This is a review of the second edition of ‘The R Book’ by Michael Crawley. The book aims to be a general purpose reference book on R, and I find it helpful to view the book as if it is split into two parts. The first part of the book, comprising the first seven chapters, is an introduction to the fundamentals of R. This part informs users how to input and manipulate data. The second (and larger) part of the book aims to comprehensively cover a number of statistical techniques. For example, there are chapters on generalised linear models, time series analysis and spatial analysis, to name but a few. The book contains much valuable information about reading data, reorganising data, running statistical tests, interpreting R's output, and making figures. Many examples are used throughout to help the reader fully understand the material included. An accompanying web page including the data analysed in the book is provided.
In general, this is an outstanding reference text which would be useful on the shelf of an R novice, as well as an R expert. The book is clearly and nicely written with R code helpfully written in red. The inclusion of numerous colour graphics makes the contents of the book visually appealing. On occasion, some commands are used in the book which are only properly explained in later chapters, and the book could do with a more comprehensive treatment of statistical programming and simulation. However, these minor shortcomings can be readily forgiven with reference to the books ambitious coverage. Ultimately a reader would likely require a dedicated statistics textbook or a specialised R book if one required to explore a particular analysis technique in depth, but the R Book is the perfect resource for starting the reader off in the right direction.
The main differences from the first edition are the inclusions of new chapters on Bayesian analysis and meta-analysis, as well as an updated bibliography and reference section. Much of the text from the first edition has been unchanged. However, the first edition did have a poor layout and suffered from bad typesetting. By reordering some material and improving the formatting of the text, this problem has largely been rectified in the present edition and this alone represents a significant improvement. If however you already own the first edition and are comfortable with its layout then there may be little to tempt you to buy the second edition.