The present text introduces the student to the basic ideas of estimation and hypothesis testing early in the course after a rather brief introduction to data organization and some simple ideas about probability. Estimation and hypothesis testing are discussed in terms of the two-sample problem. The book exploits nonparametric ideas that rely on nothing more complicated than sample differences Y-X, referred to as elementary estimates, to define the Wilcoxon-Mann-Whitney test statistics and the related point and interval estimates. The ideas behind elementary estimates are then applied to the one-sample problem and to linear regression and rank correlation. Discussion of the Kruskal-Wallis and Friedman procedures for the k-sample problem rounds out the nonparametric coverage. The concluding chapters provide a discussion of Chi-square tests for the analysis of categorical data and introduce the student to the analysis of binomial data including the computation of power and sample size. Most chapters in the book have an appendix discussing relevant Minitab commands.

The introductory statistics course presents serious pedagogical problems to the instructor. For the great majority of students, the course represents the only formal contact with statistical thinking that he or she will have in college. Students come from many different fields of study, and a large number suffer from math anxiety. Thus, an instructor who is willing to settle for some limited objectives will have a much better chance of success than an instructor who aims for a broad exposure to statistics. Many statisticians agree that the primary objective of the introductory statistics course is to introduce students to variability and uncertainty and how to cope with them when drawing inferences from observed data. Addi tionally, the introductory COurse should enable students to handle a limited number of useful statistical techniques. The present text, which is the successor to the author's Introduction to Statistics: A Nonparametric Approach (Houghton Mifflin Company, Boston, 1976), tries to meet these objectives by introducing the student to the ba sic ideas of estimation and hypothesis testing early in the course after a rather brief introduction to data organization and some simple ideas about probability. Estimation and hypothesis testing are discussed in terms of the two-sample problem, which is both conceptually simpler and more realistic than the one-sample problem that customarily serves as the basis for the discussion of statistical inference."