Final PDF to printer Applied Statistics in Business and Economics Fifth Edition ddooaa3377330044__ffmm__ii--11..iinndddd ii 1122//3300//1144 22::2255 PPMM Final PDF to printer ddooaa3377330044__ffmm__ii--11..iinndddd iiii 1122//3300//1144 22::2255 PPMM Final PDF to printer Applied Statistics in Business and Economics Fifth Edition David P. Doane Oakland University Lori E. Seward University of Colorado ddooaa3377330044__ffmm__ii--11..iinndddd iiiiii 1122//3300//1144 22::2255 PPMM Final PDF to printer APPLIED STATISTICS IN BUSINESS AND ECONOMICS, FIFTH EDITION Published by McGraw-Hill Education, 2 Penn Plaza, New York, NY 10121. Copyright © 2016 by McGraw-Hill Education. All rights reserved. Printed in the United States of America. Previous editions © 2013, 2011, and 2009. No part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written consent of McGraw-Hill Education, including, but not limited to, in any network or other electronic storage or transmission, or broadcast for distance learning. Some ancillaries, including electronic and print components, may not be available to customers outside the United States. This book is printed on acid-free paper. 1 2 3 4 5 6 7 8 9 0 DOW/DOW 1 0 9 8 7 6 5 ISBN 978-0-07-783730-3 MHID 0-07-783730-4 Senior Vice President, Products & Markets: Kurt L. Strand Vice President, General Manager, Products & Markets: Marty Lange Vice President, Content Design & Delivery: Kimberly Meriwether David Managing Director: James Heine Director, Product Development: Rose Koos Lead Product Developer: Michele Janicek Product Developer: Christina Holt Executive Marketing Manager: Melissa Caughlin Director, Digital Content Development: Douglas Ruby Digital Product Analyst: Kevin Shanahan Director, Content Design & Delivery: Linda Avenarius Program Manager: Mark Christianson Content Project Managers: Harvey Yep (Core); Bruce Gin (Assessment) Buyer: Jennifer Pickel Design: Matt Diamond Content Licensing Specialists: Keri Johnson (Image); Rita Hingtgen (Text) Cover Image: Tetra Images – Eric Isakson Compositor: Laserwords Private Limited Printer: R. R. Donnelley All credits appearing on page or at the end of the book are considered to be an extension of the copyright page. Library of Congress Cataloging-in-Publication Data Doane, David P. Applied statistics in business and economics / David P. Doane, Lori E. Seward. — Fifth edition. pages cm ISBN 978-0-07-783730-3 (alk. paper) 1. Commercial statistics. 2. Management—Statistical methods. 3. Economics—Statistical methods. 4. Statistics. I. Seward, Lori Welte, 1962- II. Title. HF1017.D55 2016 519.5—dc23 2014044939 The Internet addresses listed in the text were accurate at the time of publication. The inclusion of a website does not indicate an endorsement by the authors or McGraw-Hill Education, and McGraw-Hill Education does not guarantee the accuracy of the information presented at these sites. www.mhhe.com ddooaa3377330044__ffmm__ii--11..iinndddd iivv 1122//3300//1144 22::2255 PPMM Final PDF to printer ABOUT THE AUTHORS David P. Doane David P. Doane is a Certified Professional Statistician (PStatTM) by the American Statistical Association. He is professor emeritus in Oakland University’s Department of Decision and Information Sciences. He earned his Bachelor of Arts degree in mathematics and economics at the University of Kansas and his PhD from Purdue University’s Krannert Graduate School. His research and teaching interests include applied statistics, forecasting, and statistical education. He is corecipient of three National Science Foundation grants to develop software to teach statistics and to create a computer classroom. He is a longtime member of the American Statistical Association, serving in 2002 as president of the Detroit ASA. He has consulted with government, health care organizations, and local firms. He has published articles in many academic journals and is the co-author of Visual Statistics (McGraw-Hill, 1997, 2001). Lori E. Seward Lori E. Seward is a senior instructor of Operations Management in The Leeds School of Business at the University of Colorado in Boulder. She earned her Bachelor of Science and Master of Science degrees in Industrial Engineering at Virginia Tech. After several years work- ing as a reliability and quality engineer in the paper and automotive industries, she earned her PhD from Virginia Tech and joined the faculty at The Leeds School in 1998. She has been the coordinator of the undergraduate core business statistics course and currently teaches the core MBA statistics course. She is also responsible for coordinating the undergraduate program in Operations Management. She served as the chair of the INFORMS Teachers’ Workshop for the annual 2004 meeting. Her teaching interests focus on developing pedagogy that uses technology to create a collaborative learning environment in large undergraduate and MBA statistics courses. Her most recent article, co-authored with David Doane, was published in the Journal of Statistics Education (2011). DEDICATION To Robert Hamilton Doane-Solomon David To all my students who challenged me to make statistics relevant to their lives. Lori v ddooaa3377330044__ffmm__ii--11..iinndddd vv 1122//3300//1144 22::2255 PPMM Final PDF to printer FROM THE “How often have you heard people/students say about a particular subject, ‘I’ll never use this in the real world’? I thought statistics was a bit on the ‘math-geeky’ side at first. Imagine my horror when I saw α, R2, and correlations on several financial reports at my current job (an intern position at a financial services company). I realized then that I had better try to understand some of this stuff.” —Jill Odette (an introductory statistics student) As recently as a decade ago our students used to ask us, “How do I use statistics?” Today we more often hear, “Why should I use statistics?” Applied Statistics in Business and Economics has attempted to provide real meaning to the use of statistics in our world by using real business situations and real data and appealing to your need to know why rather than just how. With over 50 years of teaching statistics between the two of us, we feel we have something to offer. Seeing how students have changed as the new century unfolds has required us to adapt and seek out better ways of instruction. So we wrote Applied Statistics in Business and Economics to meet four distinct objectives. Objective 1: Communicate the Meaning of Variation in a Business Context Variation exists everywhere in the world around us. Successful businesses know how to measure variation. They also know how to tell when variation should be responded to and when it should be left alone. We’ll show how businesses do this. Objective 2: Use Real Data and Real Business Applications Examples, case studies, and problems are taken from published research or real applications whenever possible. Hypothetical data are used when it seems the best way to illustrate a concept. You can usually tell the difference by examining the footnotes citing the source. Objective 3: Incorporate Current Statistical Practices and Offer Practical Advice With the increased reliance on computers, statistics practitioners have changed the way they use statistical tools. We’ll show the current practices and explain why they are used the way they are. We will also tell you when each technique should not be used. Objective 4: Provide More In-Depth Explanation of the Why and Let the Software Take Care of the How It is critical to understand the importance of communicating with data. Today’s computer capabilities make it much easier to summarize and display data than ever b efore. We demonstrate easily mastered software techniques using the common s oftware available. We also spend a great deal of time on the idea that there are risks in decision making and those risks should be quantified and directly considered in every business decision. Our experience tells us that students want to be given credit for the experience they bring to the college classroom. We have tried to honor this by choosing examples and exercises set in situations that will draw on students’ already vast knowledge of the world and knowledge gained from other classes. Emphasis is on thinking about data, choosing appropriate analytic tools, using computers effectively, and recognizing limitations of statistics. What’s New in This Fifth Edition? In this fifth edition we have listened to you and have made many changes that you asked for. We sought advice from students and faculty who are currently using the textbook, reviewers at a variety of colleges and universities, and participants in focus groups on teaching statistics with technology. At the end of this preface is a detailed list of chapter- by-chapter improve- ments, but here are just a few of them: • Step-by-step instructions and new screen shots on using Excel 2013 for descriptive statistics, histograms, scatter plots, line charts, and pivot tables. • Updated exercises with emphasis on compatibility with Connect. • Updated test bank questions matched with topics and learning objectives. • Addition of topics requested by reviewers, including more on logistic regression. • Rewritten instructor’s manual with step-by-step solutions. • New and updated Mini Cases for economics and business. • Improved explanations of data types, random sampling, probability, and distributions. • Streamlined sections on Types I and II error, hypothesis formulation, and decision rules. • More focus on one-factor ANOVA and interpreting ANOVA results. vi ddooaa3377330044__ffmm__ii--11..iinndddd vvii 1122//3300//1144 22::2255 PPMM Final PDF to printer AUTHORS • Clarification of Wilcoxon rank sum test (Mann–Whitney test) with illustration of two versions. • End of each chapter guides to downloads from Connect® (simulations, demonstrations, tips, and video tutorials for Excel, MegaStat, and MINITAB). Software Excel is used throughout this book because it is available everywhere. Some calculations are illustrated using MegaStat, an Excel add-in whose Excel-based menus and spreadsheet format offer more capability than Excel’s Data Analysis Tools. MINITAB menus and examples are also included to point out similarities and differences of these tools. To assist students who need extra help or “catch up” work, the text website contains tutorials or demonstrations on using Excel, MINITAB, or MegaStat for the tasks of each chapter. At the end of each chapter is a list of LearningStats demonstrations that illustrate the concepts from the chapter. These demonstrations can be found in the Connect product for this text. Math Level The assumed level of mathematics is pre-calculus, though there are rare references to calculus where it might help the better-trained reader. All but the simplest proofs and derivations are omitted, though key assumptions are stated clearly. The learner is advised what to do when these assumptions are not fulfilled. Worked examples are included for basic calcula- tions, but the textbook does assume that computers will do the calculations after the statistics class is over, so, interpreta- tion is paramount. End-of-chapter references and suggested websites are given so that interested readers can deepen their understanding. Exercises Simple practice exercises are placed within each section. End-of-chapter exercises tend to be more integrative or to be embed- ded in more realistic contexts. Attention has been given to revising exercises so that they have clear-cut answers that are matched to specific learning objectives. A few exercises invite short answers rather than just quoting a formula. Answers to most odd-numbered exercises are in the back of the book (all of the answers are in the instructor’s manual). LearningStats Connect users can access LearningStats, a collection of Excel spreadsheets, Word documents, and PowerPoints for each chapter. It is intended to let students explore data and concepts at their own pace, ignoring material they already know and focusing on things that interest them. LearningStats includes explanations on topics that are not covered in other software packages, such as how to write effective reports, how to perform calculations, or how to make effective charts. It also includes topics that did not appear prominently in the textbook (e.g., partial F-test, Durbin–Watson test, sign test, bootstrap simulation, and logistic regression). Instructors can use LearningStats PowerPoint presentations in the classroom, but Connect users can also use them for self-instruction. No instructor can “cover everything,” but students can be encouraged to explore LearningStats data sets and/ or demonstrations perhaps with an instructor’s guidance. David P. Doane Lori E. Seward vii ddooaa3377330044__ffmm__ii--11..iinndddd vviiii 1122//3300//1144 22::2255 PPMM Final PDF to printer HOW ARE CHAPTERS ORGANIZED Chapter Contents CHAPTER CONTENTS Each chapter begins with a short list of sec- tion topics that are covered in the chapter. 1.1 What Is Statistics? 1.2 Why Study Statistics? 1.3 Statistics in Business 1.4 Statistical Challenges Chapter Learning Objectives 1.5 Critical Thinking Each chapter includes a list of learning objectives students should be able to attain CHAPTER LEARNING OBJECTIVES upon reading and studying the chapter When you fi nish this chapter you should be able to material. Learning objectives give students LO 1-1 Defi ne statistics and explain some of its uses. an overview of what is expected and iden- LO 1-2 List reasons for a business student to study statistics. tify the goals for learning. Learning objec- LO 1-3 Explain the uses of statistics in business. tives also appear next to chapter topics in LO 1-4 State the common challenges facing business professionals using statistics. the margins. LO 1-5 List and explain common statistical pitfalls. Section Exercises SECTION EXERCISES 4.12 (a) For each data set, find the median, midrange, and geometric mean. (b) Are they reasonable Multiple section exercises are found measures of central tendency? Explain. a. Exam scores (9 students) 42, 55, 65, 67, 68, 75, 76, 78, 94 throughout the chapter so that students can b. GPAs (8 students) 2.25, 2.55, 2.95, 3.02, 3.04, 3.37, 3.51, 3.66 c. Class absences (12 students) 0, 0, 0, 0, 0, 1, 2, 3, 3, 5, 5, 15 focus on material just learned. 4.13 (a) Write the Excel function for the 10 percent trimmed mean of a data set in cells A1:A50. (b) How many observations would be trimmed in each tail? (c) How many would be trimmed overall? 4.14 In the Excel function 5TRIMMEAN(Data,.10), how many observations would be trimmed from each end of the sorted data array named Data if (a) n 5 41, (b) n 5 66, and (c) n 5 83? 4.15 The city of Sonando Hills has 8 police officers. In January, the work-related medical expenses for each officer were 0, 0, 0, 0, 0, 0, 150, 650. (a) Calculate the mean, median, mode, midrange, and geometric mean. (b) Which measure of center would you use to budget the expected medical expenses for the whole year by all officers? doa37304_ch01_002-021.indd 2 11/21/14 4:11 PM Mini Cases Mini Case 4.2 Every chapter includes two or three mini Prices of Lipitor® cases, which are solved applications. They Prescription drug prices vary across the United States and even among pharmacies in the show and illlustrate the analytical appli- same city. A consumer research group examined prices for a 30-day supply of Lipitor® cation of specific statistical concepts at a (a cholesterol-lowering prescription drug) in three U.S. cities at various pharmacies. Atten- tion has recently been focused on prices of such drugs because recent medical research has deeper level than the examples. suggested more aggressive treatment of high cholesterol levels in patients at risk for heart disease. This poses an economic issue for government because Medicare is expected to pay some of the cost of prescription drugs. It is also an issue for Pfizer, the maker of Lipitor®, who expects a fair return on its investments in research and patents. Finally, it is an issue for consumers who seek to shop wisely. From the dot plots in Figure 4.17, we gain an impression of the variability of the data (the range of prices for each drug) as well as the center of the data (the middle or typical data values). Lipitor® prices vary from about $60 to about $91 and typically are in the $70s. The dot plots suggest that Providence tends to have higher prices, and New doa37304_ch04_106-167.indd 122 Orleans lower prices, though there is considerable variation among pharmacies. 11/21/14 5:48 PM FIGURE 4.17 Dot Plots for Lipitor® Prices Grand Rapids Providence New Orleans 60 70 80 90 viii doa37304_ch04_106-167.indd 121 11/21/14 5:48 PM ddooaa3377330044__ffmm__ii--11..iinndddd vviiiiii 1122//3300//1144 22::2255 PPMM Final PDF to printer TO PROMOTE STUDENT LEARNING? Figures and Tables Throughout the text, there are hundreds of charts, graphs, tables, and spreadsheets to illustrate statistical concepts being applied. These visuals help stimulate student interest and clarify the text explanations. FIGURE 4.19 Center versus Variability Machine A Machine B Process is correctly centered, Variability is acceptable, but but variation is excessive process is incorrectly centered 20 12 Actual Desired Actual Desired 10 15 Percent10 Percent 684 5 2 0 0 4.988 4.992 4.996 5.000 5.004 5.008 5.012 4.992 4.994 4.996 4.998 5.000 5.002 5.004 Diameter of Hole Diameter of Hole TABLE 4.7 Year Revenue JetBlue Airlines Revenue 2006 2,361 (millions of dollars) 2007 2,843 JetBlue 2008 3,392 2009 3,292 2010 3,779 Source: http://money.msn.com. Data are for December 31 of each year. Examples EXAMPLE 3.2 Figure 3.15 shows the U.S. balance of trade. The arithmetic scale shows that growth Examples of interest to has been exponential. Yet, although exports and imports are increasing in absolute U.S. Trade terms, the log graph suggests that the growth rate in both series may be slowing because students are taken from USTrade the log graph is slightly concave. On the log graph, the recently increasing trade deficit published research or real is not relatively as large. Regardless how it is displayed, the trade deficit remains a concern for policymakers, for fear that foreigners may no longer wish to purchase U.S. applications to illustrate debt instruments to finance the trade deficit. the statistics concept. For FIGURE 3.15 the most part, examples Comparison of Arithmetic and Log Scales USTrade are f ocused on business, but there are also some that U.S. Balance of Trade, 1980–2010 U.S. Balance of Trade, 1980–2010 3,000 10,000 are more general and don’t 2,500 Exports Imports Exports Imports rkseondqmooau3w7e3i0r 4lt_eechh d0a4_g1t0a e6a-n1.6r 7y.eiAn d db n 1ap3d0sr eetdrhe eoqrnue i ssatiurtee- Billions of Dollars211,,,050000000 Billions of Dollars 1,010000 11/21/14 5:48 PM dent projects. 500 0 10 1980 1985 1990 1995 2000 2005 2010 1980 1985 1990 1995 2000 2005 2010 Source: Economic Report of the President, 2011, Table B24. doa37304_ch03_056-105.indd 76 11/21/14 5:37 PM Data Set Icon A data set icon is used throughdooa3u73t04 _tchh04e_1 06t-e167x.intdd t 1o20 identify data sets used in the figures, examples, and 11/21/14 5:48 PM USTrade exercises that are included in Connect for the text. ix ddooaa3377330044__ffmm__ii--11..iinndddd iixx 1122//3300//1144 22::2255 PPMM