Authors: Andrew S. Zieffler, Jeffrey R. Harring and Jeffrey D. Long
Publisher: Wiley – 298 pages
Book Review by: Paiso Jamakar
In many newspaper reports, television broadcasts, and even in U.S. Census Bureau data found on the Internet, we come to know of comparisons among groups of different people, differentiated by ethnic origin, education, income or other factors
For example, we know generally that white households have higher incomes than African-American households, and have higher levels of education. But probably did not know that Asian Americans rank higher than whites in these two criteria of household income and educational attainment, as data compiled in the U.S. Census of 2000 shows.
This book is about the different ways of comparing groups of people. It is meant primarily for researchers and statisticians, but there is no reason why you, likely not involved in statistics, research or any allied area, cannot get practical benefit from reading it. This book helps you understand how researchers gather data about different groups of people and how they derive conclusions about those groups from the data.
George W. Cobb, who wrote the Foreword for this book, describes it as “a product of years of careful study and deep thought by its authors, who are practicing statisticians, education researchers and root-deep innovators.”
Prior to reading this valuable book, I feel it is critical for you to know these six important developments in statistics over the last 25 years that Cobb presents in his Foreword:
- Applied-statistics books should give priority to real data over “mathematically-driven abstract exposition.”
- Computer-driven methods of analyzing data have come of age.
- Computer simulations offer clearer expositions and better understanding of real-life data over abstract derivations.
- A cooperative development and broad embrace of R, the public domain, and open-source software, Its users have made its cutting-age methods freely available.
- Statistics education has come age, and people must come to know not only what is most important to learn, but what is also hardest to learn.
- What we can compute shapes what we can do; what we can do shapes what we actually do in practice; what we do in practice shapes how we think about what we do.
Because of these above developments, Cobb says that this book should be regarded as one of the most important pioneering works in the last quarter century in introducing non-statisticians to present-day thought and practice.
This book has a detailed, lengthy (eight-page) Contents section. Glancing through it, you have 12 chapters to choose from to read, whichever interests you most. Do you want to find out what “R” is? Do you want to start with how data representation and preparation is done? Or do you wan to begin with data exploration methods? Chapters on exploring data with one variable or going through data from two groups of people are available for you. This can get complex when you are sifting through multivariate data coming from many diverse groups.
As you go through the Contents section, available for you are also chapters on randomization and permutation tests, or tests done unsystematically in the first instance, and with substitution in the second instance.
Find out what bootstrap testing is, and intervals in this survey method; learn what are dependent samples; explore what are planned and unplanned contrasts in research, and other ways of data gathering and analysis.
The book has been endowed with numerous tables, charts and formulas to help you understand, absorb and retain the information presented in the book. The three authors of this book have a deep understanding of research methods and statistics and provide great value in this book for students of this subject and readers interested in it.
