A PhD Student's review

The review about this product

SPSS 10.0 Regression Models (Textbook Binding) This book is a must have for the burgeoning doctoral candidate or graduate student. I am a student working on my Ph.D. in ORSA and find SPSS to be a wonderful and powerful statistical tool; however, maneuvering around on the SPSS tool bar and interpreting output can be extremely frustrating. This book really fills in the gaps on data manipulation and analyzing output that most graduate and post graduate class texts skip over.
When I was working on my master's degree in 1996, I attended a 2-day course from the SPSS corporation on regression models. I was given a three ring binder as part of my course materials. Back in 1996, there were not a lot of books that taught you to manipulate data in SPSS. As a result, the three ring binder I got as part of my SPSS course really saved the day and helped me complete my graduate thesis. I even loaned out my binder on numerous occasions to my fellow students (and even professors).
Anyway, I was pleased to see that this Regression Models 10.0 book is the advanced, professional version of what used to be handed out in costly, pay SPSS courses. A lot of the material in this book is similar to my 1996 three-ring binder. But it is organized better and contains much more useful information (graphs, charts and explanations). You can tell the SPSS corporation has put a few years into this book and obviously updated it based on feedback from students over the many years of classroom instruction. The only criticisms I have (which prevented me from giving it all 5 stars) include:
1. This book does not come with a reference disk: The binder I got in 1996 from the SPSS corporation (which is very similar to this book in many ways), had a 3.5" disk with the same data on it that was used in the binder to get the various output models. This disk was (is) handy because when I am playing with new data sets, I sometimes go back to my old SPSS binder and plug in some of the data from that binder to make sure I am doing the procedures right on my own data set.
2. There is no glossary: If you are using SPSS; and especially running logistic regression models, you should probably already know what a -2LL and the difference between the Cox & Snell R2 and Nagelkerke R2. However, for the fatigued graduate student or Ph.D. candidate, it would sure be nice to be able to have a glossary to validate the meanings of some terms at 2AM after a long 16-hour study day. Yes, there is a very comprehensive glossary already contained on the SPSS CD-ROM that can be accessed from SPSS while on the computer - but I'd rather have some significant and selected terms in the back of my book.
3. This book doesn't give you guidelines for interpreting statistics. For instance, anyone who has ever taken one semester of statistics should know that a p-value less than .05 is significant. However, even doctoral candidates sometimes can't remember if a significant Wald statistic is suppose to greater than 1.0 or 2.0? This book has a lot of good discussion, but I'd prefer a little summary table at the back of the book with constraints for the various statistics.
All in all, a great book and a must have for any graduate or doctoral student! I am glad I can finally retire my 5-year-old binder and begin using the "evolved version" of this book to help me complete my disertation!