2
This article was downloaded by: [University of California Santa Cruz] On: 31 October 2014, At: 22:19 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK The American Statistician Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/utas20 Statistical Thinking: Improving Business Performance Robert Gould a a UCLA Published online: 01 Jan 2012. To cite this article: Robert Gould (2002) Statistical Thinking: Improving Business Performance, The American Statistician, 56:2, 157-157, DOI: 10.1198/tas.2002.s135 To link to this article: http://dx.doi.org/10.1198/tas.2002.s135 PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions

Statistical Thinking: Improving Business Performance

  • Upload
    robert

  • View
    217

  • Download
    4

Embed Size (px)

Citation preview

Page 1: Statistical Thinking: Improving Business Performance

This article was downloaded by: [University of California Santa Cruz]On: 31 October 2014, At: 22:19Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

The American StatisticianPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/utas20

Statistical Thinking: Improving Business PerformanceRobert Goulda

a UCLAPublished online: 01 Jan 2012.

To cite this article: Robert Gould (2002) Statistical Thinking: Improving Business Performance, The American Statistician,56:2, 157-157, DOI: 10.1198/tas.2002.s135

To link to this article: http://dx.doi.org/10.1198/tas.2002.s135

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose ofthe Content. Any opinions and views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be reliedupon and should be independently verified with primary sources of information. Taylor and Francis shall not beliable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilitieswhatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out ofthe use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Page 2: Statistical Thinking: Improving Business Performance

Statistical Thinking: Improving Business Performance.RogerHOERL and RonaldSNEE. Paci� c Grove,CA: Brooks/Cole ThomsonLearning, 2001, xvii + 526 pp., ISBN: 0-534-38158-8.

The business of this book is business. If you are looking for a book to teachgeneral introductory statistics, or even introductory statistics for economics stu-dents, then this is not your book. Many business statistics books, in my opinion,differ from the standard introductory texts only in that they are garnished withbusiness-related examples. Hoerl and Snee, on the other hand, offer a book com-pletely immersed in the business paradigm. So much so, in fact, that perhapsthey teach more business than statistics. The chief fault of this immersion, ifindeed it is a fault, is that the book seems more intent on convincing businesspeople to use statistics than it does on convincing statistics students to considerbusiness as a viable � eld of application.

The � rst four chapters provide an introductory overview and illustration ofbasic terms. Even though these chapters might seem lengthy, they do providedetailed and interesting case studies of real business problems, in the processvividly illustrating the application of statistical thinking. Useful techniques—statistical and otherwise—are named andexamples of their applicationare given,but no details are provided. Acronyms abound. (Is “Opportunity for Improve-ment” really worthyof an acronym?) The result is somewhat akin to lookingovera statistician’s shoulder as she walks a manager throughher analysis. The middlechapters segue from consideration of speci� c applications to more general con-cepts. Readers are introduced to “problem solving tools” (graphical summariesand organizational tools), statistical model building strategies, and experimen-tal design. The � nal chapters discuss statistical theory and concepts, althoughalways these concepts are approached from the perspective of a business ana-lyst looking to solve a problem, not a student needing to understand a generalprinciple.

The strongbusiness slant is botha strength and a weakness. Business studentswill no doubt appreciate the lack of algebra and urn-problems. They might alsoappreciate the level of detail put into the case studies, which makes for trulyinteresting reading for the motivated student. The weakness of this approachis that those wishing to teach statistics as a cohesive and uni� ed discipline—as opposed to a collection of tools—will be disappointed. Statistical tools andtests (and other problem solving tools) are treated in catalog fashion. Chapter 5presents each tool with brief sections titled “purpose,” “bene� ts,” “limitations,”“examples,” “procedure,” “variations,”and “tips.”After seven pages of sketchingthe hypothesis-testingparadigm in Chapter 8, tests are presented in catalog form,broken down into two-paragraph chunks; the � rst a description (“This test is: : : used to compare a sample average to a hypothesized value.” p. 355) and theseconda few sentences on “theoretical assumptions.”Formulas are not given, norare they described. Instead, readers are told that “statistical software packagescalculate p-values based on the relevant statistical distributions” (p. 355).

A conference held at theUniversity ofChicago’s Graduate SchoolofBusiness,“Making Statistics More Effective in Schools ofBusiness” (Easton, Roberts, andTiao 1988),agreed with reform advocates (Roberts 1987)and recommended thatbusiness statistics courses show examples of real applications, reduce the em-phasis on formal theory and formal testing, and increase attention paid to topicssuch as time series, quality and productivity, sampling, and report writing. Inaddition, they reported a “resounding” recommendation: a need for statisticalcase studies. To their credit, Hoerl and Snee’s book takes these recommenda-tions very seriously. But I fear the de-emphasis on theory has gone too far. Forexample, nowhere is there a discussion of statistical independence. Without anintroduction to this most fundamental concept, how well can managers com-municate with their statisticians? Even worse, how much damage could theydo in an analysis without understanding that they have violated a fundamen-tal assumption? There are other places where a statistically na� õ ve reader mightbe led astray. For example, the discussion on hypothesis tests says, “In manycases : : : . we do not decide to formally test a hypothesis until after the dataare collected. In such cases, analysis of the data is what led us to consider thishypothesis”(p. 352). True, but after that, I hoped for a warning about � shing ex-peditions or post-hoc tests. Instead readers are merely warned that in such casesthe quality of the data is an important consideration. There are other places inwhich the book strays from statistical convention. For example, the average isvaguely de� ned as “the central value around which the process varies.” (p. 12).A subsequent example illustrates how to calculate the sample average, but thisde� nition, besides begging the question of what is meant by “central,” makes itdif� cult for the student to place the sample mean in an inferential context.

This book makes heavy use of software, which I see as a strength. Shortintroductions are provided to Minitab, JMP, and Excel. I willingly concede that

Excel is in the business community to stay, but I don’t like it. I would like tothink that my job as a statistics educator is to convince students that they willbe more successful if they do not use Excel. The authors use Excel for mostof their examples of “basic” statistics, but steer students away from Excel andrecommend Minitab or JMP for the more formal analyses. I wish they were alittle more emphatic; rather than present Excel as one of three possible choices,I would like to see it pointed out that Excel is not a statistical analysis packageand should not be used as one.

Whether or not you think this is a goodbookwill depend on who your studentsare and what you’d like them to get out of yourstatistics course. Studentswho arealready working as managers or who aspire to managerial positions will comeaway with an appreciation of the usefulness of statistics. But I fear that studentswho wish to function as statisticians in a business setting will be short-changed.

Robert GOULDUCLA

REFERENCES

Easton, G., Roberts, H. V., and Tiao, G. C. (1988),“Conference Report,”Journalof Business and Economic Statistics, 6, 247–260.

Roberts, H. V. (1987),“Data Analysis for Managers,” TheAmerican Statistician,41, 270–278.

Teaching Statistics, Resources for Undergraduate Instructors.Thomas J. MOORE (ed.). Washington, DC: The Mathematical Associationof America and the American Statistical Association, 2001, xii + 222 pp.,$31.95(P), ISBN: 0-88385-162-8.

As I started to read Teaching Statistics I jotted down two questions: Howwill this collection of essays change the way I teach my undergraduate statisticscourses? How will the essays in Teaching Statistics change the way nonstatisti-cians teach undergraduate statistics? I am pleased to report that the answers toboth questions are “for the better!”

Teaching Statistics is published jointly by the MAA and the ASA within theMAA Notes series. MAA Notes volumes are not intended to be used as primarytexts, but they support undergraduate instruction in mathematics and, in thiscase, statistics.

The essays and commentary collected and edited by Tom Moore in Teach-ing Statistics are organized into six sections: Hortatory Imperatives, Teachingwith Data, Established Projects in Active Learning, Textbooks,Technology, andAssessment.

Moore’s preface to the volume and the Section 1 (reprinted) essay titled“Teaching Statistics: More Data, Less Lecturing” by George Cobb are worthreading, but they don’t carry the speci� c practical advice present in the middlesections of the volume. Moore’s goal for the volume is explicitly stated in thepreface: Teaching Statistics “aims to be an ‘instructor’s manual’ for statisticseducational reform for teachers of statistics at the undergraduate and secondaryschool levels.” Sections 2 through5 of the volumeare � lled with practical advicethat really does read like a user-friendly instructor’s manual.

Notice that Moore’s objective does not distinguish between nonstatisticiansand statisticianswho teach statistics. In fact, nonstatisticianswho � nd themselvesteaching statistics are the primary audience for this volume.

Section 2 offers guidance and suggestions for incorporating real datasets,projects, and case studies into statistics courses, and Section 4 provides essayscontaining advice on the selection of textbooksfor introductorystatistics coursesand also for mathematical statistics courses. The tone of the essays in Teach-ing Statistics is not dogmatic but is supportive, and it appears that great efforthas been made to avoid singling out speci� c products and textbooks as beingexceptionally good or bad.

What is remarkable about Teaching Statistics is the balance between overviewand philosophyversus advice on practical implementation. In Section 3 “Estab-lishedProjects in Active Learning”and Section 5“Technology,”overviewessaysor excerpts from published materials are followed by commentary written by un-dergraduate statistics instructors who are not the authors of the primary material.In fact, many of the commentators are not statisticians by training. For example,Michael Seyfried, a geometer from Shippensburg University of Pennsylvania,wrote the companion piece for Allan Rossman’s “Excerpts from ‘WorkshopStatistics: Discovery with Data’.” The essay “Using Graphing Calculators forData Analysis in Teaching” was contributed to the “Technology” section by PatHopfensperger, a high school teacher from Wisconsin.

The American Statistician, May 2002, Vol. 56, No. 2 157

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 22:

19 3

1 O

ctob

er 2

014