TEST BANK FOR PROBABILITY AND STATISTICS FOR ENGINEERING AND THE SCIENCES 8TH EDITION BY DEVORE (Complete Download) All Chapters 1-16

TEST BANK FOR PROBABILITY AND STATISTICS FOR ENGINEERING AND THE SCIENCES 8TH EDITION BY DEVORE (Complete Download) All Chapters 1-16. Test Bank with Complete Questions and Solutions. To clarify, this is the test bank, n ...

TEST BANK FOR PROBABILITY AND STATISTICS FOR ENGINEERING AND THE SCIENCES 8TH EDITION BY DEVORE (Complete Download) All Chapters 1-16. Test Bank with Complete Questions and Solutions. To clarify, this is the test bank, not the textbook. You get immediate access to download your test bank. You will receive a complete test bank; in other words, all chapters will be there. Test banks come in PDF format; therefore, you do not need specialized software to open them. PREVIEW.... Contents 1 Overview and Descriptive Statistics Introduction 1 1.1 Populations, Samples, and Processes 2 1.2 Pictorial and Tabular Methods in Descriptive Statistics 12 1.3 Measures of Location 28 1.4 Measures of Variability 35 Supplementary Exercises 46 Bibliography 49 2 Probability Introduction 50 2.1 Sample Spaces and Events 51 2.2 Axioms, Interpretations, and Properties of Probability 55 2.3 Counting Techniques 64 2.4 Conditional Probability 73 2.5 Independence 83 Supplementary Exercises 88 Bibliography 91 3 Discrete Random Variables and Probability Distributions Introduction 92 3.1 Random Variables 93 3.2 Probability Distributions for Discrete Random Variables 96 3.3 Expected Values 106 3.4 The Binomial Probability Distribution 114 3.5 Hypergeometric and Negative Binomial Distributions 122 3.6 The Poisson Probability Distribution 128 Supplementary Exercises 133 Bibliography 136 vii Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.viii Contents 4 Continuous Random Variables and Probability Distributions Introduction 137 4.1 Probability Density Functions 138 4.2 Cumulative Distribution Functions and Expected Values 143 4.3 The Normal Distribution 152 4.4 The Exponential and Gamma Distributions 165 4.5 Other Continuous Distributions 171 4.6 Probability Plots 178 Supplementary Exercises 188 Bibliography 192 5 Joint Probability Distributions and Random Samples Introduction 193 5.1 Jointly Distributed Random Variables 194 5.2 Expected Values, Covariance, and Correlation 206 5.3 Statistics and Their Distributions 212 5.4 The Distribution of the Sample Mean 223 5.5 The Distribution of a Linear Combination 230 Supplementary Exercises 235 Bibliography 238 6 Point Estimation Introduction 239 6.1 Some General Concepts of Point Estimation 240 6.2 Methods of Point Estimation 255 Supplementary Exercises 265 Bibliography 266 7 Statistical Intervals Based on a Single Sample Introduction 267 7.1 Basic Properties of Confidence Intervals 268 7.2 Large-Sample Confidence Intervals for a Population Mean and Proportion 276 Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.Contents 7.3 Intervals Based on a Normal Population Distribution 285 7.4 Confidence Intervals for the Variance and Standard Deviation of a Normal Population 294 Supplementary Exercises 297 Bibliography 299 8 Tests of Hypotheses Based on a Single Sample Introduction 300 8.1 Hypotheses and Test Procedures 301 8.2 Tests About a Population Mean 310 8.3 Tests Concerning a Population Proportion 323 8.4 P-Values 328 8.5 Some Comments on Selecting a Test 339 Supplementary Exercises 342 Bibliography 344 9 Inferences Based on Two Samples Introduction 345 9.1 z Tests and Confidence Intervals for a Difference Between Two Population Means 346 9.2 The Two-Sample t Test and Confidence Interval 357 9.3 Analysis of Paired Data 365 9.4 Inferences Concerning a Difference Between Population Proportions 375 9.5 Inferences Concerning Two Population Variances 382 Supplementary Exercises 386 Bibliography 390 10 The Analysis of Variance Introduction 391 10.1 Single-Factor ANOVA 392 10.2 Multiple Comparisons in ANOVA 402 10.3 More on Single-Factor ANOVA 408 Supplementary Exercises 417 Bibliography 418 ix Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.x Contents 11 Multifactor Analysis of Variance Introduction 419 11.1 Two-Factor ANOVA with Kij ⫽ 1 420 11.2 Two-Factor ANOVA with Kij ⬎ 1 433 11.3 Three-Factor ANOVA 442 11.4 2p Factorial Experiments 451 Supplementary Exercises 464 Bibliography 467 12 Simple Linear Regression and Correlation Introduction 468 12.1 The Simple Linear Regression Model 469 12.2 Estimating Model Parameters 477 12.3 Inferences About the Slope Parameter ␤1 490 12.4 Inferences Concerning and the Prediction of Future Y Values 499 mY #x* 12.5 Correlation 508 Supplementary Exercises 518 Bibliography 522 13 Nonlinear and Multiple Regression Introduction 523 13.1 Assessing Model Adequacy 524 13.2 Regression with Transformed Variables 531 13.3 Polynomial Regression 543 13.4 Multiple Regression Analysis 553 13.5 Other Issues in Multiple Regression 574 Supplementary Exercises 588 Bibliography 593 14 Goodness-of-Fit Tests and Categorical Data Analysis Introduction 594 14.1 Goodness-of-Fit Tests When Category Probabilities Are Completely Specified 595 Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.Contents 14.2 Goodness-of-Fit Tests for Composite Hypotheses 602 14.3 Two-Way Contingency Tables 613 Supplementary Exercises 621 Bibliography 624 15 Distribution-Free Procedures Introduction 625 15.1 The Wilcoxon Signed-Rank Test 626 15.2 The Wilcoxon Rank-Sum Test 634 15.3 Distribution-Free Confidence Intervals 640 15.4 Distribution-Free ANOVA 645 Supplementary Exercises 649 Bibliography 650 16 Quality Control Methods Introduction 651 16.1 General Comments on Control Charts 652 16.2 Control Charts for Process Location 654 16.3 Control Charts for Process Variation 663 16.4 Control Charts for Attributes 668 16.5 CUSUM Procedures 672 16.6 Acceptance Sampling 680 Supplementary Exercises 686 Bibliography 687 Appendix Tables A.1 Cumulative Binomial Probabilities A-2 A.2 Cumulative Poisson Probabilities A-4 A.3 Standard Normal Curve Areas A-6 A.4 The Incomplete Gamma Function A-8 A.5 Critical Values for t Distributions A-9 A.6 Tolerance Critical Values for Normal Population Distributions A-10 A.7 Critical Values for Chi-Squared Distributions A-11 A.8 t Curve Tail Areas A-12 A.9 Critical Values for F Distributions A-14 A.10 Critical Values for Studentized Range Distributions A-20 Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it. xixii Contents A.11 Chi-Squared Curve Tail Areas A-21 A.12 Critical Values for the Ryan-Joiner Test of Normality A-23 A.13 Critical Values for the Wilcoxon Signed-Rank Test A-24 A.14 Critical Values for the Wilcoxon Rank-Sum Test A-25 A.15 Critical Values for the Wilcoxon Signed-Rank Interval A-26 A.16 Critical Values for the Wilcoxon Rank-Sum Interval A-27 A.17 ␤ Curves for t Tests A-28 Answers to Selected Odd-Numbered Exercises A-29 Glossary of Symbols/Abbreviations G-1 Index I-1 Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.Preface Purpose The use of probability models and statistical methods for analyzing data has become common practice in virtually all scientific disciplines. This book attempts to provide a comprehensive introduction to those models and methods most likely to be encoun- tered and used by students in their careers in engineering and the natural sciences. Although the examples and exercises have been designed with scientists and engi- neers in mind, most of the methods covered are basic to statistical analyses in many other disciplines, so that students of business and the social sciences will also profit from reading the book. Approach Students in a statistics course designed to serve other majors may be initially skeptical of the value and relevance of the subject matter, but my experience is that students can be turned on to statistics by the use of good examples and exercises that blend their every- day experiences with their scientific interests. Consequently, I have worked hard to find examples of real, rather than artificial, data—data that someone thought was worth col- lecting and analyzing. Many of the methods presented, especially in the later chapters on statistical inference, are illustrated by analyzing data taken from published sources, and many of the exercises also involve working with such data. Sometimes the reader may be unfamiliar with the context of a particular problem (as indeed I often was), but I have found that students are more attracted by real problems with a somewhat strange context than by patently artificial problems in a familiar setting. Mathematical Level The exposition is relatively modest in terms of mathematical development. Substantial use of the calculus is made only in Chapter 4 and parts of Chapters 5 and 6. In particu- lar, with the exception of an occasional remark or aside, calculus appears in the inference part of the book only—in the second section of Chapter 6. Matrix algebra is not used at all. Thus almost all the exposition should be accessible to those whose mathematical background includes one semester or two quarters of differential and integral calculus. Content Chapter 1 begins with some basic concepts and terminology—population, sample, descriptive and inferential statistics, enumerative versus analytic studies, and so on— and continues with a survey of important graphical and numerical descriptive methods. A rather traditional development of probability is given in Chapter 2, followed by prob- ability distributions of discrete and continuous random variables in Chapters 3 and 4, respectively. Joint distributions and their properties are discussed in the first part of Chapter 5. The latter part of this chapter introduces statistics and their sampling distri- butions, which form the bridge between probability and inference. The next three chapters cover point estimation, statistical intervals, and hypothesis testing based on a single sample. Methods of inference involving two independent samples and paired data are presented in Chapter 9. The analysis of variance is the subject of Chapters 10 and 11 (single-factor and multifactor, respectively). Regression makes its initial appearance in Chapter 12 (the simple linear regression model and correlation) and xiii Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.xiv Preface returns for an extensive encore in Chapter 13. The last three chapters develop chi- squared methods, distribution-free (nonparametric) procedures, and techniques from statistical quality control. Helping Students Learn Although the book’s mathematical level should give most science and engineering students little difficulty, working toward an understanding of the concepts and gain- ing an appreciation for the logical development of the methodology may sometimes require substantial effort. To help students gain such an understanding and appreci- ation, I have provided numerous exercises ranging in difficulty from many that involve routine application of text material to some that ask the reader to extend con- cepts discussed in the text to somewhat new situations. There are many more exer- cises than most instructors would want to assign during any particular course, but I recommend that students be required to work a substantial number of them; in a problem-solving discipline, active involvement of this sort is the surest way to iden- tify and close the gaps in understanding that inevitably arise. Answers to most odd- numbered exercises appear in the answer section at the back of the text. In addition, a Student Solutions Manual, consisting of worked-out solutions to virtually all the odd-numbered exercises, is available. To access additional course materials and companion resources, please visit www.cengagebrain.com. At the CengageBrain.com home page, search for the ISBN of your title (from the back cover of your book) using the search box at the top of the page. This will take you to the product page where free companion resources can be found. New for This Edition • A Glossary of Symbols/Abbreviations appears at the end of the book (the author apologizes for his laziness in not getting this together for earlier editions!) and a small set of sample exams appears on the companion website (available at www.cengage.com/login). • Many new examples and exercises, almost all based on real data or actual prob- lems. Some of these scenarios are less technical or broader in scope than what has been included in previous editions—for example, weights of football players (to illustrate multimodality), fundraising expenses for charitable organizations, and the comparison of grade point averages for classes taught by part-time faculty with those for classes taught by full-time faculty. • The material on P-values has been substantially rewritten. The P-value is now ini- tially defined as a probability rather than as the smallest significance level for which the null hypothesis can be rejected. A simulation experiment is presented to illustrate the behavior of P-values. • Chapter 1 contains a new subsection on “The Scope of Modern Statistics” to indicate how statisticians continue to develop new methodology while working on problems in a wide spectrum of disciplines. • The exposition has been polished whenever possible to help students gain an intuitive understanding of various concepts. For example, the cumulative distribution function is more deliberately introduced in Chapter 3, the first example of maximum likeli- hood in Section 6.2 contains a more careful discussion of likelihood, more attention is given to power and type II error probabilities in Section 8.3, and the material on residuals and sums of squares in multiple regression is laid out more explicitly in Section 13.4. Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.Preface Acknowledgments My colleagues at Cal Poly have provided me with invaluable support and feedback over the years. I am also grateful to the many users of previous editions who have made suggestions for improvement (and on occasion identified errors). A special note of thanks goes to Matt Carlton for his work on the two solutions manuals, one for instructors and the other for students. The generous feedback provided by the following reviewers of this and previous editions has been of great benefit in improving the book: Robert L. Armacost, University of Central Florida; Bill Bade, Lincoln Land Community College; Douglas M. Bates, University of Wisconsin–Madison; Michael Berry, West Virginia Wesleyan College; Brian Bowman, Auburn University; Linda Boyle, University of Iowa; Ralph Bravaco, Stonehill College; Linfield C. Brown, Tufts University; Karen M. Bursic, University of Pittsburgh; Lynne Butler, Haverford College; Raj S. Chhikara, University of Houston–Clear Lake; Edwin Chong, Colorado State University; David Clark, California State Polytechnic University at Pomona; Ken Constantine, Taylor University; David M. Cresap, University of Portland; Savas Dayanik, Princeton University; Don E. Deal, University of Houston; Annjanette M. Dodd, Humboldt State University; Jimmy Doi, California Polytechnic State University–San Luis Obispo; Charles E. Donaghey, University of Houston; Patrick J. Driscoll, U.S. Military Academy; Mark Duva, University of Virginia; Nassir Eltinay, Lincoln Land Community College; Thomas English, College of the Mainland; Nasser S. Fard, Northeastern University; Ronald Fricker, Naval Postgraduate School; Steven T. Garren, James Madison University; Mark Gebert, University of Kentucky; Harland Glaz, University of Maryland; Ken Grace, Anoka-Ramsey Community College; Celso Grebogi, University of Maryland; Veronica Webster Griffis, Michigan Technological University; Jose Guardiola, Texas A&M University–Corpus Christi; K. L. D. Gunawardena, University of Wisconsin–Oshkosh; James J. Halavin, Rochester Institute of Technology; James Hartman, Marymount University; Tyler Haynes, Saginaw Valley State University; Jennifer Hoeting, Colorado State University; Wei-Min Huang, Lehigh University; Aridaman Jain, New Jersey Institute of Technology; Roger W. Johnson, South Dakota School of Mines & Technology; Chihwa Kao, Syracuse University; Saleem A. Kassam, University of Pennsylvania; Mohammad T. Khasawneh, State University of NewYork–Binghamton; Stephen Kokoska, Colgate University; Hillel J. Kumin, University of Oklahoma; Sarah Lam, Binghamton University; M. Louise Lawson, Kennesaw State University; Jialiang Li, University of Wisconsin–Madison; Wooi K. Lim, William Paterson University; Aquila Lipscomb, The Citadel; Manuel Lladser, University of Colorado at Boulder; Graham Lord, University of California–Los Angeles; Joseph L. Macaluso, DeSales University; Ranjan Maitra, Iowa State University; David Mathiason, Rochester Institute of Technology; Arnold R. Miller, University of Denver; John J. Millson, University of Maryland; Pamela Kay Miltenberger, West Virginia Wesleyan College; Monica Molsee, Portland State University; Thomas Moore, Naval Postgraduate School; Robert M. Norton, College of Charleston; Steven Pilnick, Naval Postgraduate School; Robi Polikar, Rowan University; Ernest Pyle, Houston Baptist University; Steve Rein, California Polytechnic State University–San Luis Obispo; Tony Richardson, University of Evansville; Don Ridgeway, North Carolina State University; Larry J. Ringer, Texas A&M University; Robert M. Schumacher, Cedarville University; Ron Schwartz, Florida Atlantic University; Kevan Shafizadeh, California State University–Sacramento; Mohammed Shayib, Prairie View A&M; Robert K. Smidt, California Polytechnic State University–San Luis Obispo; Alice E. Smith, Auburn University; James MacGregor Smith, University of Massachusetts; Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it. xvxvi Preface Paul J. Smith, University of Maryland; Richard M. Soland, The George Washington University; Clifford Spiegelman, Texas A&M University; Jery Stedinger, Cornell University; David Steinberg, Tel Aviv University; William Thistleton, State University of New York Institute of Technology; G. Geoffrey Vining, University of Florida; Bhutan Wadhwa, Cleveland State University; Gary Wasserman, Wayne State University; Elaine Wenderholm, State University of New York–Oswego; Samuel P. Wilcock, Messiah College; Michael G. Zabetakis, University of Pittsburgh; and Maria Zack, Point Loma Nazarene University. Danielle Urban of Elm Street Publishing Services has done a terrific job of supervising the book's production. Once again I am compelled to express my grat- itude to all those people at Cengage who have made important contributions over the course of my textbook writing career. For this most recent edition, special thanks go to Jay Campbell (for his timely and informed feedback throughout the project), Molly Taylor, Shaylin Walsh, Ashley Pickering, Cathy Brooks, and Andrew Coppola. I also greatly appreciate the stellar work of all those Cengage Learning sales representatives who have labored to make my books more visible to the statistical community. Last but by no means least, a heartfelt thanks to my wife Carol for her decades of support, and to my daughters for providing inspiration through their own achievements. Jay Devore Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.1 Overview and Descriptive Statistics “I am not much given to regret, so I puzzled over this one a while. Should have taken much more statistics in college, I think.” —Max Levchin, Paypal Co-founder, Slide Founder Quote of the week from the Web site of the American Statistical Association on November 23, 2010 “I keep saying that the sexy job in the next 10 years will be statisticians, and I’m not kidding.” —Hal Varian, Chief Economist at Google August 6, 2009, The New York Times INTRODUCTION Statistical concepts and methods are not only useful but indeed often indis- pensable in understanding the world around us. They provide ways of gaining new insights into the behavior of many phenomena that you will encounter in your chosen field of specialization in engineering or science. The discipline of statistics teaches us how to make intelligent judgments and informed decisions in the presence of uncertainty and variation. Without uncertainty or variation, there would be little need for statistical methods or stat- isticians. If every component of a particular type had exactly the same lifetime, if all resistors produced by a certain manufacturer had the same resistance value, if pH determinations for soil specimens from a particular locale gave identical results, and so on, then a single observation would reveal all desired information. An interesting manifestation of variation arises in the course of performing emissions testing on motor vehicles. The expense and time requirements of the Federal Test Procedure (FTP) preclude its widespread use in vehicle inspection pro- grams. As a result, many agencies have developed less costly and quicker tests, which it is hoped replicate FTP results. According to the journal article “Motor 1 Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.2 CHAPTER 1 Overview and Descriptive Statistics Vehicle Emissions Variability” (J. of the Air and Waste Mgmt. Assoc., 1996: 667–675), the acceptance of the FTP as a gold standard has led to the widespread belief that repeated measurements on the same vehicle would yield identical (or nearly identical) results. The authors of the article applied the FTP to seven vehicles characterized as “high emitters.” Here are the results for one such vehicle: HC (gm/mile) 13.8 18.3 32.2 32.5 CO (gm/mile) 118 149 232 236 The substantial variation in both the HC and CO measurements casts consider- able doubt on conventional wisdom and makes it much more difficult to make precise assessments about emissions levels. How can statistical techniques be used to gather information and draw conclusions? Suppose, for example, that a materials engineer has developed a coating for retarding corrosion in metal pipe under specified circumstances. If this coating is applied to different segments of pipe, variation in environmental conditions and in the segments themselves will result in more substantial cor- rosion on some segments than on others. Methods of statistical analysis could be used on data from such an experiment to decide whether the average amount of corrosion exceeds an upper specification limit of some sort or to pre- dict how much corrosion will occur on a single piece of pipe. Alternatively, suppose the engineer has developed the coating in the belief that it will be superior to the currently used coating. A comparative experiment could be carried out to investigate this issue by applying the current coating to some segments of pipe and the new coating to other segments. This must be done with care lest the wrong conclusion emerge. For example, perhaps the aver- age amount of corrosion is identical for the two coatings. However, the new coating may be applied to segments that have superior ability to resist corrosion and under less stressful environmental conditions compared to the segments and conditions for the current coating. The investigator would then likely observe a difference between the two coatings attributable not to the coatings themselves, but just to extraneous variation. Statistics offers not only methods for analyzing the results of experiments once they have been carried out but also suggestions for how experiments can be performed in an efficient manner to mitigate the effects of variation and have a better chance of producing correct conclusions. 1.1 Populations, Samples, and Processes Engineers and scientists are constantly exposed to collections of facts, or data, both in their professional capacities and in everyday activities. The discipline of statistics provides methods for organizing and summarizing data and for drawing conclusions based on information contained in the data. Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.1.1 Populations, Samples, and Processes 3 An investigation will typically focus on a well-defined collection of objects constituting a population of interest. In one study, the population might consist of all gelatin capsules of a particular type produced during a specified period. Another investigation might involve the population consisting of all individuals who received a B.S. in engineering during the most recent academic year. When desired informa- tion is available for all objects in the population, we have what is called a census. Constraints on time, money, and other scarce resources usually make a census impractical or infeasible. Instead, a subset of the population—a sample—is selected in some prescribed manner. Thus we might obtain a sample of bearings from a par- ticular production run as a basis for investigating whether bearings are conforming to manufacturing specifications, or we might select a sample of last year’s engineer- ing graduates to obtain feedback about the quality of the engineering curricula. We are usually interested only in certain characteristics of the objects in a pop- ulation: the number of flaws on the surface of each casing, the thickness of each cap- sule wall, the gender of an engineering graduate, the age at which the individual graduated, and so on. A characteristic may be categorical, such as gender or type of malfunction, or it may be numerical in nature. In the former case, the value of the characteristic is a category (e.g., female or insufficient solder), whereas in the latter case, the value is a number (e.g., age 5 23 years or diameter 5 .502 cm ). A variable is any characteristic whose value may change from one object to another in the population. We shall initially denote variables by lowercase letters from the end of our alphabet. Examples include z 5 braking distance of an automobile under specified conditions y 5 number of visits to a particular Web site during a specified period x 5 brand of calculator owned by a student Data results from making observations either on a single variable or simultaneously on two or more variables. A univariate data set consists of observations on a single variable. For example, we might determine the type of transmission, automatic (A) or manual (M), on each of ten automobiles recently purchased at a certain dealer- ship, resulting in the categorical data set M A A A M A A M A A The following sample of lifetimes (hours) of brand D batteries put to a certain use is a numerical univariate data set: 5.6 5.1 6.2 6.0 5.8 6.5 5.8 5.5 We have bivariate data when observations are made on each of two variables. Our data set might consist of a (height, weight) pair for each basketball player on a team, with the first observation as (72, 168), the second as (75, 212), and so on. If an engineer determines the value of both x 5 component lifetime and y 5 reason for component failure, the resulting data set is bivariate with one variable numeri- cal and the other categorical. Multivariate data arises when observations are made on more than one variable (so bivariate is a special case of multivariate). For exam- ple, a research physician might determine the systolic blood pressure, diastolic blood pressure, and serum cholesterol level for each patient participating in a study. Each observation would be a triple of numbers, such as (120, 80, 146). In many multivariate data sets, some variables are numerical and others are categorical. Thus the annual automobile issue of Consumer Reports gives values of such variables as type of vehicle (small, sporty, compact, mid-size, large), city fuel efficiency (mpg), highway fuel efficiency (mpg), drivetrain type (rear wheel, front wheel, four wheel), and so on. Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.4 CHAPTER 1 Overview and Descriptive Statistics Branches of Statistics An investigator who has collected data may wish simply to summarize and describe important features of the data. This entails using methods from descriptive statistics. Some of these methods are graphical in nature; the construction of histograms, boxplots, and scatter plots are primary examples. Other descriptive methods involve calculation of numerical summary measures, such as means, standard deviations, and correlation coefficients. The wide availability of statistical computer software packages has made these tasks much easier to carry out than they used to be. Computers are much more efficient than human beings at calculation and the creation of pictures (once they have received appropriate instructions from the user!). This means that the investigator doesn’t have to expend much effort on “grunt work” and will have more time to study the data and extract important messages. Throughout this book, we will present output from various packages such as Minitab, SAS, S-Plus, and R. The R software can be downloaded without charge from the site http://www.r-project.org. Example 1.1 Charity is a big business in the United States. The Web site charitynavigator.com gives information on roughly 5500 charitable organizations, and there are many smaller charities that fly below the navigator’s radar screen. Some charities operate very efficiently, with fundraising and administrative expenses that are only a small percentage of total expenses, whereas others spend a high percentage of what they take in on such activities. Here is data on fundraising expenses as a percentage of total expenditures for a random sample of 60 charities: 6.1 12.6 34.7 1.6 18.8 2.2 3.0 2.2 5.6 3.8 2.2 3.1 1.3 1.1 14.1 4.0 21.0 6.1 1.3 20.4 7.5 3.9 10.1 8.1 19.5 5.2 12.0 15.8 10.4 5.2 6.4 10.8 83.1 3.6 6.2 6.3 16.3 12.7 1.3 0.8 8.8 5.1 3.7 26.3 6.0 48.0 8.2 11.7 7.2 3.9 15.3 16.6 8.8 12.0 4.7 14.7 6.4 17.0 2.5 16.2 Without any organization, it is difficult to get a sense of the data’s most prominent features—what a typical (i.e. representative) value might be, whether values are highly concentrated about a typical value or quite dispersed, whether there are any Stem–and–leaf of FundRsng N = 60 Leaf Unit = 1.0 0 0111112222333333344 0 55556666666778888 1 0001222244 1 55666789 2 01 2 6 3 3 4 4 8 5 5 6 6 7 7 8 3 4 40 30 20 10 0 0 10 20 30 40 50 FundRsng 60 70 80 90 Figure 1.1 A Minitab stem-and-leaf display (tenths digit truncated) and histogram for the charity fundraising percentage data Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it. Frequency1.1 Populations, Samples, and Processes 5 gaps in the data, what fraction of the values are less than 20%, and so on. Figure 1.1 shows what is called a stem-and-leaf display as well as a histogram. In Section 1.2 we will discuss construction and interpretation of these data summaries. For the moment, we hope you see how they begin to describe how the percentages are dis- tributed over the range of possible values from 0 to 100. Clearly a substantial major- ity of the charities in the sample spend less than 20% on fundraising, and only a few percentages might be viewed as beyond the bounds of sensible practice. ■ Having obtained a sample from a population, an investigator would frequently like to use sample information to draw some type of conclusion (make an inference of some sort) about the population. That is, the sample is a means to an end rather than an end in itself. Techniques for generalizing from a sample to a population are gathered within the branch of our discipline called inferential statistics. Example 1.2 Material strength investigations provide a rich area of application for statistical meth- ods. The article “Effects of Aggregates and Microfillers on the Flexural Properties of Concrete” (Magazine of Concrete Research, 1997: 81–98) reported on a study of strength properties of high-performance concrete obtained by using superplasticizers and certain binders. The compressive strength of such concrete had previously been investigated, but not much was known about flexural strength (a measure of ability to resist failure in bending). The accompanying data on flexural strength (in MegaPascal, MPa, where 1 Pa (Pascal) 5 1.45 3 1024 psi ) appeared in the article cited: 5.9 7.2 7.3 6.3 8.1 6.8 7.0 7.6 6.8 6.5 7.0 6.3 7.9 9.0 8.2 8.7 7.8 9.7 7.4 7.7 9.7 7.8 7.7 11.6 11.3 11.8 10.7 Suppose we want an estimate of the average value of flexural strength for all beams that could be made in this way (if we conceptualize a population of all such beams, we are trying to estimate the population mean). It can be shown that, with a high degree of confidence, the population mean strength is between 7.48 MPa and 8.80 MPa; we call this a confidence interval or interval estimate. Alternatively, this data could be used to predict the flexural strength of a single beam of this type. With a high degree of confidence, the strength of a single such beam will exceed 7.35 MPa; the number 7.35 is called a lower prediction bound. ■ The main focus of this book is on presenting and illustrating methods of infer- ential statistics that are useful in scientific work. The most important types of infer- ential procedures—point estimation, hypothesis testing, and estimation by confidence intervals—are introduced in Chapters 6–8 and then used in more com- plicated settings in Chapters 9–16. The remainder of this chapter presents methods from descriptive statistics that are most used in the development of inference. Chapters 2–5 present material from the discipline of probability. This material ultimately forms a bridge between the descriptive and inferential techniques. Mastery of probability leads to a better understanding of how inferential procedures are developed and used, how statistical conclusions can be translated into everyday language and interpreted, and when and where pitfalls can occur in applying the methods. Probability and statistics both deal with questions involving populations and samples, but do so in an “inverse manner” to one another. In a probability problem, properties of the population under study are assumed known (e.g., in a numerical population, some specified distribution of the population values may be assumed), and questions regarding a sample taken from the population are posed and answered. In a statistics problem, characteristics of a Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.6 CHAPTER 1 Overview and Descriptive Statistics Probability Population Inferential statistics Figure 1.2 The relationship between probability and inferential statistics sample are available to the experimenter, and this information enables the experi- menter to draw conclusions about the population. The relationship between the two disciplines can be summarized by saying that probability reasons from the population to the sample (deductive reasoning), whereas inferential statistics rea- sons from the sample to the population (inductive reasoning). This is illustrated in Figure 1.2. Before we can understand what a particular sample can tell us about the pop- ulation, we should first understand the uncertainty associated with taking a sample from a given population. This is why we study probability before statistics. Example 1.3 As an example of the contrasting focus of probability and inferential statistics, con- sider drivers’ use of manual lap belts in cars equipped with automatic shoulder belt systems. (The article “Automobile Seat Belts: Usage Patterns in Automatic Belt Systems,” Human Factors, 1998: 126–135, summarizes usage data.) In probability, we might assume that 50% of all drivers of cars equipped in this way in a certain metropolitan area regularly use their lap belt (an assumption about the population), so we might ask, “How likely is it that a sample of 100 such drivers will include at least 70 who regularly use their lap belt?” or “How many of the drivers in a sample of size 100 can we expect to regularly use their lap belt?” On the other hand, in infer- ential statistics, we have sample information available; for example, a sample of 100 drivers of such cars revealed that 65 regularly use their lap belt. We might then ask, “Does this provide substantial evidence for concluding that more than 50% of all such drivers in this area regularly use their lap belt?” In this latter scenario, we are attempting to use sample information to answer a question about the structure of the entire population from which the sample was selected. ■ In the foregoing lap belt example, the population is well defined and concrete: all drivers of cars equipped in a certain way in a particular metropolitan area. In Example 1.2, however, the strength measurements came from a sample of prototype beams that had not been selected from an existing population. Instead, it is conven- ient to think of the population as consisting of all possible strength measurements that might be made under similar experimental conditions. Such a population is referred to as a conceptual or hypothetical population. There are a number of prob- lem situations in which we fit questions into the framework of inferential statistics by conceptualizing a population. The Scope of Modern Statistics These days statistical methodology is employed by investigators in virtually all dis- ciplines, including such areas as • molecular biology (analysis of microarray data) • ecology (describing quantitatively how individuals in various animal and plant populations are spatially distributed) Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it. Sample1.1 Populations, Samples, and Processes 7 • materials engineering (studying properties of various treatments to retard corrosion) • marketing (developing market surveys and strategies for marketing new products) • public health (identifying sources of diseases and ways to treat them) • civil engineering (assessing the effects of stress on structural elements and the impacts of traffic flows on communities) As you progress through the book, you’ll encounter a wide spectrum of different sce- narios in the examples and exercises that illustrate the application of techniques from probability and statistics. Many of these scenarios involve data or other material extracted from articles in engineering and science journals. The methods presented herein have become established and trusted tools in the arsenal of those who work with data. Meanwhile, statisticians continue to develop new models for describing random- ness, and uncertainty and new methodology for analyzing data. As evidence of the con- tinuing creative efforts in the statistical community, here are titles and capsule descriptions of some articles that have recently appeared in statistics journals (Journal of the American Statistical Association is abbreviated JASA, and AAS is short for the Annals of Applied Statistics, two of the many prominent journals in the discipline): • “Modeling Spatiotemporal Forest Health Monitoring Data” (JASA, 2009: 899–911): Forest health monitoring systems were set up across Europe in the 1980s in response to concerns about air-pollution-related forest dieback, and have continued operation with a more recent focus on threats from climate change and increased ozone levels. The authors develop a quantitative descrip- tion of tree crown defoliation, an indicator of tree health. • “Active Learning Through Sequential Design, with Applications to the Detection of Money Laundering” (JASA, 2009: 969–981): Money laundering involves con- cealing the origin of funds obtained through illegal activities. The huge number of transactions occurring daily at financial institutions makes detection of money laundering difficult. The standard approach has been to extract various summary quantities from the transaction history and conduct a time-consuming investiga- tion of suspicious activities. The article proposes a more efficient statistical method and illustrates its use in a case study. • “Robust Internal Benchmarking and False Discovery Rates for Detecting Racial Bias in Police Stops” (JASA, 2009: 661–668): Allegations of police actions that are attributable at least in part to racial bias have become a contentious issue in many communities. This article proposes a new method that is designed to reduce the risk of flagging a substantial number of “false positives” (individuals falsely identified as manifesting bias). The method was applied to data on 500,000 pedestrian stops in New York City in 2006; of the 3000 officers regu- larly involved in pedestrian stops, 15 were identified as having stopped a sub- stantially greater fraction of Black and Hispanic people than what would be predicted were bias absent. • “Records in Athletics Through Extreme Value Theory” (JASA, 2008: 1382–1391): The focus here is on the modeling of extremes related to world records in athletics. The authors start by posing two questions: (1) What is the ultimate world record within a specific event (e.g. the high jump for women)? and (2) How “good” is the current world record, and how does the quality of current world records compare across different events? A total of 28 events (8 running, 3 throwing, and 3 jumping for both men and women) are considered. For example, one conclusion is that only about 20 seconds can be shaved off the Copyright 2010 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. Due to electronic rights, some third party content may be suppressed from the eBook and/or eChapter(s). Editorial review has deemed that any suppressed content does not materially affect the overall learning experience. Cengage Learning reserves the right to remove additional content at any time if subsequent rights restrictions require it.8 CHAPTER 1 Overview and Descriptive Statistics men’s marathon record, but that the current women’s marathon record is almost 5 minutes longer than what can ultimately be achieved. The methodology also has applications to such issues as ensuring airport runways are long enough and that dikes in Holland are high enough. • “Analysis of Episodic Data with Application to Recurrent Pulmonary Exacerbations in Cystic Fibrosis Patients” (JASA, 2008: 498–510): The analysis of recurrent medical events such as migraine headaches should take into account not only when such events first occur but also how long they last—length of episodes may contain important information about the severity of the disease or malady, associated medical costs, and the quality of life. The article proposes a technique that summarizes both epis