Customer Reviews

Data Manipulation with R (Use R!)

5 star:		(9)
4 star:		(2)
3 star:		(2)
2 star:		(1)
1 star:		(1)

 

 

This book along with Jim Albert's should be read by every statistician that does a lot of statistical computing. Both books help you learn R quickly and apply it to many important problems in research both applied and theoretical. Albert emphasizes applications in Bayesian statistics whereas Spector is teaching how to do data manipulation, things like merging and transposing data sets. These techniques can be easy to do in a language like SAS after a little training but in other programming languages it can be very difficult.

This concise 150 page book contains a wealth of information, writen clearly and with many well-chosen examples. I liked it a lot. It covers reading and writing data in/out of the R workspace, including access to databases. The names of other chapters suggest the topics covered: "Dates", "Factors", "Subscripting", "Character manipulation", "Data aggregation", "Reshaping data".

This book will be helpful to any but the most absolutely new to R, and even the seasoned user will find interesting hints and examples. I cannot recommend it enough.

One minor qualm I have is the absence of references. Some topics (for instance, regular expressions) are fairly complex, and well documented elsewhere: a pointer or two would be helpful. Same with, for instance, SQL, which is mentioned and demonstrated briefly.

Another not-so-minor qualm is price. A book of this size from, for instance, Dover classics collection, with similar paper quality and covers, is about a third or fourth of the price. Although this is a new book I find the $54.95 tag (Amazon discounted price is about $44.50) fairly high. But this has nothing to do with the quality of the book, rather it has to do with the Springer pricing policies.

All in all, if you don't mind the price, this is a good buy.

All too often novices wanting to use R for an analysis never get to the analysis because they can't successfully import, clean-up and restructure their data for the analysis functions. This book prevents those problems by telling you the critical data and file manipulation materials that are usually briefly (and inadequately) covered in stat books. It is a short easy read that will give you the tools to get your data ready to go.

You can see the table of contents and read the other reviews but areas that really shine include: dealing with categorical (named or ordered) factor variables, recoding numeric data into categorical variables, and also making and working with summary tables.

When it comes to data manipulation and clean-up Spector has the best coverage of any book or web FAQ. This book is very expensive for its size but it is worth every cent.

I am neither a statistician nor a computer programmer yet I used statistics and programming on a daily basis and that is my perspective on this book. I am an ecologist with messy data with relational databases, missing data, dates etc. Most statisticians will already have more than a dollop of programming under their belts and will have a completely different experience with this book. I found the book disorganized with sections that should occur together in different sections or in the wrong order of how a user would approach a problem. For example, Chapter 3 goes over SQL coding but I wasn't sure what package we were supposed to be using or if this coding was used for all of them and, if so, what are the caveats to each. It would have been incredibly useful to discuss these three packages with strengths, weaknesses, and examples of each. It's great that there's code to use MySQL but it would have been useful to provide some information on the MySQL side so one could get going on queries without referring to yet another manual. Much of this information was redundant with other manuals that one should already have such as Introductory Statistics with R (Statistics and Computing), Modern Applied Statistics with S, and The R Book.

I was pretty disappointed with the thoroughness of this book on R. I ordered the book "R in a Nutshell" as well and Nutshell does a much better job explaining the same topics and does it in a much more through manner and you don't feel like you got ripped off spending $50 on this very thin book. I would not recommend it and instead would choose R in a Nutshell for learning how to do R data cleaning, manipulation, processing and analysis.

This is a great book for new users to R, especially those who are used to Stata and other programs. Recoding, and working with data in different formats, in particular, can be challenges for new users, this book discusses these options in depth. I had been searching for a good data manipulation book on R for a while, since I was tired of having to go to websites constantly; this book serves that purpose.

Data manipulation can often take as long as or longer than the actual analysis. The ability to manipulate data is not emphasized enough. This book gives a great introduction to this skill set. It is clearly written and provides good examples.

After using R for many statistical analyses for several years, I finally began using R for manipulating data when my SAS license expired. This book has been essential, my copy is now completely dog-eared. I found that 150 pages was a good length to provide plenty of challenging, yet masterable, material. All is clear and well organized. I recommend this book without hesitation to those who want to move beyond spreadsheets for data manipulation, or who seek an alternative to commercial products.

Whenever R is doing something I don't expect when managing data, this is the first book I turn to. Every page is filled with deep insights into how R works. It could go into some topics in more depth, but you could spend hours in R's help files and with trial and error to discover what this book provides in clear, concise form. I highly recommend it.

Data manipulation, in my opinion, is one of R's weak points, and I have found the help available directly in R to be inadequate to understanding how things work. This book fills a valuable gap, allowing R users to more easily learn how to manipulate data and work with data frames and other data structures.