This book blends the necessary theory with practical application to provide the reader with a riviting insight into the world of linear statistical models.

This is one of the few stat books that I'm keeping. I've used it to help me through Linear Regression, Experimental Design, and Multiple Regression classes. This book will be a good one to keep, even after I graduate.

It was a bit over my head, at first. After 1/2 a semester (and after reading many text books that are much worse), I soon was able to get into the text.

I was only interested in regression, when younger, as this seemed to be useful in physics labs and in scientific literature. Early programmable calculators could do regressions. But what was behind this mysterious line through the points? This book explains it every way they can think of...there are so many ways they apply the simple y=mx+b that once you can master the line in cartesian space the rest is easy. I took this book in the two semester suggested way with regression first, ANOVA second. I did very well with these course and got a stats degree. I am very fond of this book now. Yes, my data diskette was corrupted at one point and I did have to ask our professor to make me a copy. I also agree with another reviewer that the examples are good. These allowed me to learn about the application areas and remain critical of the models. However, I can see that this book could confuse some and it is very expensive. Also I agree that when solving the later problems in the book one has to flip back to the first occurance of the dataset but if your working on a computer this isn't really as much of a problem. But it is still a problem because you have to know what variable is what. A favorite school book? Yes it is.

Graduate students, and people professionally involved into dealing with data every now and then will find this book to be an indispensable comprehensive reference for most types of statistical tasks you may encounter; deep, solid, non nonsense, and all contained in one book. This is probably the only stat/math book you will need to keep.

Save Your Money. Spend your money on more usefull things like paying all those parking tickes your have received thus far. This student solutions manual has the soulutions to some and I repeat some of the many problems in the text. Having mear soulutions to problems is not enough depth to place in a manual of solutions. Most solutions manuals work out the problem to show where the answer came from, not this one.

When studying about linear models, this book is a very good resource. The very title gives one a sense of this book's target audience: _Applied_ Linear Statistical Models. Throughout its 1400 some odd pages examples and applications are sprinkled, neatly illustrating concepts relevant to the section. One problem: the end-of-chapter exercises. All of the data for use in these exercises are contained on a diskette. Fine. What happens if one were to lose this disk? Better make friends quick in order to have someone from whom to copy the data. Hypothetically, you've gone the entire semester without losing the disk. Datasets are reused throughout the book, yet explanations are only given the first time the data are used. When reusing the data one has to turn to the section where it was first used and decipher the book's difficult labelling system. The exercises do just that, and well, yet they take much more time than if their organisation were better.

I'm a senior, business administration, Yonsei Univ. in Korea. I'm studying Regression Analysis with this book. This book is easy for students specilizing in other fields to understand Regression Analysis. Very good. I could understand Regression Analysis easily because all theorems are proved in the level of introductory statistics. There are, however, no solutions of problems. In preface I read that instructors may order from the publisher, Irwin, copies of the Student Solutions Manual. Can't a student like me studying by himself receive 'Student Solutions Manual'? I'm eager to check my solutions. I don't want to stop to study Regression Analysis