The primary questions that must be answered when a new statistics text for engineers and scientists is written
relate to the issue of the contribution of the textbook to the pedagogy of teaching statistics to this audience
of students and to how the text will differ from the many texts that are already available. These questions can
be answered for the proposed text only in the context of recommendations that have been made as the result of a
1984 conference on the statistical education of engineers Hogg(1985) and a 1993 Quality Engineering Workshop Hogg(1994).
Among the recommendations made was that engineers need to appreciate the following statistical concepts:
omnipresence of variability;
the use of simple graphical tools such as run charts, histograms, scatter plots, probability plots, residual
plots, and control charts;
basic concepts of statistical inference;
the importance and essentials of carefully planned design of experiments;
the philosophies of Shewhart, Deming, and other practitioners concerning the quality of products and services.
The Hogg(1994) article proposed a core course of topics for engineering students. This proposed text is based
on the curriculum model presented in that article.
Educational Philosophy
In our many years of teaching introductory statistics courses to students majoring in a wide variety of disciplines,
we have continually searched for ways to improve the teaching of these courses. Over the years, our vision has
come to include the following:
Students need a frame of reference when learning a subject, especially one that is not their major. This frame
of reference for engineering and science students should be applications to the various areas of engineering and
the sciences. The discussion of each statistical topic should include references to at least one of these areas
of application.
Virtually all the students taking introductory statistics courses for engineers and scientists are majoring
in areas other than statistics. Introductory courses should, therefore, focus on the underlying principles that
are important for nonstatistics majors.
The use of spreadsheet and/or statistical software should be integrated into all aspects of the introductory
statistics course. The reality that exists in the workplace is that spreadsheet software (and sometimes statistical
software) is typically available on one's computer desktop. Our teaching approach needs to recognize this reality
and make our courses more consistent with the workplace environment.
Textbooks that use software must provide detailed instructions that maximizes the student's ability to use
the software with a minimum risk of failure.
Instruction for each topic should focus on (1) the application of the topic to an area of engineering or the
sciences, (2) the interpretation of results, (3) the presentation of assumptions, (4) the evaluation of the assumptions,
and (5) the discussion of what should be done if the assumptions are violated. These 'points are particularly important
in regression, forecasting and in hypothesis testing. Although the illustration of some computations is inevitable,
the focus on computations should be minimized.
Both classroom examples and homework exercises should relate to actual or realistic data as much as possible.
Introductory courses should avoid an over-concentration on one topic area and instead provide breadth of coverage
of a variety of statistical topics. This will help students avoid the "I can't see the forest for the trees"
syndrome.
The main features of this proposed text are summarized in the following sections.
Main Feature: Emphasis on Data Analysis and Interpretation of Computer Output
The personal computer revolution has dramatically changed how information is analyzed in the workplace and how
statistics should be taught in the classroom. In this text, we take the position that the use of computer software
in the form of a spreadsheet application such as Microsoft Excel or a statistical package such as MINITAB is an
integral part of learning statistics. We emphasize analyzing data, interpreting the output from Microsoft Excel
and MINITAB, and explaining how to use this software while reducing the emphasis on computation. In order to carry
out our approach, we have integrated this output into the fabric of the text. For example, our coverage of tables
and charts in Chapter 2 focuses on the interpretation of various charts, not on their construction by hand. In
Chapter 9 on hypothesis testing, we have made sure to include extensive computer output so that the p-value
approach can be used. The presentation of simple linear regression in Chapter 12, assumes that software such as
Microsoft Excel or MINITAB will be used, and thus our focus is on the interpretation of the output, not on hand
calculations (which have been placed in a separate section of the chapter).
Main Feature: Problems, Case Studies, and Team Projects
"Learning" results from "doing." This text provides the student with the opportunity to
select from many problems (most with multiple parts) presented at the ends of sections as well as at the ends of
chapters. Most of these problems use real data and apply to realistic situations in various fields of engineering
and the sciences. Students can aid their comprehension by engaging in multiple hands-on exercises as detailed below.
The end-of-section problems give the students the opportunity to reinforce what they have just learned.
The chapter review problems included at the end of each chapter are based on the concepts and methods learned
throughout the chapter.
Answers to Selected Odd-Numbered Problems appear at the end of the text.
Detailed Case Studies are included at the end of several chapters.
The Whitney Gourmet Cat Food Company case study is included at the end of most chapters, as an integrating
theme.
Main Feature: Appendices on Using Microsoft Excel and MINITAB
Rather than rely on the supplementary manuals, that accompany statistical software packages, it is a much better
pedagogical approach to provide an explanation of how the software is used in the text while employing the in-chapter
examples. Detailed appendices are included at the end of all chapters that explain how to use MINITAB, the most
popular statistical software for introductory business statistics, and Microsoft Excel, the dominant spreadsheet
package. In addition, an appendix is provided after Chapter 1 that explains the basics of the Windows operating
environment.
Main Feature: Statistics Add-In for Microsoft Excel--PHStat
The CD-ROM that accompanies the text includes the PHStat Statistics add-in for Microsoft Excel that facilitates
its use in introductory statistics courses. Although Microsoft Excel is a spreadsheet package, it contains features
that enable it to perform statistical analysis for many of the topics in this text. In some cases, however, such
analyses are cumbersome in the off-the-shelf version of Microsoft Excel. The PHStat statistics add-in provides
a custom menu of choices that leads to dialog boxes which enable users to make entries and selections to perform
specific analyses. PHStat minimizes the work associated with setting up statistical solutions in Microsoft Excel
by automating the creation of spreadsheets and charts. PHStat, along with Microsoft Excel's Data Analysis tool,
now allows users to perform statistical analyses on virtually all topics covered in this text.
Main Feature: Pedagogical Aids
Numerous features designed to create a more stimulating learning environment throughout the text include:
conversational writing style;
a Using Statistics example that illustrates the application of at least one of the statistical methods
covered in each chapter in engineering and the sciences;
real data for many of the examples and problems;
exhibit boxes that highlight important concepts;
comment boxes that focus on assumptions of statistical methods;
problem sets with varied levels of difficulty and complexity;
key terms;
chapter opening quotes from a philosopher, historical figure, well-known statistician or from literature;
chapter ending problems that begin with Checking Your Understanding problems that require students to
demonstrate their understanding of concepts;
explanation and illustration of statistical tables;
side notes in which additional material appears adjacent to where it is referenced.
Main Feature: Statistical Topics Covered
The text focuses on such topics as tables and charts (Chapter 2), descriptive statistics (Chapter 3), control
charts (Chapters 6 and 7), experimental design (Chapters 10 and 11), regression (Chapters 12 and 13), and statistical
inference (Chapters 8 and 9). This emphasis is consistent with the recommendations presented by Hogg(1994).
Perhaps the important statistical method used by engineers in industry is experimental design. Simply stated,
engineers need to know how to conduct experiments where multiple factors are varied. Thus, in addition to coverage
of one- and two-factor designs, this text discusses the concept of interaction in depth. Further, it provides coverage
of factorial and fractional-factorial designs, using both a graphical approach and a confirmatory hypothesis-testing
approach. In addition, the contributions of the Japanese engineer Genichi Taguchi are introduced.
By providing this comprehensive coverage of quality and experimental design, the text provides an orientation
that allows the presentation of statistical tools in an organizational context, instead of in isolation. The goal
is for students to learn not just how to use the tools but why and how statistical methods are useful in a wide
variety of industrial settings. A portion of Chapter 1 is devoted to quality management, including both key themes
and the contribution of individuals such as W Edwards Deming, Joseph Juran, and Walter Shewhart.
Main Feature: Full Supplement Package
The supplement package that accompanies this text includes:
Instructor's Solution Manual--written by M&N Toscano. This solutions manual is enriched with extra
detail in the problem solutions and many Excel solutions. ISBN 0-13027423-2.
Student Solutions Manual--written by M&N Toscano. To order this supplement for your students, use
ISBN 0-13-028681-8.
About the World Wide Web
The text has a home page on the World Wide Web with an address of http://www.prenhall.com/levine. There
is a separate home page on the World Wide Web for the PHStat add-in, http://www.prenhall.com/phstat that
provides user assistance and periodic updates.
Acknowledgements
We are extremely grateful to the many organizations and companies that allowed us to use their data in developing
problems and examples throughout the text. We would like to thank American Cyanamid Company, American Society for
Testing and Materials, Biometrika, Environmental Progress, Graphics Press, Journal of Energy Resources Technology,
Journal of Engineering for Industry, Journal of Structural Engineering, Journal of the Minerals, Metals and Materials
Society, Journal of Water Resources Planning and Management, New England Journal of Medicine, Newsday, Noise Control
Engineering Journal, Philosophical Transactions of the Royal Society, Quality and Reliability Engineering International,
Quality Engineering, Quality Progress, Technometrics, The American Statistician, and The Free Press.
We would also like to express our gratitude to David Cresap, University of Portland, Dr. C. H. Aikens, The University
of Tennesee-Knoxville, and Dr. Robert L. Armacost, University of Central Florida for their constructive comments
during the writing of this text.
We offer special thanks to Kathy Boothby Sestak, Joanne Wendelken, Ann Heath, and Gina Huck of the editing team
at Prentice Hall, and to Bob Walters our Production Editor. Thanks also to Brian Baker for his copyediting and
M&N Toscano for their accuracy checking.
David M. Levine
Patricia P. Ramsey
Robert K. Smidt
Summary
This applied book for engineers and scientists, written in a non-theoretical manner, focuses on underlying principles
that are important in a wide range of disciplines. It emphasizes the interpretation of results, the presentation
and evaluation of assumptions, and the discussion of what should be done if the assumptions are violated. Integration
of spreadsheet and statistical software complete this treatment of statistics. Chapter topics include describing
and summarizing data; probability and discrete probability distributions; continuous probability distributions
and sampling distributions; process control charts; estimation procedures; hypothesis testing; the design of experiments;
and simple linear and multiple regression models. For individuals interested in learning statistics--without a high
level of mathematical sophistication.
Table of Contents
1. Introduction to Statistics and Quality Improvement.
What Is Statistics? Why Study Statistics? Statistical Thinking: Understanding and Managing Variability. Variables,
Types of Data, and Levels of Measurement. Operational Definitions. Sampling. Statistical and Spreadsheet Software.
Introduction to Quality. A History of Quality and Productivity. Themes of Quality Management. The Connection between
Quality and Statistics. Appendix 1.1: Basics of the Windows User Interface. Appendix 1.2: Introduction to Microsoft
Excel. Appendix 1.3: Introduction to MINITAB.
2. Tables and Charts.
Introduction and the History of Graphics. Some Tools for Studying a Process: Process Flow Diagrams and Cause-and-Effect
Diagrams. The Importance of the Time-Order Plot. Tables and Charts for Numerical Data. Checksheets and Summary
Tables. Concentration Diagrams. Graphing Categorical Data. Tables and Charts for Bivariate Categorical Data. Graphical
Excellence. Appendix 2.1: Using Microsoft Excel for Tables and Charts. Appendix 2.2: Using MINITAB for Tables and
Charts.
3. Describing and Summarizing Data.
Introduction: What's Ahead. Measures of Central Tendency, Variation, and Shape. The Box-and-Whisker Plot. Appendix
3.1: Using Microsoft Excel for Descriptive Statistics. Appendix 3.2: Using MINITAB for Descriptive Statistics.
4. Probability and Discrete Probability Distributions.
Introduction. Some Rules of Probability. The Probability Distribution. The Binomial Distribution. The Hypergeometric
Distribution. The Negative Binomial and Geometric Distributions. The Poisson Distribution. Summary and Overview.
Appendix 4.1: Using Microsoft Excel for Probability and Probability Distributions. Appendix 4.2: Using MINITAB
for Probability and Probability Distributions.
5. Continuous Probability Distributions and Sampling Distributions.
Introduction to Continuous Probability Distributions. The Uniform Distribution. The Normal Distribution. The
Standard Normal Distribution as an Approximation to the Binomial and Poisson Distributions. The Normal Probability
Plot. The Lognormal Distribution. The Exponential Distribution. The Weibull Distribution. Sampling Distribution
of the Mean. Sampling Distribution of the Proportion. Summary. Appendix 5.1: Using Microsoft Excel for Continuous
Probability Distributions and Sampling Distributions. Appendix 5.2: Using MINTAB for Continuous Probability Distributions
and Sampling Distributions.
6. Process Control Charts I: Basic Concepts and Attribute Charts.
Introduction to Control Charts and Their Applications. Introduction to the Theory of Control Charts. Introduction
to Attributes Control Charts. np and p Charts. Area of Opportunity Charts (c Charts and u
Charts). Summary. Appendix 6.1: Using Microsoft Excel for Attribute Control Charts. Appendix 6.1: Using Microsoft
Excel for Attribute Control Charts. Appendix 6.2: Using MINITAB for Attribute Control Charts.
7. Statistical Process Control Charts II: Variables Control Charts.
Introduction to Variables Control Charts. Rational Subgroups and Sampling Decisions. Control Charts for Central
Tendency (X Charts) and Variation (R and s Charts). Control Charts for Individual Values (X
Charts). Special Considerations with Variable Charts. The Cumulative Sum (CUSUM) and Exponentially Weighted Moving
Average (EWMA) Charts. Process Capability. Summary. Appendix 7.1: Using Microsoft Excel for Variables Control Charts.
Appendix 7.2: Using MINITAB for Variables Control Charts.
8. Estimation Procedures.
Introduction. Properties of Estimators. Confidence Interval Estimation of the Mean. Confidence Interval Estimation
for the Variance. Prediction Interval Estimate for a Future Individual Value. Tolerance Intervals. Confidence Interval
Estimation for the Proportion. Summary. Appendix 8.1: Using Microsoft Excel for Confidence Interval Estimation.
Appendix 8.2: Using MINITAB for Confidence Interval Estimation.
9. Introduction to Hypothesis Testing.
Introduction. Basic Concepts of Hypothesis-Testing. One-Sample Tests for the Mean. t Test for the Difference
between the Means of Two Independent Groups. Testing for the Difference between Two Variances. The Repeated Measures
or Paired t Test. Chi-Square Test for the Differences among Proportions in Two or More Groups. X2
Test of Hypothesis for the Variance or Standard Deviation (Optional Topic). Wilcoxon Rank Sum Test for the Difference
between Two Medians (Optional Topic). Summary. Appendix 9.1: Using Microsoft Excel for Hypothesis Testing. Appendix
9.2: Using MINITAB for Hypothesis Testing.
10. The Design of Experiments: One Factor and Randomized Block Experiments.
Introduction and Rationale. Historical Background. The Concept of Randomization. The One-Way Analysis of Variance
(ANOVA). The Randomized Block Model. Kruskal-Wallis Rank Test for Differences in c Medians (Optional Topic).
Appendix 10.1: Using Microsoft Excel for the Analysis of Variance. Appendix 10.2: Using MINITAB for the Analysis
of Variance.
11. The Design of Experiments: Factorial Designs.
Two-Factor Factorial Designs. Factorial Designs Involving Three or More Factors. The Fractional Factorial Design.
The Taguchi Approach. Summary and Overview. Appendix 11.1: Using Microsoft Excel for the Two-Factor Factorial Design.
Appendix 11.2: Using MINITAB for the Two-Factor Factorial Designs.
12. Simple Linear Regression and Correlation.
Introduction. Types of Regression Models. Determining the Simple Linear Regression Equation. Measures of Variation
in Regression and Correlation. Assumptions of Regression and Correlation. Residual Analysis. Inferences about the
Slope. Confidence and Prediction Interval Estimation. Pitfalls in Regression and Ethical Issues. Computations in
Simple Linear Regression. Correlation--Measuring the Strength of the Association. Appendix 12.1: Using Microsoft
Excel for Simple Linear Regression and Correlation. Appendix 12.2: Using MINITAB for Simple Linear Regression and
Correlation.
13. Multiple Regression.
Developing the Multiple-Regression Model. Residual Analysis for the Multiple-Regression Model. Testing for the
Significance of the Multiple-Regression Model. Inferences Concerning the Population Regression Coefficients. Testing
Portions of the Multiple-Regression Model. The Quadratic Curvilinear Regression Model. Dummy-Variable Models. Using
Transformations in Regressions Models. Collinearity. Model-Building. Pitfalls in Multiple Regression. Appendix
13.1: Using Microsoft Excel for Multiple-Regression Models. Appendix 13.2: Using MINITAB for Multiple Models Regression.
Appendices.
Appendix A: Tables. Appendix B: Statistical Forms. Appendix C: Documentation for the Data Files. Appendix D:
Installing the PHStat Microsoft Excel Add-In. Appendix E: Answers to Selected Odd Problems.