Preface

This text evolved from a set of class notes for an MBA course and from teaching materials developed as part of in-house training on quantitative analysis tools in several industrial and financial corporations, some as part of Six Sigma training. Our MBA students, who often have limited inclination for formula-based statistical instruction, need statistical tools to perform course work in finance, economics, operations, and quality management. Students in industry often have job-related projects with strong statistical content. We have found that both groups are quite willing to delve into fairly sophisticated statistical methods if they are given a practical and problem-oriented approach that involves the assistance of modern statistical software.

Accordingly, we aim this text at business and management students, both undergraduate and graduate, and at professionals in industry with a need to use and interpret statistics. The majority of the examples and exercises have been gathered from various sources in manufacturing and service industries. Of course, we have modified many problems to suit the introductory level of the text and to make pedagogical points, but have retained the flavor of the originals.

We offer a standard selection of introductory topics in classical statistics. We have chosen them on the basis of their usefulness to business and management and their necessity in other management courses. We have also included questionnaire design because of its importance in surveys, and exponential smoothing because of its importance in operations. We also cover the statistical steps for dealing with the capital asset pricing model and price elasticity of demand. Page limitations required that we omit topics such as quality methods, design of experiments, simulation, or partitioning.

Most of our students had a recent exposure to probability for about half a semester. We offer two probability chapters to review probability and to demonstrate the capabilities of JMP. These two chapters should enable readers to understand the remainder of the text.

We have attempted to avoid statistical formulas as much as possible and to define important concepts in plain English and graphically. Most chapters demonstrate a statistical tool by working through a detailed example according to the following template:

• problem statement

• data requirements

• implementation in JMP

• discussion of the JMP results

• interpretation to address the problem statement

We used JMP 8 to develop the examples. We have found that the JMP platforms used in this text did not undergo major changes from version to version. However, some option menus have been extended. In some instances, such as the case of sample sizes for proportions, a newer method has replaced an older one.

Each chapter includes a problem set and all but one ends with at least one case study. The problems range from simple to intermediate in their level of difficulty, with specific questions intended to reinforce key concepts. The case studies offer more unstructured tasks and are more open-ended than the problems.

The following supportive materials are available from http://support.sas.com/authors:

• data sets for the examples in the text

• data sets for the problems at the end of each chapter

• solutions to most problems

• Microsoft PowerPoint presentations of the main points in each chapter

• supplementary chapters and examples

Our approach to teaching a statistics course for management has been analogous to owning a car. Whoever owned a car around 1900 needed to be a mechanic or had to hire a driver with good mechanical skills. Nowadays, car technology has advanced to a state where a car owner needs to know only how to drive a car and what the rules of the road are. Still, highly trained automotive engineers continue to be engaged in research to improve cars.

Similarly, before the widespread availability of statistical software, statistical analysis required advanced training to perform long-winded calculations. Statistical software has changed that. Complicated calculations can be performed on a laptop in a split second. Complex graphs can be drawn and redrawn with a mouse click. But the laptop owner needs to know the rules of the statistical road. Meanwhile, highly trained statisticians are engaged in researching new tools and in improving existing ones.