Statistics: The Art and Science of Learning from Data, Global Edition

Book Preface

We have each taught introductory statistics for many years, and we have witnessed the welcome evolution from the traditional formula-driven mathematical statistics course to a concept-driven approach. This concept-driven approach places more emphasis on why statistics is important in the real world and places less emphasis on mathematical probability. One of our goals in writing this book was to help make the conceptual approach more interesting and more readily accessible to students. At the end of the course, we want students to look back at their statistics course and realize that they learned practical concepts that will serve them well for the rest of their lives.

We also want students to come to appreciate that in practice, assumptions are not perfectly satisfied, models are not exactly correct, distributions are not exactly normally distributed, and different factors should be considered in conducting a statistical analysis. The title of our book reflects the experience of data analysts, who soon realize that statistics is an art as well as a science.

What’s New in This Edition

Our goal in writing the fourth edition of our textbook was to improve the student and instructor user experience. We have:

• Clarified terminology and streamlined writing throughout the text to improve ease of reading and facilitate comprehension.
• Used real data and real examples to illustrate almost all concepts discussed. Throughout the book, within three to five consecutive pages, an example is presented that depicts a real-world scenario to illustrate the statistical concept discussed.
• Introduced new web-based applets (referred to as web apps or apps) illustrating and helping students interact with key statistical concepts and techniques. These apps invite students to explore consequences of changing parameters and to carry out statistical inference. Among other relevant concepts and techniques, students are introduced to:
• Sampling distributions
• Central limit theorem
• Bootstrapping for interval estimation (Chapter 8)
• Randomization or permutation tests for significance testing (Chapter 10 for difference in two means and Chapter 11 for two categorical variables).
• Inserted brief overviews to set the stage for each chapter, introducing students to chapter concepts and helping them see how previous chapters’ concepts, tools, and techniques are related.
• Included computer output from the most recent versions of MINITAB and the TI calculator.
• Expanded Chapter 1, providing key terminology to establish a foundation to understand the big picture of the statistical investigative process—the importance of asking good statistical questions, designing an appropriate study, performing descriptive and inferential analysis, and making a conclusion.

• Reflected the latest trends in statistical education, including:
• Measures of association for categorical variables in Chapter 3
• Permutation testing in Chapters 10 and 11
• Updated coverage of McNemar’s test in Chapter 10 (previously Chapter 11)
• Moved important coverage of risk difference and relative risk to Chapter 3 (instead of first introducing these measures in Chapter 11). We believe that understanding these two statistics is a necessary part of statistical literacy for the everyday citizen as they are pervasive in mass media and the medical literature.
• Updated or replaced over 25 percent of the exercises and examples. In addition, we have updated all General Social Services (GSS) data with the most current data available.