You are here

Teaching Statistics: A Bag of Tricks

Andrew Gelman and Deborah Nolan
Oxford University Press
Publication Date: 
Number of Pages: 
[Reviewed by
Andreas Karlsson
, on

This book is written by statistics teachers for statistics teachers, with the aim of being “a convenient resource for instructors of introductory probability and statistics” at high schools and colleges. The authors draw on their own experience of teaching introductory statistics to college students, but they stress that the book is as much collected as it is authored. One of the authors' goals is to collect in one place all the good examples and demonstrations in statistics teaching that they have heard of or developed in their own courses.

Definitions are in order here: By an “example,” the authors mean an activity that is conducted by the instructor, while a “demonstration” means an activity that involves active student participation.

The book is divided into three parts: Part I, Introductory probability and statistics , makes up the main part of the book and has a large selection of examples, demonstrations and projects useful for introductory statistics courses. In nine chapters it covers such topics as the first week of class, descriptive statistics, linear regression and correlation, data collection, statistical literacy and the news media, probability, statistical inference, multiple regression and nonlinear models, and lying with statistics.

Part II, Putting it all together , focuses on the practical aspects of implementing activities in class. The authors give tips on what works and what does not work, how effective demonstrations and examples are set up, and how to manage group work in class and projects to make the students work effectively. The authors also give examples of mistakes they have made themselves in carrying out activities. This part of the book also has a chapter on the structuring of an introductory statistics course, including a detailed syllabus for a non-calculus-based semester-long introductory course that demonstrates how the activities discussed in the book can be integrated in a course.

Finally, Part III, More advanced courses , is devoted to examples, demonstrations and projects useful for more advanced statistics courses, and includes chapters on such topics as decision theory and Bayesian statistics, student activities in survey sampling, problems and projects in probability, and directed projects in a mathematical statistics course.

The book is well written and richly illustrated with figures, diagrams, tables and pictures. Each demonstration, example and project is presented in a self-contained format, with the goal of making it easy for a teacher to pick out the activities he or she likes the most without having to read the whole book. This structure is mostly successful.

There was one issue that puzzled me: The authors repeatedly talk about rolling a dice to get random numbers, especially when discussing sampling in Chapter 5. But they never explain how random number including zeros or the numbers 7 to 9 could be obtained with a dice as described in many of the activities. Not until Chapter 7 did they tell their readers that they do not use an ordinary dice but rather a 20-sided die with each of the digits 0 to 9 written twice.

This notwithstanding, this book is very pedagogical, is enjoyable to read, gives many interesting ideas and examples. It should prove to be very useful and stimulating to all statistics teachers. I highly recommend it.

Andreas Karlsson ( is a Ph. D. in statistics graduated from the Division of Statistics, Department of Information Science, Uppsala University ( He is currently working as a biostatistician at the Västerås Center for Clinical Research (part of Uppsala University) at Central Hospital in Västerås, Sweden. His primary research interest is in bootstrap methods and quantile regression with applications in biostatistics.

1. Introduction

Part I: Introductory probability and statistics
2. First week of class
3. Descriptive statistics
4. Linear regression and correlation
5. Data collection
6. Statistical literacy and the news media
7. Probability
8. Statistical inference
9. Multiple regression and nonlinear models
10. Lying with statistics
Part II: Putting it all together
11. How to do it
12. Structuring an introductory statistics course
Part III: More advanced courses
13. Decision theory and Bayesian statistics
14. Student activities in survey sampling
15. Problems and projects in probability
16. Directed projects in a mathematical statistics course