Average Reviews:
(More customer reviews)With clear writing and an emphasis on showing statistical concepts using well presented examples and code this book should be ideal for the social scientist or applied researcher who wants to know how to use R to do basic to intermediate level regression analyses but does not want a formal introduction to the math. The intended audience is people who have had a "social sciences" statistics class but no background in programming. The authors support this audience by covering basic concepts like loading data from external sources like text files or SPSS (they do not deal directly with processing Excel files) and progress from simple analyses like t-tests and basic ANOVA out to more advanced topics like bootstrapped estimates, lasso/LARs, GAM and robust regression (dont worry if you dont know what these are before opening the book). The mechanics of using R, like getting additional packages for specialized analyses and setting up programming options are covered in adequate details without bogging down the flow of the book.
A nice feature is that each chapter concludes with summary boxes that capture the key statistical and code words contained in each chapter and that information is summarized in a nice glossary. Another unusual thing is that the authors mention important topics, like ridge regression and the lasso, that do not usually show up in basic analysis books.
Amazon will not let me post the link here but ... the book has a solid website and the texts include great references to other tools (including friendly R and statistics books and the Tinn-R editor).
A down side of the book is that the table of contents does not give adequate information on what is in each chapter. For example chapter 2 is listed as "The Basic Regression" and it would be nice if it said that it covers two examples where the first includes instruction on generating a random data set and the second shows how to compare two (nested ANOVA/regression) models dealing with multiple personality disorder. I mention this because the data sets are memorable so it would be good if the reader could quickly flip back to them without having to dig in the index (which does cover the data sets by topic).
Another noteworthy fault of the book is its brevity. At under 200 pages, it is not long enough to cover regression diagnostics in adequate detail. What is here is great but I hope the authors expand the book to cover likely problems in regression.
Overall, the book includes an ideal ratio of text to code and has plentiful graphics complete with the code needed to generate the pictures. This makes it easy to learn the basics of regression with R and it is the best introduction to R for the social sciences.
Click Here to see more reviews about: Modern Regression Techniques Using R: A Practical Guide
In a way that is refreshingly engaging and readable, Daniel B. Wright and Kamala London describe the most useful of these techniques and provide step-by-step instructions, using the freeware R, to analyze datasets that can be located on the books’ webpage via the SAGE homepage. Techniques covered in this book include multilevel modeling, ANOVA and ANCOVA, path analysis, mediation and moderation, logistic regression (generalized linear models), generalized additive models, and robust methods. These are all tested using a range of real research examples conducted by the authors in every chapter.
Buy cheap Modern Regression Techniques Using R: A Practical Guide now.
No comments:
Post a Comment