Children
For Patients and Families   Your Child's Health   Clinical Services   |   For Health Care Professionals   Medical Education   Medical Research
Stats

My ten favorite books, #8 (March 24, 2006). Category: Good books

A recent query on the MedStats email discussion group reminded me that I need to continue the list of my ten favorite statistics books. The list so far is:

  • #10 Essential evidence-based medicine. Dan Meyer
  • #9 Understanding Variation: The Key to Managing Chaos. Donald J. Wheeler

Another book I really love and recommend all the time is

  • How To Report Statistics in Medicine. Thomas A. Lang, Michelle Secic (2006) Philadelphia, PA: American College of Physicians. [BookFinder4U link]

The 2006 publication is the second edition, which I have, but which I have not had time to review in detail yet. The first edition, published in 1997 is excellent.

A nice web page from the publisher

outlines some of the changes in the second edition. The publisher also includes commentary that outlines the purpose of the book (referring to the 1997 edition):

This book is easy to refer to when trying to find the information you need. Although it doesn't replace a biostatician or a thorough understanding of a subject when conducting research, it allows quick scanning of a topic, picking up some tips on how to present data and check your work, while highlighting potential problems and leading the reader via well-referenced sources to more in-depth review. It also makes excellent use of graphs, and has nice examples of common tests, their application, and their presentation. www.acponline.org/journals/impact/aug97/review.htm

 A review by Nadine Martin in the Journal of the American Statistical Association (Vol. 93, No. 441, (Mar., 1998), pp. 409-410) had the following comment

Lang and Secic do a masterly job of taking a subject that intimidates many people and sweetening it so that it is palatable. This book should be on every medical writer's desk (and many authors would benefit from it too) to be read from cover to cover and used as a reference. I also recommend that it be used as a text for journalism students and science writers, or by anyone who does not plan to become a statistician yet needs to be able to interpret and report statistics.

A review by Ivan Kre'imir Luki' (my apologies if this name gets mangled by my web translation) in the Croatian Medical Journal (December 1998, Volume 39, Number 4) had the following conclusion:

Although there are numerous statistical handbooks, textbooks, and manuals, a book like this is hard to find. The authors have provided a step by step, fully detailed guide to statistical issues addressed in each of main parts of a scientific manuscript. It is important to stress that this is not a classic statistical textbook. Rules and calculations of particular statistical tests are not included. Instead, it will help the readers to understand frequently used statistical methods and to avoid common mistakes. I cordially recommend this book to everyone: scientists, editors, peer reviewers, and students. It will help the authors to prepare their manuscripts for publishing, to give their reports a really 'professional' look, to avoid the most common 'traps', and to satisfy even the most demanding editors (e.g., the editors of the Croatian Medical Journal). It will help the editors to improve the quality of their journals. It will show reviewers what to look for in a manuscript. It will help students to understand basic concepts of statistics and use of the most common statistical methods and tests. I had a lot of fun reading and hope you will, too. www.cmj.hr/1998/39/4/390421.htm

A review by Stephen Scott in the Journal of Child Psychology and Psychiatry and Allied Disciplines (Vol. 40. No 4, p 662, 1999) starts out with a ringing recommendation:

This is the best book on statistics I have ever read by a long way. For anybody writing a research paper, or trying to evaluate one, either for their own sake or because they need to comment on its quality to somebody else, it is invaluable. What is so good about it is that it tells you precisely what to do in easily digested, step-by-step gobbets of information. There are bullet points in bold followed by an easily understood paragraph or two.

This book is just about on everybody's "resources for researchers" list. On Betty Jung's annotated biostatistics, she has the comment

FINALLY, a guide on how medical research should be reported in the literature!! With this text in print, now there really is no reason for poorly written research reports to be in print, nor for any medical researcher not to know how to write up their research in an appropriate fashion. And for research consumers - this text will tell you what you should be looking for when you read the literature. www.bettycjung.net/Biostats.htm

Related weblog entries:

07/08/2008.

CMH Employees