R for Everyone: Advanced Analytics and Graphics (Addison-Wesley Data & Analytics Series) 2nd Edition, Kindle Edition
Use the Amazon App to scan ISBNs and compare prices.
- Highlight, take notes, and search in the book
- Create digital flashcards instantly
See the Best Books of 2018 So Far
Looking for something great to read? Browse our editors' picks for the best books of the year so far in fiction, nonfiction, mysteries, children's books, and much more.
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.
Would you like to tell us about a lower price?
Top customer reviews
There was a problem filtering reviews right now. Please try again later.
Where "R for Everyone" differs from "R in Action" - and, coming to the positives, where it wins out - is in intermediate-R territory. One important example is coverage of "ggplot2". Whereas "R in Action" discusses the "old school" R graphics, "R for Everyone" goes with "ggplot2", becoming the second popular book (after Winston Chang's "R Graphics Cookbook") to discuss the package - and although its explanation of "ggplot2" syntax is sketchy, the samples found throughout the book do build into a useful "ggplot2" gallery that actually brought me over the fence. "plyr" package, an important data-manipulation aid, is another example, and another "R in Action" no-show. So is "data.table". So is "knitr", used to produce reports. So is "rcpp", used to interface R and C++. So is R package-building. (You will notice that the topics become more advanced. These are introductions rather than substantial explorations, but awareness is a valuable thing). In the book's second half, when discussion moves from R to statistics-with-R, the author continues to manage to find original material; statistical explanations may be brief - this is not a textbook - but examples, and pointers to useful R utilities, are much appreciated.
I own just one R book - literally, "The R Book", by Crawley - but "R for Everyone" will be joining it; this has got to be a compliment. Kudos to Jared Lander for writing an original, substantial, useful book.
UPD. It's June 2015, and second edition of Robert Kabacoff's "R in Action" is finally out - but the changes are incremental, and my endorsement of "R for Everyone" stands.
The book has a nice layout in color, but this is misleading. Long R output without proper formatting for a textbook is always displayed because the author wrote the book directly in the code as he himself states and printed it out as it is. And it feels like. Most of the text looks just like comments in a program code. The treatment of functions is very poor (they are also very rarely used in the book) and the explanation of the different R data types lacks depth and is misguided. Silly examples are used to show the basics as in printing the author's name. The later chapters get even worse, literally damaging all the more interesting parts, where the book leaves the very basics and moves on to data handling and then to advanced data analytics in R.
The part of the book that deals with data analytics is sincerely a bit of a tragedy. Rushed text with no clear or sometimes whatsoever explanations of what is actually being done, with just little text and lots of code output and charts taking most of the space. Ironically the book that is "for everyone" makes hard for "everyone" to understand anything that uses statistics, about 60% of the book!.
It is harder even for those trained on statistics or related "hard" sciences.
For example, In chapter 22, right in the beginning the author uses a value for the predicted number of clusters in the data under analysis. This value is taken out of the blue and only later it is shown how this value can be found using two methods. The first method doesn’t bring any useful value (and you wonder why it is shown). The second method does bring a good value but it is not explained in the text how this method's results should be interpreted to determine this value. Apart from two rushed sentences that speak of a standard deviation being used, whatever this standard deviation is coming from as the author says nothing about the algorithm clusGap that he used for such. I did some research and found out that the author's LiveLesson video course, that follows the book almost page by page, does mention, albeit quickly, how to interpret the second method’s result above. But not in his book… Unfortunately this video course also suffers from the same problems that the book does, as it is mostly a live reading of the book with the author typing the code.
In fact, almost anything related to data analytics is very poorly explained, if at all. Another example, out of many, is the section 20.3 on Generalized Adaptive Models. After preprocessing the raw data used for the analysis, a few charts are displayed (without much explanation of the code used for which) and then the data analysis code output is shown without any explanation. Two features of the data, CreditAmount and Age, are displayed in charts where they are smoothed, but there is no explanation about what for. And the analysis stops right there without any further explanation. What could be said in a few sentences is left out.
Most of the data analytics examples also show very poor performance, leading the user to think why data analysis is used if it performs so badly and, if it performs well, why the author didn’t select any better example.
There are also many pedagogical errors, minor and major ones. I will just mention a few taken from chapter 12 as an example:
1) Many variables are created with the function assign in a loop but actually only two of them are used. What for? On top of it, the same loop is coded again later with just a different variable name.
2) The function merge is used with the same column names, although the author states that “the ability to specify different column names (..) is the most useful feature of merge” before doing so with the function join from the library plyr. Then you wonder what difference is being shown.
3) It gets worse. In a rather convoluted way to show how to merge different data frames, the author introduces two new features of R, eval and parse, just by passing and without any specific examples or further explanation. In this same convoluted example the author also uses the R function Reduce in the most complicated way with the dots, without first showing simple examples and what it is for. Only then later down in the text he goes on to explain what Reduce does but fails to mention that it can only be applied to binary functions. The text states that “Reduce can be a difficult function to grasp”. If it is, it would deserve a better treatment, not as a side note, explained in an example that is related to something else (how to merge data frames). It should also have a full explanation of how it can be used.
R is a beautiful language that can be well explained. It is not hard to show its power in data analysis with short but clear explanations. It’s regrettable that this book misses its stated goals so badly, when it could have done otherwise brilliantly, as its author seems capable to do a much better job. So I can't recommend this book. There is actually a shortage of good R books in the market, but "R in Action" (second edition is coming), "The Art of R Programming" and "The R Book" are much superior choices.
many of the data sets the author references are not available where he references them. And others have very long and silly column names, which makes writing out code needlessly complicated. Worse still is when the column names don't match the book - this book is brand new so the columns should match. When they don't match, it makes a lot of unnecessary busy work matching up the author's terms with the actual data set. D for effort. These mistakes would have been easy to catch and clean up.