Government, employment, health, housing, food, entertainment, almost all areas of modern society are dominated by vast amounts of quantitative information made possible by the technological revolution. In order to function intelligently in this society, educated persons must have some understanding of data and chance since those concepts underlie the language in which many important issues are best described. Consequently, it is not surprising that introductory statistics courses have one of the highest rates of increasing enrollment among courses in the mathematical sciences. In fact, the increase in enrollment is outstripping the increase in instructors with degrees in statistics. In addition, the practice of statistics is moving at a rapid pace into methodologies built around intensive data exploration rather than mathematical theory, necessitating an upgrading of skills even for many instructors with degrees in statistics. This new book, *Teaching Statistics*, is, then, a timely and welcome addition to the resource bank of any teacher of introductory statistics, including those teaching at the high school level.

Capturing and building upon the emphasis on data in modern statistics, the theme of the book is "Teaching Statistics: More Data, Less, Lecturing," which is also the title of the opening essay. The book is faithful to that theme throughout, as it covers sections on *Teaching with Data*, *Establishing Projects in Active Learning*, *Textbooks*, *Technology*, and *Assessment*. Articles in each of these areas are written by some of the most experienced and most successful teachers of statistics in the country; all articles contain practical advice for the teacher and each section has extensive lists of references which encourage the interested reader to delve deeper into any of these subjects. Some articles develop lessons that a reader could carry directly to the classroom, while others discuss successful teaching and assessment practices that will challenge the reader to think deeply about how to improve his or her own teaching. Still others review published resources that are built around the spirit of teaching statistics through data.

*Teaching with Data* concentrates on the use of projects and case studies in the statistics classroom, emphasizing that students must, at least in part, construct their own knowledge of the subtle concepts of statistics through hands-on experience with real data and simulations of probabilistic events. The related section on *Establishing Projects in Active Learning* discusses three recently published books that emphasize this hands-on, active approach.

The section on *Textbooks* provides advice on how to evaluate introductory textbooks and textbooks on mathematical statistics. The former suggests looking at textbooks at three levels, the reading level, the technical (statistical) level, and the mathematical level. Good authors can explain much of statistics with little reference to mathematical formalism, but many introductory books require more mathematical skills than the students are likely to possess. The latter evaluates fifteen current mathematical statistics texts on criteria such as topics covered and nature and extent of examples and exercises.

The section on *Technology* begins with an outline of a series of tasks that allows a teacher to evaluate and compare statistical software packages for use in teaching statistics. Instructors whose students do not have access to computers may be interested in the companion paper on the use of graphing calculators in teaching statistics. These articles are followed by descriptions of and excerpts from six sets of material developed around innovative use of technology (including videos) in the classroom. The concluding paper of the section provides an extensive list of WWW resources for teaching statistics.

Effective *Assessment* of student learning must go beyond the fifty-minute in-class exams that so many instructors have come to know and love. Consistent with the overall theme of the book, assessment should include projects (both individual and group), portfolios of work done throughout a course (especially with examples of written communication skills), and some deeper tools such as concept maps and rubric-based open ended questions on statistics problem solving.

There are many wonderful ideas presented throughout the book, and instructors interested in improving their teaching of statistics would do well to study the ones most applicable to their classroom situation. There is one supreme enemy of all of this — time! Teaching with data in an active environment and with authentic assessment will take more time and energy than did the old lecture style. If the goal is improved student learning, though, there is no alternative. The time spent on trying some of these ideas will be well worth the effort.

Richard L. Scheaffer (

scheaffe@stat.ufl.edu) is Professor Emeritus in the Department of Statistics at the University of Florida. He is the current (2001) president of the American Statistical Association.