The stated objective of this book is to get readers “up and running quickly” using R as an alternative to point-and-click packages; it accomplishes this objective. It also accurately states that it supplies reminders about statistics methods while not claiming to teach statistics. One large benefit of R is the inclusion of a large number of datasets. Additional datasets are available for download on the publisher’s website (http://www.apress.com/9781484201404). The examples are done using these datasets, which serves to increase comprehension as well as get readers comfortable with real data.
The book has 11 chapters plus three appendices (note: Appendix B on programming is referred to incorrectly as chapter 12 in chapter 1). Each chapter has an introduction and a summary. The internal organization and the sequencing of chapters facilitate comprehension. Backward and forward references are supplied when necessary and are appropriate.
The first chapter, “R Fundamentals,” is a gentle introduction to the R interface, including downloading and installing R, the command line, and the working directory. Directions are given for Windows, Mac, and Linux both here and elsewhere in the text. The second chapter, “Working with Data Files,” starts with creating small files directly in R and works through importing text files, comma-separated values (CSV) and tab delimited, data interchange format (DIF) files, and more. Chapters 3, “Preparing and Manipulating Your Data,” and 4, “Combining and Restructuring Datasets,” address the topic of data preparation through manipulations using R functions such as subset, cut, unique, paste, stack, unstack, and reshape. These were situations where it was very helpful to see actual data in the text and also experiment with the datasets directly in R.
At this point in the exposition, data is ready for analysis. Chapter 5, “Summary Statistics for Continuous Variables,” and chapter 6, “Tabular Data,” which focuses on categorical data (factors), illustrate the many statistical functions available in R. The book has tables relating the task to the corresponding commands. I did encounter a problem with the apartments dataset. My R environment signaled an error, apparently because of the United Kingdom (UK) currency pound sign; this was easily fixed in Excel. Chapter 7, “Probability Distributions,” shows the reader how to handle various standard probability distributions using a table, examples, and figures.
Chapters 8, “Creating Plots,” and 9, “Customizing Your Plots,” do an excellent job of instructing how to produce simple to more complex charts. They use built-in datasets, datasets from the website, and datasets produced in previous chapters. This is all explained clearly in the chapter introductions. Of course, color would have been nice, but here again the reader can experiment directly in R.
The last two chapters, chapters 10, “Hypothesis Testing,” and 11, “Regression and General Linear Models,” address the significant statistical tasks of using data for decision making and prediction. The author supplies an appropriate number of reminders on the statistics to support the reader in understanding the examples. Appendix A, “Add-on Packages,” provides specific instructions for Windows, Mac, and Linux users. Appendix B, “Basic Programming with R,” explains how to write a function and describes constructs for looping. My choice would have been to include something on the vector and matrix facilities in R. Some of these were mentioned in the text (for example, sapply in chapter 5 for component-wise calculation of means). Appendix C, “Datasets,” describes the datasets provided in the zipped file on the publisher’s site.
All in all, this is a worthwhile book for anyone doing statistical analysis.
More reviews about this item: Amazon
