P Armitage et al (eds)
Published by Blackwell Science (2002)
794 pages plus index.
This is the fourth edition of this text. Again, it is a comprehensive volume from three eminent authors. This text should be available in all workplaces that deal with epidemiological data. It provides not only a clearly written theoretical approach to statistical analysis, but also practical examples that the readers can use to get them started in their work. I would have liked to see more examples to assist the readers in grasping the techniques, but numerous references to other texts may help in this regard. This is a weighty volume at just over 800 pages, and should therefore be treated as a reference book. The comprehensive index guides the readers to techniques that are appropriate for their situation.
The introductory chapters should be read by all as they remind readers of the basic approaches to analysis that are applicable to more complex techniques. Chapters on probabilities, analysing count data, regression, correlation and comparative methods are all as one would expect. The new chapter on Bayesian methods is worth reading and the chapters on clinical trials, survival analysis and statistical methods in epidemiology are very useful to those in the cancer epidemiology and trials field. Throughout the book there are numerous practical tips in the application of statistical techniques contained in the discussion.
The book is well laid out, with formula referencing, clear presentation of algebraic characters, good graphs, tables and highlighted examples. The book is very dense with text, takes careful reading and covers the topics comprehensively.
In summary, the authors state their purpose as “to guide the medical research worker with no particular mathematical expertise but with the ability to follow algebraic formulae, and more particularly the concepts behind them”(xi). I think that in this volume they go a long way to achieving this purpose. At a price of just under $200 it is a valuable investment.