Customer Reviews

Primer of Applied Regression & Analysis of Variance

5 star:		(3)
4 star:		(0)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

Like all advanced stats books, this one has mathematical rigor and plenty of examples. But unlike the others, this one is written from the point of view of a biologist. You won't just learn the math, you'll learn how to make sense of the results. The title is a bit misleading. This is not a "primer" of statistics. But once you've learned the basic principles of statistics, this is THE book to learn about various kinds of ANOVAS and regressions.

Once you've learned the basic principles of statistics, how can a biologist learn more advanced techniques? Many books focus on math rather than on understanding concepts. Other books are too narrow -- discussing only a single method. And books that focus on multiple regression and ANOVA tend to have examples from psychology and social sciences. Glantz and Slinker do a great job of explaining the principles of multiple regression, analysis of variance, and analysis of covariance. The focus is not on mathematical proofs, but rather on making sense of the results in the context of biological and medical research.

This book also has excellent chapters on linear regression, nonlinear regression (curve fitting) and logistic and proportional hazards regression (regression when the outcome is an either-or binary variable).

New to the second edition are a chapter on analysis of covariance, more extensive discussions of multiple comparisons methods, and a discussion of Cox proportional hazards regression for analyses of survival data.

The title is a bit misleading. This is not a "primer" of statistics. But once you've learned the basic principles of statistics, this is THE book for biologists to learn about various kinds of ANOVAS and regressions.

I looked at several options for a regression textbook that would be both understandable and relatively complete for my introduction to the topic. This book won hands down. The authors keep it simple and use a wide variety of examples to get the point across. I felt that the sections on logistic and Cox regression could have been a bit better, but these subjects are best learned by dedicated textbooks such as Hosmer and Lemeshow and Collett.

I think that this may be the best introductory regression book out there.