A Second Course in Statistics: Regression Analysis

by William Mendenhall, Terry L. Sincich

This reader-friendly book focuses on building linear statistical models and developing skills for implementing regression analysis in real-life situations. It includes applications for a range of fields including engineering, sociology, and psychology, as well as traditional business applications. The authors use the latest material available from news articles, magazines, professional journals, the Internet, and actual consulting problems to illustrate real business situations and how to solve them using the tools of regression analysis. In addition, this book emphasizes model building and multiple regression models and pays special attention to model validation and spline regression. For professionals in any number of fields, including engineering, sociology, and psychology, who would benefit from learning how to use regression analysis to solve problems.

1028 pages, Hardcover

First published January 1, 1981

Reviews

[alexis peat]

do not recommend getting a masters degree in statistics

[Matt · Rating 5 out of 5]

Ohh yeah, regress my ass

[fedya]

This book is not really your general “math” textbook. The authors clearly aim to make the material accessible even to readers without a strong mathematical background, which mathematicians or statisticians may find insufficiently rigorous. While the exposition and overall structure are not my favorite, the book does cover a wide and comprehensive range of topics. Importantly, it goes beyond model construction and assumptions to emphasize what regression models actually mean, both conceptually and in practice, and how to exercise sound judgment not only about the model itself but also about the context in which it is applied, the consequences of its underlying assumptions, and the caveats and potential pitfalls that may arise. This interpretive and context-aware emphasis is sometimes forgotten in practice, despite being crucial to the discipline and one of the features that distinguishes statistics from areas where interpretability is often secondary, such as our neighbouring discipline of machine learning.

[Dr. Jake Brock · Rating 1 out of 5]

Disorganized, disjointed and unclear on explanations. Have seen far better statistics books.

[Rusty · Rating 4 out of 5]

This is a textbook used in an undergraduate statistics class. Good book.

[Khoa · Rating 3 out of 5]

I had to read this my class, so it's okay. Could've had more explanations.