Scientific data analysis

A practical guide to scientific data analysis

David Livingstone Chichester, UK: John Wiley 2009 | 358pp | ?50.00 (HB)

ISBN 9780470851531Reviewed by Jeremy Frey

Displaying data, modelling data, developing and fitting models, and estimating uncertainties are all crucial aspects of understating experimental results. Many old and new techniques have been facilitated by the ready availability of computer power and quality graphical displays. The range of possibilities can be confusing even to those regularly analysing data.

This book is a guide to the wide range of methods available. Not surprisingly given the author’s background, the examples in the book are all chemical and hence it will be of most interest and value to chemistry researchers.

The book helpfully starts with a description of what is meant by ’data’ and introduces some of the ideas and terms used in data analysis. Importantly, a step back from analysis is taken in the discussion of ’design of experiments’, as decisions made before any data is collected can have a major influence on subsequent analysis. 

The stages of data analysis, from initial data reduction, data display, followed by model fitting in terms of clustering, regression, supervised-learning, a wide range of multi-linear techniques and neural networks are discussed at an accessible level of detail. 

The book finishes with a chapter on molecular design. The format of the book gives a very dense impression of text and the feeling of compressed graphics, however, the material is clearly set out and very readable. The 10 chapters contain plenty of well referenced example data which as well as illustrative of the techniques will be a great source of examples for those teaching in this area.