- Paperback: 608 pages
- Publisher: Manning Publications; 2 edition (June 6, 2015)
- Language: English
- ISBN-10: 1617291382
- ISBN-13: 978-1617291388
- Product Dimensions: 7.8 x 1.5 x 9.2 inches
- Shipping Weight: 2.3 pounds (View shipping rates and policies)
- Average Customer Review: 53 customer reviews
- Amazon Best Sellers Rank: #28,340 in Books (See Top 100 in Books)
Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your mobile phone number.
R in Action: Data Analysis and Graphics with R 2nd Edition
Use the Amazon App to scan ISBNs and compare prices.
"Rebound" by Kwame Alexander
Don't miss best-selling author Kwame Alexander's "Rebound," a new companion novel to his Newbery Award-winner, "The Crossover,"" illustrated with striking graphic novel panels. Pre-order today
Frequently bought together
Customers who bought this item also bought
Customers who viewed this item also viewed
About the Author
Dr. Rob Kabacoff is a seasoned researcher and teacher who specializes in data analysis. He also maintains the popular Quick-R website at statmethods.net.
Author interviews, book reviews, editors picks, and more. Read it now
Top customer reviews
There was a problem filtering reviews right now. Please try again later.
The changes from the first edition are helpfully summarized on page xxvii, but I will distinguish five strands. First, there is a nod to the computer-science (vs. "proper" statistics) sensibility, via addition of a chapter on clustering and another chapter on classification, discussing CART and SVM methods. Second, two new chapters deal with R as a programming language, with one chapter dedicated to package-building. Third, there is a new chapter on producing reports, focusing on, but not limited to, the capability of "rmarkdown" R package. Fourth, second edition boosts coverage of "ggplot2" package. Finally, statistics repertoire is expanded with a short chapter on (very basic) time-series analysis, relying on "forecast" R package.
I may have wished for a few different choices. The package-building chapter could have been dropped, and more space given to either statistics (one could say more about regression - maybe taking a page from "Modern Regression Techniques Using R" by Wright and London) or machine learning. I would introduce RStudio from the outset, not in an appendix. I would say goodbye to R's base graphics, and, like Lander, embrace "ggplot2", instead of relegating it to Chapter 19. I would give space to "data.table" - like "ggplot2", this is a necessity - and I would give more space to "plyr"-and-friends. Turning to Manning, I would ask them why Addison Wesley could publish "R for Everyone" in color, and still charge only 60% of the price of the black-and-white "R in Action". I would also draw their attention to a typo ("Stark Trek"?) in line 5 (!) of the book's preface. This slip-up is odd considering how long the book was delayed.
This criticism aside, "R in Action" is an excellent, well-thought-out and well-presented book, with second edition offering incremental but important changes from the first. In the field of "generalist" R books, two titles stand out - "R in Action" and "R for Everyone" - and "R in Action" is the heavier part of the duo.
UPD. With the benefit of a little more life experience, I would say: don't spend your time on *any* R book. Python is the way to go.
As a bit of background, I am a relative novice in R and, though a statistician, I actually had little need for R until recently. I’m mid-career and, to be honest, learning R seemed a difficult challenge. I now know I have missed out on a lot not getting into R sooner, and this book has aided my progress more than any other resource. The pros of this book are not mutually exclusive, so I apologize that they are not in a list. It’s just that one pro relates to another.
This book is very comprehensive. It walks you through all of the issues a researcher faces when he or she tackles real data. We are often not presented with cleaned data and we have to recode, aggregate, restructure, etc. Many books walk you through this process, but this one does so in a novel way. Rather than merely present code, it explains what form code for a particular task takes, why you would need it, runs it, and then walks you through it again line by line. My main concern with other texts is not their fault, but my own. I can’t just read code. Not that I don’t understand it when I read it; rather, I tend to gloss over it (like my students do with math equations). This book presents the code and then, with numeric labels for lines, walks you through it piece by piece and line by line. I find this method works better to keep my attention. It also does not skip pieces of code like other texts. Sometimes I have wondered what a very specific piece of a code does, but, perhaps because it is a minor detail, other texts fail to clarify. Consider:
“fit <- lm(weight ~ height, data=women)
produces the graphs shown in figure 8.6. The par(mfrow=c(2,2)) statement is used to combine the four plots produced by the plot()”
Other texts often would have left me wondering what “par” did because this specific feature was not central to the topic at hand. This might not be the single best example, but sometimes authors of other texts seemed to assume I knew things I did not about why they ran code a specific way.
The book covers basic and some advanced statistics as well as data cleaning. If the book covered multilevel models, longitudinal data (as a special case of MLMs), and SEM, I would probably never need another book on R. Realistically, that is a ridiculous wish list, but the writing and presentation are so good that I REALLY wish these methods were covered too. One caveat, the book does not try to teach you statistics. It assumes you are familiar with them and want to apply your knowledge to R. However, it does not completely gloss over the issues either. For example, it reminds you of what OLS regression is, what the assumptions are, and how to run/test such models. If you want to learn statistics, this is not the book for that, but it does not pretend to be either. A good companion might be Andy Field’s Discovering Statistics in R, which goes into the details of basic statistics. Please do not interpret this as me saying the book falls short of covering essential topics – it does not. Take a look at the list of topics in chapters and in the appendices; it’s rather impressive! And the appendices should not be skipped.
All the code is available to download from the publisher’s website. This, in itself, is not novel to this text, but when you combine the code with the level of detail and step-by-step instruction from the text, you have an amazing resource. I can easily copy code from these documents and paste in my own variables. This saves me a great deal of time that I use to spend reading through long manuals for each R package and trying to understand what settings I needed. This is especially important given the fact that the text explains how to test model/statistical assumptions (often using multiple methods).
I think the most favorable single conclusion I can provide for a possible reader of this text is this: Had I picked this book first, I would have saved myself a great deal of frustration with R. Some people say R is not that hard to learn. Maybe I am obtuse, but I disagree. Coming from a background in MPlus, SPSS and SAS, some of the code in R seems arbitrary and nonsensical to me. This book does the best job I have seen of placing that nonsensical code in an interpretable framework. This book goes on the list of texts for which I feel extreme gratitude for the author’s hard work.