Gudmund R. Iversen

discussed in this book. It is clear that with an understanding of the main ideas of statistics, engaged citizens can grasp what the professional number crunchers have produced and evaluate the results. This book grew out of a course designed by Gudmund R. Iversen to meet the challenges created by this greater reliance on statistical It was one of a series of courses designed at Swarthmore information. College to fulfill the mission of a liberal arts college to educate its students for the challenges of the twenty-first century. The idea was that students should not become so involved with the intricacies of a single discipline that they lose sight of the big picture. These courses were intended to educate students to understand how the major ideas of a field relate to the world. In many respects statistics seemed an ideal subject for one such course. While statistics could be a mystifying, self­ aggrandized, and esoteric discipline, it could also be a key to under­ standing many other disciplines. The course, Stat 1: Statistical Think­ ing, was created to produce this understanding. The course proved to be very popular, and each year it grew in size. Over time the lecture notes for the course became more refined and extensive, and eventu­ ally the course material served as the basis for this book. Fonnulas As most statistics instructors are keenly aware, the teaching of statistics has changed dramatically.

Aimed at readers who may be more familiar with statistics than calculus and mathematics, this carefully written volume gives an overview of the central ideas in calculus. Author Gudmund R. Iversen shows examples of how calculus is used to translate many real-world phenomena into mathematical functions. Beginning with an explanation of the two major parts of calculus, differentiation and integration, Iversen illustrates how calculus is used in statistics to distinguish between the mean and the median, to derive the least squares formulas for regression coefficients, to find values of parameters from theoretical distributions, and to find a statistical p value when we using one of the continuous test variables like the t variable. Social scientists who either never took a calculus course or who want to "brush up" on their understanding of calculus will find this book a necessity.

Contextual analysis, the study of the role of the group context on actions and attitudes of individuals, is a useful technique in the study of education, neighborhoods, census tracts, election districts, and the family. However, the effective use of contextual analysis has involved overcoming a number of issues, such as group boundaries, the mobility of the individuals within a group, overlapping groups, missing individual data, and the choice of statistical models. Contextual Analysis offers researchers a guide for selecting the best model to use. Written in a straightforward style, the book explores such topics as contextual analysis with absolute effects, with relative effects, and the choice between regression coefficients as fixed parameters or as random variables.

The authors have improved on their widely used first edition by providing updated examples, adding material on how to do ANOVA using statistical packages for microcomputers, linking the use of ANOVA to regression analysis, and enchancing their discussion on using ANOVA for experimentally gathered data.

Statisticians now generally acknowledge the theorectical importance of Bayesian inference, if not its practical validity. According to Gudmund R. Iversen, one reason for the lag in applications is that empirical researchers have lacked a grounding in the methodology. His volume provides this introduction and serves as a companion to #4, Tests of Significance.