This fifth edition has undergone major restructuring, with some sections completely rewritten; it is now more logically organized and more user friendly (with the addition of 'summary boxes' throughout the text). It incorporates new statistical techniques and approaches that have made an appearance since the last edition. In addition, some chapters or chapter headings are specifically marked to signify material that is more difficult than the material in which it is embedded - such sections or chapters can be omitted at first reading.
Several new chapters have been added . "Associations: Chance, Confounded and Causal?" explains without any formulae the concepts underlying confounding, confidence intervals and p values, and the interpretation of associations observed in research investigations. Another new chapter considers sample size calculations in some detail and provides, in addition to the relevant formulae, useful tables that should give the researcher an indication of the order of magnitude of the number of subjects he or she might require in different situations.
