing briefs, to guide pretrial conversations with their statisticians, and to produce effective courtroom testimony that is understandable to judges and jurists. Statisticians will find the advice given to be useful until they have entered the legal arena a few times, either by testifying, assisting in the writing of briefs, or preparing cases. After that, they will likely find what approaches work best for them and might refer back to this book only when they are faced with a type of case that they have not previously encountered. This book is easily readable and follows the style of the other familiar CRC publications on various aspects of criminal sciences.