Customer Reviews

Multiple regression in behavioral research

5 star:		(6)
4 star:		(2)
3 star:		(1)
2 star:		(2)
1 star:		(4)

 

 

Pedhazur's third edition of this text that he began with Kerlinger is an excellent introduction to linear regression. I use this text in a statistics course that I teach to first-year graduate students and they have responded favourably to me about the book. Pedhazur is a stickler for 'getting it right,' which is critically important for the serious student to learn at this stage. The book is replete with examples of both doing it right and doing it wrong (the latter culled from the literature). If you want to learn how to do linear regression correctly, this text is excellent within its scope as a primer. The introductory chapters on multilevel modelling and logistic regression are nice touches. All in all a very useful text.

This is a wonderful text for a class primarily dedicated to multiple regression. I own and have read many regression textbooks, and this continues to be one of the very best! The reader reviews below appear to mostly be expressing frustration with instructors, and not this text. If you are taking an introductory course in which regression is one of many topics, this book may be too advanced for your needs (and may be a poor choice by your instructor!). However, if you are taking a course dedicated to regression (linear and logistic; or general linear models in general), this is one of the very best text available!

My graduate class had this text last semester for psychology statistics. No one in the class liked the text very much, and some refrained from reading it for exams due to the difficulty in extracting out useful concepts and materials. Much of the text is devoted to various statistical software packages, but this may be of limited value if you don't have the exact software version listed in the text. I had SPSS 9.0 which has different syntax from the earlier SPSS version in the text. In addition, the author appears to editorialize too much, and with too much animosity, further hindering the learning process. I am interested in (re)learning some more of the details of the various regression techniques, but do not want to wade through all the irrelevant material to do so. I'm actually looking for a better text on the subject on Amazon now. This semester, we are focusing on ANOVAs and are using "Design and Analysis" (3rd Ed) by Keppel. It's an excellent text, very clear, and (the perhaps "easier" material notwithstanding) a direct contrast to Pedhazur. However, I'm keeping Pedhazur as a last-ditch reference, as it does appear to have broad coverage of regression. This is the main reason I gave it two stars instead of one.

If you're serious about learning quantitative research, well, this man takes his work very, very seriously, much to our benefit. He has an extraordinary intellect. I would call the book "magisterial" if his self-importance weren't taken to such hilarious heights. But that comic relief is welcome, too.

The book plumbs the depths of the general linear model, providing extensive, rigorous, extremely well-considered treatment of statistical control, linear regression, analysis of covariance, logistic regression, and structural equation modeling. Includes important sections on topics such as nonlinear relationships and interactions. Has sections that I reread every few months to a year because the ideas are so important to my understanding of analytic issues and so hard to come by anywhere else.

It's not for beginners, but don't wait too long! You'll kick yourself for all the insights you could have developed, and all the conceptual and technical mistakes you could have avoided.

Two drawbacks. There is too much attention to SPSS, SAS, and other code for many people's taste. Also, it is weighted down by unnecessarily precise calculation, overemphasizing hypothesis testing and statistical significance at the expense of the size of effects. One of the more insidious statistical mistakes (see [...]).

The author explains multiple regression in a way that is easy to follow. I am actually a beginner in statistics and had experiences in reading some other books which looked to me as if they were written by aliens.

I would recommend this book for courses offered to graduate students.

Pedhazur is fat. The book is as thick as someone seeking the help of weight watchers. The inclusion of material from multiple statistical packages is a waste of space. That information should be left to the instruction manuals from the companies that produce the statistical packages. The packages are constantly being updated and the information is quickly outdated. In all honesty (and with no disrespect intended to a great man, Pedhazur) is seems like a not-so-suttle attempt by an oldtimer to seem hip. If code really needs to be added, then pseudo-code (generic) would be enough or pick only one package. Pedhazur's notation is problematic for statistical youngsters. Personally I don't like it. His shorthand, such as Sum of xy meaning Sum (x-xmean)(y-ymean) is not helpful for beginners. I suggest 'youngsters' stay away until you've matured a bit. More mature readers like the book because it is full of juicy bits of information, however, if you're an instructor beware: your kids are not you! The book deserves space on the shelf as a reference (it looks impressive). For early learning, I don't recommend it.

This textbook is VERY comprehensive and detailed. The author provides several methods to arrive at the same conclusion, and is very clear in delievering those methods. For the beginning statistics student, this text is overwhelming at first, and will need several reads before the content becomes clear. In any case, the author errs on the side of being super comprehensive and readers will have to realize that he presents several concepts multiple times using multiple methods. The content is a bit dense; it is something to get used to. Also, the author tends to point out errors that researchers have made and uses them as examples in his explanations of what NOT to do, which comes off as a little narcissistic. Personal woes aside, this is a good buy.

Looking at the other reviews of this book you will notice that one is favorable, one is against, and one is neutral. The book is fat (most comprehensive stats books are) - at least you can use it as a paper wieght. The demonstration of multiple computer packages may be a waste (especially since the packages keep adding new functions, etc.), however, pseudo code would not be helpful to newbies, who probably wouldn't understand it. Yes, Pedhazur's notation is lame, but the book is fairly comprehensive about regression (it would be nice if repeated measures type studies were mentioned though). ...

This is an easy to read text on multiple regression analysis. It has numerous examples to illustrate the various methods available to analyze data, to spot correlations and understand the relative importance of independent variables in explaining the variations in a dependent variable. I would recommend this without hesitation.

very pleased with the prompt delivery and ease of placing my order - will certainly order my books this way in the future!