ABSTRACT
In the preface, we identied the aim of this book as being to provide an introductory, practical and illustrative guide to the design of experiments and subsequent data analysis in the biological and agricultural sciences. We have provided a brief overview of basic statistical concepts and terminology in Chapter 1, and ideas of summary statistics, probability distributions and simple statistical estimates and tests in Chapter 2. The bulk of the rest of the book has introduced and developed various statistical approaches associated with designing experiments and analysing the data generated (Chapters 3 to 11) or with analysing regression models (Chapters 12 to 15). We have tried to use common terminology across these sections to emphasize that the same form of model, the linear model, underlies all of these situations. We then described some more advanced techniques. Chapter 16 introduced linear mixed models that allow analysis of models with a structural component and any mixture of factors and variates in the explanatory component, with no requirement for a balanced structure. Chapter 17 extended the regression modelling approach to allow curved responses and non-linear models, and Chapter 18 introduced generalized linear models (GLMs) that allow analysis of models with any mixture of factors or variates in the explanatory component for data with certain types of non-Normal distribution. Throughout the book we have introduced real examples, either drawn from or inspired by our own experiences of working with scientists in research institutes and university departments. Our aim has been to show how the statistical approaches in this book can be used to address a range of real-life research problems across a number of application areas.
