Quality ManagementJournal

Does Six Sigma Improve Performance?S. Thomas Foster

The Long-Term Impact of ISO 9000 Certification on Business Performance:A Longitudinal Study Using Turkish Stock Market Returns

Musa Pinar and Ceyhun Ozgur

Applying the Toyota Production System to a Healthcare Organization:A Case Study on a Rural Community Healthcare Provider

Kevin F. Collins and Senthil K. Muthusamy

Ranking Customer’s Requirements in QFD by Factoring inTheir Interrelationship Values

J. R. Sharma and A. M. Rawani

V O L U M E 1 4 , I S S U E 4

144

QUALITY M

ANAG

EMEN

T JOURN

AL2007

www.asq.org

Cover spine v14 i4 9/20/07 10:12 AM Page 1

ContentsVolume 14 Issue 4

AIMS AND SCOPES: The Quality Management Journal (QMJ ) is a rigorously refereed periodicalpublished quarterly by the American Society for Quality (ASQ). It is the first journal to link the efforts of acad-emic researchers and quality management practitioners. QMJ provides a forum for communicating and discussingresearch findings. New research that scientifically explores the principles of quality management is activelysought for the journal.

All papers must be original works and be grounded in research and experience. Each article mustprovide readers with significant new information that contributes to the development of the field of qualitymanagement. Highly technical papers of limited scope are not appropriate for this publication. Manuscriptssubmitted to QMJ must be original works not previously published or under review by another publication.

Note that ASQ and QMJ cannot be responsible for unsolicited manuscripts.SUBSCRIPTIONS: For subscriptions to QUALITY MANAGEMENT JOURNAL, please contact ASQ’s Customer Care

Center at 800-248-1946 or 414-272-8575 (outside North America) or visit the Web site at http://www.asq.org.IDENTIFICATION STATEMENT: Copyright 2007 by ASQ. QUALITY MANAGEMENT JOURNAL (ISSN 1068-6967) is published quarterly (January, April, July, October) by

ASQ, 600 N. Plankinton Ave., P.O. Box 3005, Milwaukee, WI 53201-3005. Preferred periodicals postage paid at Milwaukee, WI, and additional mailing offices. CanadianGST #128717618. Canadian Publications Mail Agreement #40030175, Returns: 4960-2 Walker Rd., Windsor, ON N9A 6J3. Back issues are available by calling ASQ at 800-248-1946 (U.S. or Canada) or 414-272-8575. ASQ members $15 per copy; nonmembers $23 per copy and based on availability.

Authorization to photocopy items for internal or personal use is granted by QUALITY MANAGEMENT JOURNAL provided that the fee of $1 per copy is paid to theCopyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. Copying for purposes other than internal or personal reference requires theexpress permission of QUALITY MANAGEMENT JOURNAL. For permission, contact Valerie Funk at ASQ, P.O. Box 3005, Milwaukee, WI 53201-3005, or call 414-272-8575.To request bulk reprints (100 or more) contact Barbara Mitrovic at the same address or phone number.

For microform, contact ProQuest Information and Learning, 300 N. Zeeb Road, Ann Arbor, MI 48106, 800-521-0600, ext. 2888, international 734-761-4700,http://www.il.proquest.com.

MANUSCRIPT REQUIREMENTS: Manuscripts should be submitted to James R. Evans, Editor, Quality Management Journal c/o ASQ, 600 N. Plankinton Ave., Milwaukee, WI53203. See the Author Guidelines in the back of this issue for details about manuscript requirements and submission or see ASQ’s Web site at http://www.asq.org.

CLAIMS: No claim for missing issues will be accepted after three months following the month of publication of the issue for domestic addresses and six months for Canadianand international addresses.

POSTMASTER: Please send address changes to Quality Management Journal, ASQ, P.O. Box 3005, Milwaukee, WI 53201-3005. Printed in USA

Annual Subscription RatesMembers Nonmembers Institutional

U.S. $50.00 $75.00 $130.00

International $74.00 $90.00 $150.00

Canadian $80.00 $95.00 $160.00

Rates subject to change without prior notification. Institutional rates includefirst-class delivery. Canadian rates include 7% GST.Institutional subscriptions are held in the name of a company, corporation,government agency, or library.

ARTICLES

7 Does Six Sigma Improve Performance?S. Thomas Foster

A look at the long-term financial and operational results of companiesimplementing Six Sigma.

21 The Long-Term Impact of ISO 9000 Certification on BusinessPerformance: A Longitudinal Study Using Turkish Stock Market ReturnsMusa Pinar and Ceyhun Ozgur

Do ISO 9000 firms have higher returns and lower variability than non-ISO 9000 firms?

41 Applying the Toyota Production System to a Healthcare Organization:A Case Study on a Rural Community Healthcare ProviderKevin F. Collins and Senthil K. Muthusamy

The TPS can be a powerful technique—even to companies unaccustomedto advanced production techniques.

53 Ranking Customer’s Requirements in QFD by Factoring in TheirInterrelationship ValuesJ. R. Sharma and A. M. Rawani

The most neglected part of the QFD process—the interrelationship matrixamong customer requirements—is examined.

DEPARTMENTS

4 Introduction

5 Executive Briefs

61 Book Reviews

The annual index for Volume 14can be found online:http://www.asq.org/pub/qmj/.

Contents v14 i4 9/20/07 10:11 AM Page 1

EDITORJames R. Evans

University of Cincinnati

FOUNDING EDITORWilliam A. Golomski

BOOK REVIEW EDITORJames B. Kohnen

St. Mary’s College of California

PUBLISHERWilliam Tony

MANUSCRIPT COORDINATORValerie Funk

COPY EDITORSLeigh Ann KlausKris McEachern

JoAnn Regenauer

PRODUCTIONCathy Schnackenberg

DIGITAL PRODUCTION SPECIALISTLaura Franceschi

QMJ ADMINISTRATION/PRODUCTION

QMJ EDITORIAL

Masthead v14 i4 9/20/07 10:13 AM Page 2

Editorial EditorialAdvisory Board Review Board

Stefano Biazzo*University of Padova

Ching-Wen Chen*National Kaohsiung First University of Scienceand Technology

Michael B.C. Khoo*Universiti Sains Malaysia

Adrian S. ChooRensselaer Polytechnic Institute

Edward DuplagaWinona State University

Susan West EngelkemeyerBabson College

Byron FinchMiami University

Mark P. Finster University of Wisconsin

Thomas S. Foster Jr.Boise State University

Brian Fynes*The Michael Smurfit Graduate School of Business

Thomas GattikerBoise State University

Soumen GhoshGeorgia Institute of Technology

Glenn H. GilbreathVirginia Commonwealth University

John M. Groocock*TRW (Retired)

Robert HandfieldNorth Carolina State University

Sandra J. HartmanUniversity of New Orleans

David HartmannUniversity of Central Oklahoma

Marilyn HelmsDalton State College

Mary Collins Holcomb University of Tennessee

Hale KaynakUniversity of Texas-Pan American

Gary Kern Indiana University South Bend

David Kerridge*Aberdeen University

Ray A. Klotz Qualcomm Inc.

Frank KnightFISI Madison Financial

Ronald D. KurtzmannDiamond Management Systems

Yongjin (James) KwonDrexel University

Keong LeongThe University of Nevada–Las Vegas

A. Magid MazenSuffolk University

Kathleen L. McFaddenNorthern Illinois University

Kate McKone-SweetBabson University

Satish MehraUniversity of Memphis

Kim I. MeltonNorth Georgia College and State University

Henry R. Neave*British Deming Association

Yoram NeumannCalifornia State University

William Newman Miami University

Gertrude P. PannirselvamSouthern Illinois University Edwardsville

Darrell RadsonDrexel University

Gary Ragatz Michigan State University

Gipsie B. RanneyBelmont University

Borut Rusjan*University of Ljubljana

Brooke Saladin Wake Forest University

Joseph SarkisClark University

Helmut Schneider Louisiana State University

Nirmal Sethia California State Polytechnic University

Rui Sousa*Universidade Catolica Portuguesa

John G. Surak Clemson University

Mile Terziovski*The University of Melbourne

Michael D. Tveite Polaris Industries

Peter Ward The Ohio State University

L. David Weller University of Georgia

*International reviewer

John AndersonUniversity of Minnesota

Selwyn BeckerUniversity of Chicago

Robert E. ColeUniversity of California

James W. Dean Jr.University of North

Carolina

James R. EvansUniversity of Cincinnati

John P. EvansUniversity of North

Carolina

Frank M. GrynaUniversity of Tampa

John HamburgAPEX, Inc.

David LutherLuther Quality Associates

Ram NarasimhanMichigan State University

Duke OkesQuality Management

Division, ASQ

Roger G. SchroederUniversity of Minnesota

Kalyan SinghalUniversity of Baltimore

Michael J. StahlUniversity of Tennessee

Editorial Board v14 i4 9/20/07 10:12 AM Page 3

I am extremely honored to have been selected as the new editor of the Quality Management Journal.I would like to thank the ASQ Publications Management Board for putting their trust in me, and the out-going editor, Barbara Flynn, for her help during the transition process and the preparation of this issue.Barb’s stewardship of the journal will certainly be missed, but she has joined its Editorial AdvisoryBoard to continue to provide counsel and advice. I wish her well in her work at Indiana University.

For those of you who may not know me, my formal educational background is in industrial engineer-ing, and I have been on the faculty in the College of Business at the University of Cincinnati for more than32 years. My involvement in quality began some 20 years ago with the preparation of the first edition ofa textbook entitled The Management and Control of Quality, with my coauthor William Lindsay. Wewrote this book to fill a need for a comprehensive text that focused more on managerial issues taught inbusiness schools, rather than just technical topics such as statistical process control and reliability. Now inits 7th edition with Thomson/South-Western, and retitled Managing for Quality and PerformanceExcellence, this book has helped me maintain currency in the field. I continue to actively write both booksand research papers on various quality topics. In addition, I have engaged in a variety of professionalactivities in the field, including serving on the Board of Examiners and the Panel of Judges for theMalcolm Baldrige National Quality Award for a total of 11 years since 1994.

The frequent citations of quality issues in the automotive industry, food safety, and in developing coun-tries that we so often read in the press point to the continued need to better understand both the theoreticalfoundations of quality management and how to translate these into better management practice. This canbe facilitated through good applied research, reflecting upon W. Edwards Deming’s philosophy that thereis no knowledge without theory. That’s what this journal is all about—providing relevant knowledge aboutquality management practice that is grounded in rigorous research. In the next issue, I will present a moredetailed editorial philosophy that will guide my tenure as editor, a new Editorial Review Board andEditorial Advisory Board, and some key changes to the Guidelines for Authors, all of which are gearedtoward better serving the readership of this journal and the professional quality community.

This issue begins with two empirical research papers that add to our understanding of the impact onquality-related initiatives. Tom Foster’s article, “Does Six Sigma Improve Performance?” examines theeffects of Six Sigma programs on long-term financial and operating results in a similar fashion by whichothers have examined the effects of more general TQM initiatives. Musa Pinar and Ceyhun Ozgur’s article,“The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study UsingTurkish Stock Market Returns,” takes a similar focus, comparing long-term stock returns and variabilityof returns of ISO 9000 certified versus noncertified firms in the Turkish stock market. Both of these articlesprovide more rigorous evaluation of quality initiatives beyond the anecdotal knowledge commonly seen inother publications. In “Applying the Toyota Production System to a Healthcare Organization: A Case Studyon a Rural Community Healthcare Provider,” Kevin F. Collins and Senthil K. Muthusamy present an inter-esting case study of how healthcare can apply the principles of the Toyota Production System, focusingon the challenges and techniques for effective implementation. This case study not only provides lessonsfor practitioners, but also opportunities and ideas for further empirical research. Finally, J. R. Sharmaand A. M. Rawani propose a new approach to enhancing quality function deployment by consideringthe interrelationships of customer requirements in their article, “Ranking Customer’s Requirements inQFD by Factoring in Their Interrelationship Values.”

I look forward to your support of the journal. Please feel free to contact me with any questions orsuggestions.

James R. EvansEditor

4 QMJ VOL. 14, NO. 4/© 2007, ASQ

Introduction

Introduction v14 i4 9/20/07 10:13 AM Page 4

BriefsExecutive

Does Six Sigma ImprovePerformance (pp. 7 – 20).S. Thomas Foster, Brigham YoungUniversity

What are the long-term financialand operational results from theimplementation of Six Sigma? Thisarticle seeks to answer this question.In recent years, many firms haveadopted Six Sigma in an effort toimprove quality and reduce costs.Whether they actually succeeded indoing so, however, is not clear. Thereis a need for more research eluci-dating the benefits and costs of SixSigma implementation.

The authors present two propo-sitions:

• Proposition 1: Six Sigma adop-tion will positively affect financialresults. This proposition is thentranslated into five hypothesesfor this study.

• Proposition 2: Six Sigma adop-tion results in better operationalresults. This proposition is alsotranslated into five hypotheses.

To identify a group of companiesto study, the authors performed akeyword Lexis/Nexis online searchof corporate financial reports for theterms Six Sigma, ISO 9000, TQM,and Baldrige. They then performeda content analysis to determinewhich firms had announced theimplementation of Six Sigma. Theirsearch eventually resulted in 41companies to study.

The authors studied the follow-ing outcome variables: free cashflow per share; cost per dollar sales;earnings before interest, taxes,depreciation, and amortization(EBITDA); sales; sales per employee;asset turnover; return on assets;return on investment; total assets;and number of employees.

The results of the study weresomewhat mixed. It seems that SixSigma has a significant effect onfree cash flow, EBITDA, and assetturnover. It did not, however, seemto affect sales return on assets,return on investment, for firmgrowth.

The Long-Term Impact ofISO 9000 Certification onBusiness Performance: A Lon-gitudinal Study Using TurkishStock Market Returns (pp.21– 40). Musa Pinar and CeyhunOzgur, Valparaiso University

Despite the expected benefits ofISO 9000 certification, only a fewempirical studies have examinedwhether ISO 9000 certified compa-nies actually perform better thannon-ISO 9000 certified companies.This article looks to identify the lon-gitudinal effect of ISO 9000 certifi-cation on business performance bycomparing the stock performancesof ISO 9000 certified firms vs. non-ISO 9000 certified firms traded onTurkey’s Instabul Stock Exchange.

It is thought that the impact of ISO9000 success should be reflected intheir stock prices and returns.Specifically, this article seeks to: 1) provide information about thecharacteristics of ISO 9000 andnon-ISO 9000 firms; 2) analyzethe impact of ISO 9000 certifica-tion on stock returns and stockreturn variability; and 3) deter-mine if the stock returns and thevariability of returns of ISO 9000firms are influenced by certifyingorganization.

For their study, the authors usedthe monthly stock returns of allmanufacturing firms listed on theISE market of Turkey from January1997 to September 2005. Then, tobe included in the analysis, the ISO9000 certified firms had to: 1) havereceived their certification beforeJanuary 1997; and 2) have been list-ed on the ISE during the entirestudy period. For non-ISO 9000firms, the requirement was thatthey were traded on the ISE for theentire study period. The studyexamined the yearly returns; vari-ous combinations of two-, three-,and four-year returns; and returnsover the nine-year period.

Using annual two-year, three-year, and four-year averages of themonthly stock returns, the resultsindicated that ISO 9000 certifiedfirms generally have higher returnsand lower variability than non-ISO

www.asq.org 5

Exec Briefs v14 i4 9/20/07 10:12 AM Page 5

Briefs

9000 firms. Also, results showedthat ISO 9000 certified firms con-sistently had higher means andlower variances. Finally, the certi-fying agency had little effect on thestock performance and variabilityof returns.

Applying the Toyota Produc-tion System to a HealthcareOrganization: A Case Study ona Rural Community HealthcareProvider (pp. 41–52). Kevin F.Collins, Healthcare Consultant, andSenthil K. Muthusamy, BowlingGreen State University

Over the last five decades theToyota Production System (TPS)has evolved from an advancedsociotechnical concept in manufac-turing to a participative design forlarge-scale change management.Toyota has been able to sustain astrategic competitive advantage byapplying TPS as a process innova-tion and intervention, as measuredby quality, reliability, productivity,cost reduction, sales and marketgrowth, and market capitalization.Many organizations are trying toreplicate Toyota’s success with TPSin their respective business andindustry environments.

Community Health Systems(CHS) is a network of rural hospitalsthat is trying to improve perfor-mance using advanced managementtechniques. This article looks atCHS’s application of TPS as a process

innovation/intervention technique toimprove quality, safety, and costreduction, and survive accreditationaudits.

To achieve a competitive advan-tage using TPS requires adherence tothe fundamental technical rules,developing suitable organizational/structural design, and fostering cul-tural norms and values that facilitatethe TPS processes. The TPS methodis very scientific in that it clearlydelineates the responsibilities at theoperational level to a very highdegree of specification, and it doesnot promote vertical command andcontrols. TPS facilitates creativebehaviors and also emphasizes cer-tain leadership qualities in managersimplementing TPS.

Using TPS, CHS devised a busi-ness model that provides guidance toits small rural community providersin the area of quality and patient out-come management, specificallyrequiring standardization of accred-itation documentation. This is amodel that, when enforced, can pro-duce positive outcomes and result inbuilding competitive advantages.

Ranking Customer Require-ments in QFD by Factoring in Their InterrelationshipValues (pp. 53–60). J. R. Sharma,Manoharbhai Patel Institute ofEngineering, and A. M. Rawani,National Institute of Technology,Raipur

Quality function deployment(QFD) is a methodology used toachieve higher customer satisfac-tion. The engineering characteristicsaffecting product performance aredesigned to match the customerrequirements. The study discussedin this article deals with the mostneglected and often-omitted partof the QFD process—the interrela-tionship matrix among customerrequirements.

This article presents a mathe-matical solution to the correlationtriangle problem that incorporatesboth customer requirements andengineering characteristics. This newapproach addresses relationships ina different and improved numericalmethod.

This article shows the effective-ness and preciseness of the proposedmodel. Numbers are used to repre-sent the imprecise nature of thejudgments and to define moreappropriately the relationshipsamong customer requirements.The model is shown to be compu-tationally feasible for realisticproblems and outperforms tradi-tional approaches on the basis ofbeing relatively straightforwardand simplified. The proposedapproach is shown to producemore informative results, addingcredibility to the outcome and itsanalysis, for the benefit of theproduct designers and developers.

6 QMJ VOL. 14, NO. 4/© 2007, ASQ

Exec Briefs v14 i4 9/20/07 10:12 AM Page 6

www.asq.org 7

Does Six SigmaImprove Performance?

S. THOMAS FOSTER JR. BRIGHAM YOUNG UNIVERSITY

© 2007, ASQ

This article presents the long-term financial and oper-ational results coming from the implementation of SixSigma. The results from Six Sigma programs weresomewhat mixed. The author found a significant effecton free cash flow (FCF), earnings before interest,taxes, depreciation, and amortization (EBITDA), andasset turnover. Six Sigma did not seem to affect salesreturn on assets, return on investment, or firmgrowth. As a result, if firms want to improve cash,earnings, or productivity in using assets, Six Sigmamight be of use. In 1998, companies with high cashflows and no quality management program (QMP)had lower FCF in 2002 than companies that hadadopted Six Sigma. Companies with low cash flow andno QMP did better than companies using Six Sigma,suggesting that for cash-poor firms, Six Sigma may bea drain on resources. Also, these companies may nothave the cash necessary to sustain effective Six Sigmaresults over four years. Among companies with lowand medium asset turnover, Six Sigma led to higherasset turnover. It could be that companies with lowasset turnover could benefi t more from processimprovement implicit in Six Sigma than firms withhigh asset turnover.

Key words: financial and operational outcomes, qualitymanagement, Six Sigma

INTRODUCTIONIn recent years, many firms have adopted Six Sigma inan effort to improve quality and reduce costs (Pyzdek2003). Six Sigma has been attractive to business exec-utives, as it is thought to overcome some of the pitfallsof historical quality management implementations(Linderman et al. 2003). Quality management pro-grams have been criticized for relying on improvementwithout mechanisms for ensuring that positive resultswill ensue (Howard, Foster, and Shannon 2005). Thatis, if employees are trained and empowered in qualityimprovement approaches, quality improvements andbenefits will “percolate to the top,” thereby improvingcompany performance.

On the other hand, the Six Sigma approachrequires more direction and leadership from topmanagement than traditional quality management.This is termed “leadership for Six Sigma” (Foster2007; Treichler et al. 2002). Along with this higherdegree of leadership is a more structured process forimproving performance. One approach to Six Sigmais the five-phase define, measure, analyze, improve,and control (DMAIC) process. Included with DMAICis a method for leaders to priorit ize potentialimprovement projects based on the probability thatsuch projects will result in financial benefit to theorganization.

Similarly, another defining aspect of Six Sigma isits greater emphasis on cost reduction through qualityimprovement. This aspect includes a target of at least$200,000 in cost reduction for each Six Sigma projecttaken on with two to three projects per Black Belt overa one- to two-year period (Bisgaard and DeMast2006). Such returns are the results expected from thehigh cost of investing in Black Belt employees.

Foster v14 i4 9/20/07 10:10 AM Page 7

8 QMJ VOL. 14, NO. 4/© 2007, ASQ

Does Six Sigma Improve Performance?

The convergence of these dimensions of leadership,structured process for improvement, and focus onfinancial results present in Six Sigma was designed toaddress the perceived weaknesses of traditional qualitymanagement. It is expected that this increased focuson financial outcomes should result in improvementsin financial and operational results. The purpose ofthis research is to determine if this is indeed the case.

LITERATURE REVIEW ANDHYPOTHESIS DEVELOPMENTSeveral studies have examined the impact of qualitypractices on financial and operational results(Kaynak 2003). Most of these studies compare theresults of firms that adopt quality practices versuscontrol groups to determine if quality practices signif-icantly influence financial and operational results.The purpose of this literature review is threefold. Theauthor defines Six Sigma. Then he identifies similarstudies that have examined the financial and opera-tional impacts of different quality improvementapproaches such as total quality management (TQM).Finally, he uses this review to provide a basis forselecting variables to be studied in this research.

Six Sigma The author’s literature review revealed only one articlerelating to Six Sigma in an A-level journal (Lindermanet al. 2003). Linderman et al. studied Six Sigma froma goal-theoretic perspective. They examined the rela-tionship between goals and Six Sigma success anddeveloped a series of propositions suggesting that high,yet attainable, goals were important to the success ofSix Sigma programs. They also examined the inter-vening effects of effort, persistence, and directionresulting from explicit Six Sigma goals. Given the wideadoption of Six Sigma in many organizations aroundthe world, there is a need for more research elucidat-ing the benefits and costs of Six Sigma implementa-tion. Currently, no studies investigate the relationshipbetween Six Sigma implementation and financial andoperational results.

The most common process for Six Sigma is DMAIC.This is similar to the plan, do, check, act (PDCA) cycleproposed by Walter Shewhart and W. Edwards Deming(Deming 1986). Since results from Six Sigma areclosely related to the DMAIC process, the author dis-cusses it in detail. It should be noted that as recentlyas the 1990s, many companies used DMAIC to guideSix Sigma projects.

In the define phase, projects are identified andselected. Project selection is performed under thedirection and with the participation of a Six Sigmachampion. Also involved are Master Black Belts andBlack Belts or Green Belts. Project selection is per-formed in four steps: 1) developing the business case;2) project evaluation; 3) Pareto analysis; and 4) projectdefinition. Business case development involves identify-ing a group of possible projects, writing the businesscase, and stratifying the business case into problem andobjective statements. Project evaluation often involvesrisk and return assessment. The key individuals in per-forming this analysis are project champions. Projectchampions are usually top management executiveswho have legitimate and financial authority to supportSix Sigma projects (Treichler et al. 2002). The processfor defining Six Sigma projects helps in prioritizingwhich projects will provide the greatest financial andoperational returns.

The measure phase involves two major steps: 1) selecting process outcomes; and 2) verifying meas-urements. To select process outcomes, process mappingis used to help understand and define the processitself. A process map is a flowchart showing responsi-bilities (Gourishankar 2003). The goal of a processmap is to identify nonvalue-added activities. Twoimportant measures that are monitored are defects perunit (DPU) and defects per million opportunities(DPMO). Measurement systems analysis (MSA) isused to determine if measurements are consistent(Conklin 2006).

The analyze phase involves gathering and ana-lyzing data relative to a particular Black Belt project(Pyzdek 2003). The analyze-phase steps are as follows:1) define the performance objectives; 2) identify inde-pendent variables (Xs); and 3) analyze sources ofvariability. Defining objectives involves determining

Foster v14 i4 9/20/07 10:10 AM Page 8

Does Six Sigma Improve Performance?

what characteristics of the process need to be changed toachieve improvement. Next, independent variables areidentified for gathering data. These are variables thatsignificantly contribute to process or product variation.The goal of analyzing sources of variability is to usevisual and statistical tools to better understand the rela-tionships between dependent (X) and independent (Y)variables for use in future experimentation. A varietyof tools are used in this analysis, including his-tograms, box plots, scatter plots, regression analysis,and hypothesis tests.

The improve phase of the DMAIC process involvesoff-line experimentation (Antony and Esamilla 2003;Montgomery 2004). Off-line experimentation involvesstudying the identified variables and using design ofexperiments (DOE) to determine whether the inde-pendent variables significantly affect variation independent variables.

The control phase involves putting into placeprocess checks to ensure that improvements are longlasting. The DMAIC process provides a process forimproving operational results that is cost-reductionoriented. This results in the removal of waste fromprocesses. These improvements are expected to result inoperational and financial improvements. In thisresearch, the author examines whether this is the case.

Financial ResultsPrevious studies on the financial and operationalimpacts of quality efforts have provided mixed results.The Jacobsen and Aaker (1987) study using the ProfitImpact of Market Strategy (PIMS) database examinedthe relationship between quality and market share,return on investment (ROI), relative price, relativecost, and relative quality. Jacobsen and Aaker (1987)found the impacts of quality improvement practices tobe positive and significant. Depending on how busi-nesses were grouped, however, they found variance inresults that might be reflective of business strategies.Phillips, Chang, and Buzzell (1983) found similarrelationships and added that product quality appearedto be negatively related to cost.

Several researchers have investigated the impactof quality practices on stock prices. Easton and Jarrell

(1998) reviewed the financial results of 108 firms thatstarted TQM programs between 1981 and 1991. Theyfound that these firms’ stocks outperformed a controlgroup by finding excess cumulative returns. Hendricksand Singhal (1996; 1997; 2001a; 2001b) found similarresults by analyzing the market returns of qualityaward winners. Hendricks and Singhal used winners ofvarious awards to determine whether implementingeffective TQM resulted in improved operating perform-ance in firms. They used the winning of a qualityaward as a proxy for effective implementation of TQM.In their research, they found strong evidence thatwinners of quality awards outperformed control firmson operating-income-based measures. Subsequent totheir studies, other researchers have continued tostudy receipt of awards such as the Malcolm BaldrigeNational Quality Award as indicative of mature qualityimplementation (Rajan and Tamimi 1999; Wilson andCollier 2000; Dean and Tomovic 2004). The literature,however, is not unanimous. York and Miree (2004)studied the links between TQM and financial perform-ance. Conducting a comparison of the financial per-formance of quality award winners against theircontrol firms by SIC groups, they studied performanceboth before and after the winning of the BaldrigeAward. York and Miree found that TQM firms had betterfinancial performance before and after they won theawards—some for even 20 years prior to winning theawards. The author suggested that winning the awardwas a covariate for financial success.

Adams, McQueen, and Seawright (1999) studiedthe stock performance of Baldrige Award winnersfrom the day their award was announced. They foundonly limited evidence to support the hypothesis thatstockholders are rewarded with abnormal returns onthe day of the quality award announcement. Theysuggested that stock analysts may have been fore-warned that the company was winning or that theywere at least aware of prior TQM efforts. They alsosuggested that there may be little impact on stockprice, as stock impact is not the purpose of theBaldrige Award. They do propose that stock impactsrelating to announcements are not as important asimprovements relating to quality improvementefforts. While many studies have focused on the

www.asq.org 9

Foster v14 i4 9/20/07 10:10 AM Page 9

10 QMJ VOL. 14, NO. 4/© 2007, ASQ

Does Six Sigma Improve Performance?

financial and operational results of TQM and theBaldrige Award (Pannirselvam, Siferd, and Ruch1998), the author’s literature review revealed no suchstudies relative to Six Sigma.

It is expected that Six Sigma should result inimproved financial performance. Since Six Sigmaemphasizes reductions in cost and increases in meas-ures such as rolled-throughput, one should find thatcash flows, earnings, and other financial measures willshow improvement. General Electric (GE) has reportedmore than $12 billion in savings due to Six Sigma.Similarly, Motorola reported $15 billion in savings over11 years from Six Sigma implementation. As processesand products are improved, it is also expected that saleswill improve. This research was performed to see if theexpected financial results occurred for companies thatimplemented Six Sigma overall.

As the author has discussed, when compared withtraditional quality management, Six Sigma is muchmore cost-reduction and financial-results oriented. Acomprehensive review of the quality management lit-erature reveals that financial and operational perform-ance could generally be categorized as measuringprofitability, cost, efficiency, and growth (Phillips,Chang, and Buzzell 1983; Jacobsen and Aaker 1987;Ward, Leong, and Boyer 1994; Flynn, Schroeder, andSakakibara 1995; Mohrman et al. 1995; Hendricks andSinghal 1996; 1997; Easton and Jarrell 1998; Samsonand Terziovski 1999; Curkovic, Vickery, and Droge2000; Lapre, Mukherjee, and VanWassenhove 2000;Wilson and Collier 2000; Cua, McKone, and Schroeder2001; Devaraj, Matta, and Conlon 2001; Douglas andJudge 2001; Fynes and Voss 2001; Hendricks andSinghal 2001a; 2001b; Park, Hartley, and Wilson 2001;Sousa and Voss 2001; Eriksson and Hansson 2003;Fullerton, McWatters, and Fawson 2003; Kaynak2003). For this study, the author used a modified ver-sion of Hendricks and Singhal’s (1997) performancemeasures. Financial measures used in this studyinclude free cash flow per share; cost per dollar share;earnings before interest, taxes, depreciation, andamortization (EBITDA); sales; and sales per employee.The author starts with the following proposition:

Proposition 1: Six Sigma adoption will positivelyaffect financial results.

Next, this proposition is translated into five hypothe-ses for this study. The author posits that investment inSix Sigma training and implementation signals afocus on aggressive cost reduction coupled withprocess and organizational improvement. Given thatcosts are reduced, cash is freed up for other uses. Thisresults in hypotheses H1a and H1b:

• H1a: Six Sigma adoption is positively associated withfree cash flow per share.

• H1b: Six Sigma adoption is positively associated withcost per dollar sales.

As a result of this signaling of a focus on processand organizational improvement, operating marginsshould improve. This argument is similar to Deming’svalue chain—that focusing on quality will result inlower costs and improved performance. Since operatingmargin can be expressed as EBITDA/sales, the authorproposes the following three hypotheses:

• H1c: Six Sigma adoption is positively associatedwith EBITDA.

• H1d: Six Sigma adoption is positively associatedwith sales.

• H1e: Six Sigma adoption is positively associatedwith sales per employee.

Operational ResultsIn addition to looking at financial results, many stud-ies have examined the relationship between qualityimplementation and operational results. Again, theauthor reviews these studies to understand impactsfound by traditional quality improvement approaches.Dow, Samson, and Ford (1999) studied the effect ofquality practices on quality outcomes. They categorizedquality practices into nine dimensions. Not all of theirdimensions, however, contributed to superior qualityperformance. Employee commitment, shared vision,and customer focus yielded positive correlations withquality outcomes. Conversely, other hard qualitypractices, such as benchmarking, work teams,advanced manufacturing technologies, and closesupplier relations, were not related to superior qualityoperational results.

Foster v14 i4 9/20/07 10:10 AM Page 10

Does Six Sigma Improve Performance?

Focusing on the competitive aspects of quality per-formance, Douglas and Judge (2001) found strongempirical support for a positive relationship betweenthe degree of TQM implementation and organizationalperformance. They also found some empirical evidencethat this relationship between TQM implementationand organizational performance was moderated byorganizational structure.

Das et al. (2000) used structural equation modelingto study the impacts of international competition onquality. They found that competitive intensity providedan explanation for the absence of returns from invest-ments in quality capital. Obtaining customer satisfactionperformance from quality practices was shown to becontingent on the degree of international competitionpresent in the business environment.

Curkovic, Vickery, and Droge (2000) studied thedirect effects of 10 quality action programs on six firmperformance outcomes as well as their indirect effectsthrough eight quality performance dimensions. Twopaths from action programs through quality perform-ance to firm performance in the automotive supplyindustry were identified. The first path was productquality, whose hallmarks were superior performance onconformance and design quality. The second pathinvolved relationship quality. This path included supe-rior customer responsiveness and service. Both theproduct-quality and the relationship-quality paths ledto superior operational ROI. Product quality led toenhanced return on assets (ROA), and relationshipquality led to enhanced market share performance.

Flynn, Schroeder, and Sakakibara (1995) con-structed a framework that focuses on both core qualitymanagement practices and the infrastructure used toengender an environment supportive of their use. Theyincorporated two measures of quality performance andtheir roles in establishing and sustaining competitiveadvantage. The author used path analysis to test a pro-posed model that explained that perceived qualitymarket outcomes were primarily related to statisticalcontrol/feedback and the product design process, whilethe internal measure of percent that passed finalinspection without requiring rework was stronglyrelated to process flow management and statisticalcontrol/feedback. The focus of most Six Sigma efforts

is on operational improvements. By focusing on theprocesses and variables associated with operationaloutcomes (for example, y=f(x)), it is expected thatSix Sigma programs will result in more efficient usesof assets and higher ROI. The general proposition isstated as:

Proposition 2: Six Sigma adoption results in betteroperational results.

Using COMPUSTAT, the author computed severaloperational measures using productivity ratios. Part ofthe process of Six Sigma is to improve the use of assets.During the improve phase of the DMAIC cycle, DOEinvolves tolerance design, systems design, and parame-ter design (Foster 2006). Systems design involveschoices, tradeoffs, and improvements to existing assets.These improvements may include the optimization ofplant, equipment, or technology. Implicit in thisimproved usage is a more productive usage of assets.This results in hypotheses H2a through H2d.

• H2a: Six Sigma adoption is positively associated withasset turnover.

• H2b: Six Sigma adoption is positively associated withreturn on assets.

• H2c: Six Sigma adoption is positively associated withreturn on investment.

• H2d: Six Sigma adoption is positively associated withtotal assets.

Improvement efforts such as Six Sigma could help acompany to grow as profitability improves. Often theseefforts result in downsizing and the more productive useof employees. Therefore, Six Sigma could result in eithergrowth in the number of employees or a reduction in thenumber of employees. The literature is not clear on thisissue. Therefore, hypothesis H2e is stated in null form.

• H2e: Six Sigma adoption is not related to number ofemployees.

METHODSThis study examines the long-term financial and opera-tional impacts of implementing Six Sigma. In order tocreate a population of Six Sigma-adopting firms, theauthor performed a keyword Lexis/Nexis (LexisONE)

www.asq.org 11

Foster v14 i4 9/20/07 10:10 AM Page 11

12 QMJ VOL. 14, NO. 4/© 2007, ASQ

Does Six Sigma Improve Performance?

online search of corporate financial reports for the termsSix Sigma, ISO 9000, TQM, and Baldrige through theLexisONE subscription database. Since annual reportsare used to transmit information to shareholders, themention of a quality management initiative such as SixSigma is interpreted as signaling an organizationalcommitment to that effort. Since annual reports areused to elucidate strategic emphases, it can be said thatthese quality improvement efforts are strategic in nature.Using this rationale, the author performed a contentanalysis to determine which firms had announced theimplementation of Six Sigma. Initially, the year 1998was searched. Since only 12 Six Sigma firms werereturned, the author also searched 1996 and 1997. Thiscontent analysis of annual reports resulted in a list of 30companies that had adopted Six Sigma as a means ofimproving performance. In a further step, they assem-bled a panel of six judges, including academics andquality professionals, to perform a content analysis ofthe annual reports found in LexisONE to determinewhich firms showed strategic commitment to Six Sigma.Based on this content analysis, six firms were removedfrom the analysis, resulting in a final set of 24 Six Sigmafirms. They also found 24 Baldrige firms, 26 TQM firms,and 23 ISO 9000 firms. In addition, a randomly selectedlist of 50 companies from the 1998 Fortune 500 list wasused as a control group.

Company ticker symbols were obtained usingHoover’s Online, and the database of firms the authorcreated was searched for duplicates and firms that werereturned in more than one search. If a firm in the controlgroup was also returned in the list of Six Sigma compa-nies, it was deleted from the control group and anotherfirm in the Fortune 500 was randomly selected. In addi-tion, firms no longer in business or that had beenacquired since 1998 were also stricken. In the end, thedatabase of control firms was reduced to 41 companies.

As stated, a modified version of Hendricks andSinghal’s (1997) performance measures was used.Profitability measures used were earnings (EBITDA),ROA, operating margin (EBITDA/sales), and net sales.Cost structure was analyzed by calculating cost perdollar of sales and FCF per share. Relative firm growthmeasures were obtained using total assets and numberof employees.

Firm data were collected using Standard and Poor’sCOMPUSTAT database. Data were collected for the firmsdrawn from the Lexis/Nexis database for fiscal years1998 through 2004. This provided data for the author’slongitudinal analysis.

In this article, the author compares results from thetime Six Sigma was mentioned in the annual report tofour years later. This four-year interval is needed, as ittakes time for quality improvement programs to yieldsignificant results. In a study of best quality practices,Ozan (1992) found that quality improvement pro-grams should be implemented gradually. A study by theUnited States General Accounting Office (1991) statedthat, on average, 3.5 years were required to see signifi-cant results for TQM programs. In a study of the U.S.auto industry, Narasimhan, Ghosh, and Mendez (1993)found a 2.26-year lag between quality improvementefforts and sales improvements. Foster (1996) studiedthe speed of quality improvement in five different pro-duction facilities and found that plants that improvedmore slowly actually had better financial results thanplants that attempted to improve rapidly. Given theseprior research results, it was thought that a four-yearinterval would provide necessary time to see significantimprovement in results if it was present.

RESULTSThe results are organized according to the outcomevariables studied. For each dependent variable theauthor discusses data transformation (when needed).Next, analysis of covariance (ANCOVA) results for maineffects are presented. ANCOVA was used to test where sig-nificant differences exist between outcomes for each ofthe treatments. ANCOVA is a general linear model withone explanatory variable and multiple factors. It wasapplied where a potentially strong correlation existsbetween independent and dependent variables. Then,the author explains interactions between various qualityprograms (Six Sigma, Baldrige, TQM, and ISO 9000)and no quality program relative to each financial orperformance outcome. The interaction tests usedANCOVA to determine if there are significant interactioneffects between quality management programs (QMP)and outcomes. Model-based estimates are presented for

Foster v14 i4 9/20/07 10:10 AM Page 12

Does Six Sigma Improve Performance?

companies in the 10th percentile, median, and 90thpercentile. The percentiles were determined according tofirm size, as prior research has suggested that firm sizehas a moderating effect on QMP effects (Hendricks andSinghal 2001a; York and Miree 2004). Finally, hypothe-sis tests were performed to study the author’s researchpropositions. The results are presented according tofinancial or performance outcome in order to enhanceclarity of presentation.

Free Cash Flow Per Share Summary statistics for free cash flow per share (FCF) in1998 and 2002 are presented in Table 1. This summaryshows that 138 companies were evaluated for thisresearch. These companies are rank ordered, with the34th company in the 25th percentile, company 69 at themedian, and the 104th company at the 75th percentile.These statistical tests were performed using SAS.

The model for FCF in 2002 was fit with all datapresent and no transformations. Table 2 shows theANCOVA results for FCF (as well as the other measures).The author has combined tables here for efficient pres-entation. Table 2 shows that the main effect for FCFwas not significant. However, as is shown in Table 3,the interaction term for FCF by QMP was significant (p < 0.0002). With a low R2 of 0.22, this suggests thatother variables are having an effect on an otherwisenull model. The first five rows in Table 4 present

model-based estimates of FCF in 2002 at key levels ofFCF in 1998. For reference, Table 5 shows the countsof companies in various percentiles of FCF in 1998 toprovide an indication of the range of data behind theestimates shown in previous tables and figures.

Companies without a QMP appear to have a differ-ent relationship between FCF in 2002 and 1998 thanall other companies. While FCF in 2002 appears to bemore or less positively related to FCF in 1998, these

www.asq.org 13

Table 2 ANCOVA results main effects.

Term F value Pr>F

Free cash flow .94 0.3300

Log cost per dollar sales 36.07 <.0001

EBITDA 9.57 0.0027

Log of sales 579.24 <.0001

Sales per employee 306.97 <.0001

Asset turnover 3.01 0.0800

ROA 1.12 0.2910

ROI 1.81 0.1806

Total assets 1.66 0.2006

Log number of employees 860.94 <.0001 ©20

07, A

SQ

Table 1 Summary statistics for FCF in 1998and 2002.

1998 2002

Number of companies 138 138

Minimum -14.05 -22.18

25th percentile -0.54 0.33

Median 0.25 1.05

75th percentile 0.85 2.38

Maximum 17.58 12.40

Mean 0.19 1.19

Standard deviation 2.82 3.63 ©20

07, A

SQ

Table 3 ANCOVA results interaction terms.

Term F Value Pr>f

Free cash flow by QMP 6.08 0.0002

Log cost per dollar sales by QMP 16.99 <.0001

EBITDA by QMP 8.56 <.0001

Log sales by QMP .83 0.5101

Sales per employee by QMP 0.46 0.7600

Asset turnover by QMP 14.94 <.0001

ROA by QMP 2.45 0.0490

ROI by QMP 2.42 0.0517

Total assets by QMP 17.4 <.0001

Log employees by QMP 1.14 0.3400 ©20

07, A

SQ

Foster v14 i4 9/20/07 10:10 AM Page 13

14 QMJ VOL. 14, NO. 4/© 2007, ASQ

Does Six Sigma Improve Performance?

companies have a negative relation-ship between FCF in 2002 and 1998.Figure 1 plots this interaction.

The author evaluates H1a in Table 6.Among companies with high FCF in1998, those with no QMP have an esti-mated FCF for 2002 that is substantiallylower than all other companies, whereascompanies using Six Sigma have an esti-mated 2002 FCF that is substantiallyhigher than all other companies. Thus,they see Six Sigma companies and theaverage of all companies with a QMP ashaving significantly higher FCF in 2002than those companies without a QMP.At low FCF in 1998, however, thosecompanies without a QMP had signifi-cantly higher FCF in 2002, leading tothe patterns of significance displayedin Table 6.

Cost Per Dollar SalesIt was evident from the review of the1998 data that cost per dollar sales valuewas extremely skewed, and to betteraccommodate model fitting, the authorfitted the model to the natural log ofcost per dollar sales. The 2002 values,however, were evenly distributed.

Table 2 shows the ANCOVA resultsfor the model fit to logged cost per dol-lar sales. As shown in Table 3, becauseof the significant interaction between1998 cost per dollar sales and QMP, theeffect of QMP is different depending onthe level of cost per dollar sales in1998. Table 4 gives the model-basedestimates for mean (logged) cost perdollar sales in 2002 by QMP and levelof 1998 logged cost per dollar sales.

As can be seen in the plots ofmeans in Figure 1, the interactionappears to result from different char-acteristics with respect to cost per

Table 4 Model-based mean estimates (std. error of estimates)of effects in year 5, based on levels of year 1.

Free Cash Flow 10th Median 90th

Six Sigma 1.59 (0.75) 1.64 (0.68) 1.7 (0.83)Baldrige 1.27 (0.76) 1.94 (0.73) 2.65 (1.04)TQM -0.29 (1.35) 1.61 (0.69) 3.61 (1.38)ISO 0.79 (1.03) 1.21 (0.77) 1.65 (1.43)No QMP 2.59 (0.67) 1.13 (0.55) -0.4 (0.56)

Cost Per Dollar Sales

Six Sigma -0.18 (0.06) -0.13 (0.02) -0.11 (0.04)Baldrige -0.34 (0.04) -0.13 (0.02) -0.02 (0.03)TQM -0.1 (0.02) -0.1 (0.02) -0.098 (0.02)ISO -0.16 (0.03) -0.07 (0.02) -0.02 (0.03)No QMP -0.35 (0.03) -0.16 (0.02) -0.07 (0.02)

EBITDA

Six Sigma 2.76 (3.56) 11.88 (2.97) 23.99 (3.14)Baldrige 15.11 (4.96) 8.73 (3.49) 0.263 (3.79)TQM 2.63 (4.8) 7.58 (3.07) 14.15 (7.04)ISO 10.98 (4.27) 13.68 (3.29) 17.26 (3.79)No QMP 5.69 (3.17) 13.8 (2.19) 24.57 (3.22)

Asset Turnover

Six Sigma 15,272 (2,550) 10,116 (2,083) 812 (2,725)Baldrige -2,352 (3,392) -627 (2,685) 2,487 (2,826)TQM 11,426 (5,604) 4,765 (2,519) -7,255 (6,004)ISO 9,435 (8,878) 4,937 (2,614) -3,182 (12,084)No QMP 6,266 (1,957) 6,011 (1,579) 5,551 (2,252)

ROA

Six Sigma 2.18 (2.25) 2.28 (1.75) 2.37 (2.34)Baldrige -1.17 (2.77) -1.12 (1.77) -1.13 (2.42)TQM 3.74 (3.76) 3.08 (1.84) 2.55 (2.54)ISO -1.52 (1.96) 2.65 (1.88) 5.96 (2.08)No QMP 1.98 (1.71) 2.87 (1.29) 3.58 (1.8)

ROI

Six Sigma 3.91 (3.8) 4.25 (2.98) 4.47 (3.67)Baldrige -6.05 (4.76) -0.82 (3.02) 2.63 (3.68)TQM 7.26 (7.96) 6.88 (3.06) 6.63 (5.32)ISO -2.08 (3.77) 6.81 (3.45) 12.67 (4.26)None 8.2 (3.3) 6.46 (2.19) 5.3 (2.93)

Total Assets

Six Sigma -705.3 (1659.4) 1894 (1 j610.9) 6140.6 (1562.7)Baldrige 7682.9 (1937.1) 7523.7 (1804.5) 7263.5 (1741.2)TQM -28.3 (3916.3) 1784.6 (1638.3) 4746.2 (6746.5)ISO 1502.8 (1846.6) 1950.3 (1718.7) 2681.4 (2220.2)None 4135.2 (1193.8) 4545.1 (1167.6) 5214.9 (1396.8) ©

2007

, ASQ

Foster v14 i4 9/20/07 10:10 AM Page 14

Does Six Sigma Improve Performance?

dollar sales among compa-nies using TQM and thoseusing either no QMP or oneof the other three understudy. All companies otherthan those with TQM appearto have a direct relationshipbetween cost per dollar salesin 2002 and 1998, whileTQM companies appear tohave no such relationship.

Evaluation of the hypotheses of interest is given inTable 6. Null hypotheses regarding Six Sigma compa-nies cannot be rejected. Among companies with low costper dollar sales in 1998, companies using the BaldrigeAward criteria have a significantly lower predicted costper dollar sales than companies using any of the otherthree QMPs. Therefore, H1b is not supported.

EBITDAA review of the data showed that the distribution ofEBITDA in 1998 was skewed because of the presence ofoutlying companies. In 2002 the distribution of EBITDA

was fairly symmetric, but it was peaked. Therefore, themodel for EBITDA in 2002 was fit with all data presentand no transformations.

Table 2 shows significant ANCOVA results (p < .0027).Table 3 shows the model-based estimates for EBITDA in2002. Because of the significant interaction betweenEBITDA 1998 and QMP, the effect of QMP is differentdepending on the level of EBITDA in 1998. Table 4shows model-based estimates of average EBITDA in2002 at levels of 1998 EBITDA.

Inspection of Figure 1 suggests the following as aninterpretation of the interaction: Among companies withthe highest EBITDA in 1998, those using the Baldrige

www.asq.org 15

Table 5 Count of companies in each percentile category and QMP.

Six Sigma Baldrige TQM ISO None Total

0–10th percentile 3 6 1 4 14

10th–50th percentile 7 10 14 16 7 54

50th–90th percentile 12 5 11 6 22 56

90th percentile and higher 2 3 1 8 14

Total 24 24 26 23 41 138 ©20

07, A

SQ

FCF 1998 by QMP interaction

FCF

2002

FCF 1998

Six SigmaBaldrigeTQMISONone

Six SigmaBaldrigeTQMISONone

Six SigmaBaldrigeTQMISONone

Six SigmaBaldrigeTQMISONone

Six SigmaBaldrigeTQMISONone

Six SigmaBaldrigeTQMISONone

Six SigmaBaldrigeTQMISONone

0

1

2

3

3.5

4

2.5

1.5

0.5

-0.5

-1

0 1 2 3-1-2

Return on Assets by QMP interaction

1998 Return on Assets

2002

Ret

urn

on A

sset

s

0 2 4 6 8 10-4 -20

1

2

3

4

5

6

7

-1

-2

Asset Turnover

1998 Asset Turnover level

2002

Ass

et T

urno

ver

20,000

15,000

10,000

5,000

(5,000)

(10,000)

0-1000-2000-3000 1000 2000 3000 4000 5000 6000

Logged Cost per dollar sales by QMP interaction

Logg

ed c

ost/

dolla

r sal

es, 2

002

Logged Cost/dollar sales, 1998

0-0.35 -0.3 -0.25 -0.2 -0.15 -0.1 -0.050

-0.05

-0.1

-0.15

-0.2

-0.25

-0.3

-0.35

-0.4

ROI 1998 by QMP interaction

ROI,

2002

0

5

15

10

-5

-10

-4.1 8.4 16.6

ROI, 1998

Mean EBITDA in 2002, by QMP and EBITDA in 1998

EBIT

DA

200

2

EBITDA 1998

30.00

25.00

20.00

15.00

10.00

5.00

0.000 5 10 15 20 25 30

Total assets by QMP interaction

Tota

l ass

ets

2002

Total Assets 1998

0-1000 1000 2000 3000 4000 5000 6000 70000

8000

7000

6000

5000

4000

3000

2000

1000

-1000

-2000

9000

Figure 1 Interaction Plots by Effect.

©20

07, A

SQ

Foster v14 i4 9/20/07 10:10 AM Page 15

16 QMJ VOL. 14, NO. 4/© 2007, ASQ

Does Six Sigma Improve Performance?

Award criteria had the low-est EBITDA in 2002, andthose using Six Sigma or noQMP had the highest. Thatresult is somewhat reversedamong companies with thelowest EBITDA in 1998.

Evaluation of thehypotheses of interest isgiven in Table 6. TheEBITDA in 2002 is statisti-cally equivalent betweenSix Sigma companies andcompanies with no QMP,regardless of EBITDA in1998. Among companieswith a QMP, Six Sigmacompanies have a higherEBITDA in 2002. However,this only holds amongcompanies with a higherEBITDA in 1998; for allothers, performance ofEBITDA in 2002 is equiva-lent. This provides partialsupport for H1c.

SalesDespite a large range ofsales data, the raw datawere left in their originalscales. The author exam-ined the summary of salesdata and found they wereextremely skewed. To bet-ter accommodate modelfitting, he modeled the natural log of sales.

Table 2 gives the ANCOVA table for the model forlogged sales. While the overall R2 for the model is 0.93,no terms other than sales in 1998 have any explanatorypower for sales in 2002 (see Table 3). For this reason, theauthor concludes that the null hypotheses of interestcannot be rejected and that Six Sigma has no influenceon sales. Therefore, H1d is not supported.

Sales Per EmployeeThe raw data were divided by 1,000 to drop the scale to amore useable level. It was evident from the summarydata that this value was extremely skewed, and to betteraccommodate model fitting, the author used the naturallog of sales per employee (in thousands). Table 2 givesthe ANCOVA table for the model fit. While the overall

Table 6 Hypothesis results.

Free cash flow 10th1 Median2 90th3

Six Sigma vs. No QMP -1 (0.99) 0.5 (0.87) 2.1 (1.01)*

Six Sigma vs. Other QMP 1 (0.99) 0.05 (0.81) -0.94 (1.12)*

Any QMP vs. No QMP 1.75 (0.85)* -0.46 (0.67) -2.8 (0.82)*

Cost per dollar sales

Six Sigma vs. No QMP 0.17 (0.08) 0.03 (0.03) -0.04 (0.05)

Six Sigma vs. Other QMP 0.01 (0.07) -0.03 (0.03) -0.06 (0.05)

Any QMP vs. No QMP -0.16 (0.04)* -0.06 (0.02)* -0.01 (0.03)

EBITDA

Six Sigma vs. No QMP -2.93 (4.71) -1.92 (3.66) -0.57 (4.54)

Six Sigma vs. Other QMP -6.81 (4.44) 1.88 (3.52) 13.43 (4.3)*

Any QMP vs. No QMP -2.17 (3.84) 3.33 (2.76) 10.65 (4.13)*

Asset turnover

Six Sigma vs. No QMP 9005.74 (2813.7)* 4105.52 (2592.55) -4738.12 (2696.38)

Six Sigma vs. Other QMP 9101.92 (4355.24)* 7091.12 (2571.04)* 3462.13 (5065.76)

Any QMP vs. No QMP -2179.08 (3458.55) 1213.03 (2110.05) 7334.95 (3709.69)

ROA

Six Sigma vs. No QMP 0.20 (2.82) -0.58 (2.17) -1.21 (2.97)

Six Sigma vs. Other QMP 1.81 (2.85) 0.76 (2.06) -.09 (2.70)

Any QMP vs. No QMP 1.16 (2.22) 1.15 (1.63) 1.15 (2.21)

Total assets

Six Sigma vs. No QMP -4,840.5 (2,029.3) -2,651.1 (1,975.6) 925.7 (2,087.3)

Six Sigma vs. Other QMP -3,757.9 (2,293.2) -1,858.9 (1,894.2) 1,243.5 (2,895.8)

Any QMP vs. No QMP 2,022.1 (1,754) 1,257 (1,476.2) 7 (2,362.3)

* p < .0001 ©20

07, A

SQ

Foster v14 i4 9/20/07 10:10 AM Page 16

Does Six Sigma Improve Performance?

R2 for the model is a high 0.77, no terms other thansales per employee in 1998 have any explanatory powerfor sales per employee in 2002 (see Table 3). For thisreason, they concluded that the null hypotheses H1ecannot be rejected and that QMP has no influence onsales per employee.

Asset TurnoverOnce again, Table 2 shows the ANCOVA table for themodel fit, and Table 4 shows the model-based esti-mates for asset turnover in 2002. These estimatesindicate that companies with high asset turnover in1998 had a lower 2002 asset turnover than companieswith a low asset turnover in 1998. This is generallytrue for companies using Six Sigma, TQM, and ISO9000. The 2002 asset turnover for Baldrige companiesshows the opposite pattern (companies with high assetturnover in 1998 had higher asset turnover in 2002than those with lower 1998 asset turnover). Thosecompanies with no QMP showed no real difference in2002 asset turnover, regardless of their 1998 assetturnover. Figure 1 plots these estimates from Table 4.

Evaluation of the hypotheses of interest is given inTable 6. Among companies with low and median 1998asset turnover, companies using Six Sigma QMP hadsignificantly higher asset turnover in 2002 than com-panies with no QMP; the asset turnover of these SixSigma companies was also higher than the average ofcompanies with the other three programs. Amongcompanies with high asset turnover in 1998, there wasno significant difference among any of the hypothesesof interest. Therefore, H2a is partially supported.

Return on AssetsTable 2 gives the ANCOVA table for the model fit, andTable 3 shows the model-based estimates for ROA in2002. Table 3 shows a significant interaction betweenROA and sales (p < .05). Table 4 provides the model-based estimates of average 2002 return on investmentsat levels of 1998 ROA.

Evaluations of the hypotheses of interest are givenin Table 6. While the interaction is significant, noneof the hypotheses of interest are. Table 6 provides

means and standard errors of the return on asset dataused in the model by QMP, company size, and year.Therefore, H2b is not supported.

Return on InvestmentROI for either year looks bell-like in distribution exceptfor the presence of very long tails that show outlyingobservations in both the positive and negative direction.Attempts at transformation did little to pull in thesestrong outliers. Thus, the ANCOVA model was fit on theraw data. Table 2 shows the ANCOVA table. As one cansee in Table 3, at p < .05, the interaction of QMP byROI in 1998 is technically not statistically significant.However, it is of borderline significance. Table 5 givesthe model-based estimates for ROI in 2002 by per-centiles of ROI in 1998 and QMP. Figure 1 plots thesemeans for easier interpretation. From Figure 1 itappears that Baldrige and ISO 9000 companies may beresponding differently than the companies with theother three QMPs. This is descriptive only, however,owing to the lack of significance of any of the terms inthe ANCOVA model. Therefore, H2c is not supported.

Total AssetsTo bring the data to a more manageable scale, theoriginal data were divided by 1,000. The model was fitwith all available data for total assets in 2002. Thenonsignificant ANCOVA result is presented in Table 2.As is shown in Table 3, because of the significantinteraction between total assets 1998 and QMP, theeffect of QMP is different, depending on the level oftotal assets in 1998. Table 4 shows the model-basedestimates of the average total assets (in thousands) for2002 for each QMP by levels of total assets in 1998.

Figure 1 shows a plot of the estimates to facilitategeneral interpretation of the interaction. Generally, itappears that the relationship between 1998 total assetsand 2002 total assets does not hold for companiesusing Baldrige, ISO 9000, or no QMP.

Table 6 contains the tests of the hypotheses ofinterest. Among companies at the highest level of totalassets in 1998, there is no difference in 2002 totalassets. For all others, however, those using the

www.asq.org 17

Foster v14 i4 9/20/07 10:10 AM Page 17

18 QMJ VOL. 14, NO. 4/© 2007, ASQ

Does Six Sigma Improve Performance?

Baldrige Award criteria have higher total assets in2002 than those using any other QMP. Therefore, H2dis not supported.

Number of EmployeesBecause these data are skewed, the author used thenatural logarithm as the outcome in modeling efforts.Table 2 gives the results of the ANCOVA model. For thismodel, adding the indicator for company size was notappropriate, so it was not included here. Because noterm other than the number of employees in 1998 isstatistically significant, H2e is not supported.

DISCUSSION AND CONCLUSIONSAs has been seen, the results from Six Sigma programsare at best mixed. The author found a significant effecton FCF, EBITDA, and asset turnovers; however, SixSigma did not seem to affect sales, ROA, ROI, or firmgrowth. As a result, if firms want to improve cash,earnings, or productivity in using assets, Six Sigmamight be of use. An intervening variable in this analy-sis was firm size, which confirmed Hendricks andSinghal’s finding on the importance of firm size. Itappears that firm size mitigates the effects of programslike Six Sigma. Certainly, large firms such as GE andMotorola have resources and assets to invest in SixSigma programs. These firms also showed greatcapacity for improvement. Smaller firms may nothave the same resources to apply to such programs;however, smaller firms that invest in efforts such asSix Sigma may be impacted more significantly as atotal proportion of the firm’s business results.

Companies with high cash flows in 1998 and noQMP had lower FCF in 2002 than companies that hadadopted Six Sigma. Companies with low cash flowand no QMP did better than companies using SixSigma, suggesting that for cash-poor firms, Six Sigmamay in fact be a drain on resources. Also, these com-panies may not have the cash necessary to sustaineffective Six Sigma results over four years. Therefore,H1a was supported for firms with high cash flows, butmay have a negative effect for companies with poorcash flows.

Among companies with low and medium assetturnover, Six Sigma led to higher asset turnover.Therefore, H2a was supported for companies with lowand medium asset turnover in 1998. It could be thatthese companies with low asset turnover could benefitmore from process improvement implicit in Six Sigmathan firms with high asset turnover.

The adoption of Six Sigma seemed to not affect ROAand ROI significantly. This may not be surprisingbecause many factors affect ROA and ROI, as evidencedby the DuPont model. Further investigation showed thatlarge firms using Six Sigma performed better. That is,the improvement rates were in the correct direction. Thedifferences, however, were not significant. Reflectingon the original research question regarding financialand operational outcomes, it appears that at a macrolevel, the effects, while in some cases significant, aresomewhat modest. While the author did see benefits inboth financial and operational areas, quality profes-sionals should be careful to not oversell the benefits ofSix Sigma or any other QMP.

That Six Sigma did not affect firm growth was notsurprising. There were many exogenous macroeconomicfactors, including 9/11, that could have influencedgrowth during this period. Powell (1995) and othershave stated that there are many factors that affect firmgrowth other than quality management approaches.Indeed, many quality programs have been associatedwith reductions in number of employees. On the otherhand, that one is able to find any significant results forthese small numbers of firms demonstrates that qualityprograms such as Six Sigma can potentially provideimportant financial and operational returns. Thisresearch is very high level with much room for futurestudy. Future studies should examine specific effects onspecific projects in companies. It is expected that suchstudies could verify and clarify these findings.

LimitationsThere were several limitations to this research. As wasdiscussed previously, financial results during this periodwere influenced by 9/11 and other events. In addition,these firms’ use of Six Sigma or other quality programsin 1998 does not guarantee that they implemented these

Foster v14 i4 9/20/07 10:10 AM Page 18

Does Six Sigma Improve Performance?

programs effectively or maintained their strategicsupport through 2002 and beyond. Of course, thenormal provisos concerning COMPUSTAT hold, asthese data were firm-level data, and many times SixSigma programs are implemented at the business unitlevel. While the methodology the author used isadapted from prior QMP studies, it is difficult to isolatethe effects of Six Sigma at a macro level. Future stud-ies at a more micro level may help the understandingof these phenomena.

This study included firms from a variety of indus-tries. For future research, the author suggests industry-specific studies of the effects of Six Sigma adoption. Ithas been suggested that Six Sigma is best suited forprocess industries. It is expected that in some ways suchindustries would have been leading-edge adopters.

Implications for ManagersThis research has implications for managers. Theauthor found negative cash flows for small firms, butpositive cash flows for larger firms. It is apparent thatmanagers in smaller firms should carefully considerthe cash flow impacts of quality improvement invest-ments. Managers should also carefully monitor theeffects of Six Sigma efforts and focus on “hard dollar”benefits. If not skillfully implemented, the benefits ofSix Sigma may be marginal. The “fad” element of SixSigma should also be noted. It is more likely thatcompanies that understand the benefits of Six Sigmaare able to effectively manage to achieve those bene-fits. It could be that companies that implement SixSigma to merely parrot industry practice will likelyfind modest competitive benefits from Six Sigma orany other quality improvement effort.

REFERENCES

Adams, G., G. McQueen, and K. Seawright. 1999. Revisiting thestock price impact of quality award. Omega 27: 595-604.

Antony, J., and J. Esamilla. 2003. Lean sigma. ManufacturingEngineering (April): 40-42.

Bisgaard, S., and J. DeMaat. 2006. After Six Sigma, what’s next?Quality Progress 39, no. 1: 30-36.

Conklin, R. 2006. Measurement system analysis for attributemeasuring processes. Quality Progress 39, no. 3.

Cua, K. O., K. E. McKone, and R. G. Schroeder. 2001. Relationshipsbetween implementation of TQM, JIT, and TPM and manufacturingperformance. Journal of Operations Management 19: 675-694.

Curkovic, S., S. Vickery, and C. Droge. 2000. An empiricalanalysis of the competitive dimensions of quality performance inthe automotive supply industr y. Internat ional Journal ofOperations & Production Management 20, no. 3: 386-403.

Das, A., R. Handfield, R. Calantone, and S. Ghosh. 2000. Acontingent view of quality management: The impact of interna-tional competition on quality. Decision Sciences Journal 31, no. 3:649-690.

Dean, M., and C. Tomovic. 2004. Does Baldrige make a businesscase for quality? Quality Progress 37, no. 4: 40-46.

Deming, W. 1986. Out of the crisis. Cambridge, Mass.: MIT Press.

Devaraj, S., K. Matta, and E. Conlon. 2001. Product and servicequality: The antecedents of customer loyalty in the automotive indus-try. Production and Operations Management 10, no. 4: 424-439.

Douglas, T. J., and W. Q. Judge Jr. 2001. Total quality manage-ment implementation and competitive advantage: The role ofstructural control and exploration. Academy of ManagementJournal 441: 158-169.

Dow, D., D. Samson, and S. Ford. 1999. Exploding the myth: Doall quality management practices contribute to superior quality per-formance? Production and Operations Management 8, no. 1: 1-27.

Easton, G. S., and S. L. Jarrell. 1998. The effects of total qualitymanagement on corporate performance: An empirical investiga-tion. Journal of Business 71, no. 2: 253-307.

Eriksson, H., and J. Hansson. 2003. The impact of TQM onfinancial performance. Measuring Business Excellence 7, no. 1:36-50.

Flynn, B. B., R. G. Schroeder, and S. Sakakibara. 1995. The impactof quality management practices on performance and competitiveadvantage. Decision Sciences Journal 26, no. 5: 659-684.

Foster, S. T. Jr. 1996. Examining the impact of speed of qualityimprovement on quality-related costs. Decisions Sciences Journal24, no. 4: 623-646.

Foster, S. T. Jr. 2006. Managing quality: Integrating the supplychain. Upper Saddle River, N.J.: Prentice Hall.

Foster, S. T. Jr. 2007. Quality survival guide: Leadership. QualityProgress 40, no. 7: 25-35.

Fullerton, R. R., C. S. McWatters, and C. Fawson. 2003. Anexamination of the relationship between JIT and financial per-formance. Journal of Operations Management 21: 383-404.

Fynes, B., and C. Voss. 2001. A path analytic model of qualitypractices, quality performance, and business performance.Production and Operations Management 10, no. 4: 494-513.

Gourishankar, T. 2003. Back to basics: A simple process map.Quality Progress 39, no. 1: 112-120.

www.asq.org 19

Foster v14 i4 9/20/07 10:10 AM Page 19

20 QMJ VOL. 14, NO. 4/© 2007, ASQ

Does Six Sigma Improve Performance?

Hendricks, K. B., and V. R. Singhal. 1996. Quality awards andthe market value of the firm: An empirical investigation.Management Science 42, no. 3: 415-436.

Hendricks, K. B., and V. R. Singhal. 1997. Does implementing aneffective TQM program actually improve operating performance?Empirical evidence from firms that have won quality awards.Management Science 43, no. 9: 1258-1274.

Hendricks, K. B., and V. R. Singhal. 2001a. Firm characteristics,total quality management, and financial performance. Journal ofOperations Management 19: 269-285.

Hendricks, K. B., and V. R. Singhal. 2001b. The long-run stockprice performance of firms with ef fective TQM programs.Management Science 47, no. 3: 359-368.

Howard, L., S. T. Foster, and P. Shannon. 2005. Team climateand teamwork in government: The power of embedded leader-ship. International Journal of Quality and Reliability Management22, no. 8: 769-795.

Jacobsen, R., and D. Aaker. 1987. The strategic role of productquality. Journal of Marketing 51: 31-44.

Kaynak, H. 2003. The relationship between total quality man-agement practices and their effects on firm performance. Journalof Operations Management 21: 405-435.

Lapre, M., A. Mukherjee, and L. VanWassenhove. 2000. Behindthe learning curve: Linking learning activities to waste reduction.Management Science 46, no. 5: 597-611.

Linderman, K., R. Schroeder, S. Zaheer, and A. Choe. 2003. SixSigma: A goal theoretic perspective. Journal of OperationsManagement 21, no. 2: 193-204.

Mohrman, S. A., R. V. Tenkasi, E. E. Lawler III, and G. E. LedfordJr. 1995. Total quality management: Practice and outcomes in thelargest U.S. firms. Employee Relations 17, no. 3: 26-41.

Montgomery, D. 2004. Design and analysis of experiments.Hoboken, N.J.: John Wiley and Sons.

Narasimhan, R., S. Ghosh, and D. Mendez. 1993. A dynamicmodel of product quality and pricing decisions on sales response.Decision Sciences Journal 20, no. 4: 893-908.

Ozan, T. 1992. International quality study: Best practices report.New York: Ernst and Young.

Pannirselvam, G., S. Siferd, and W. Ruch. 1998. Validation ofthe Arizona governor’s quality award: A test of the Baldrige cri-teria. Journal of Operations Management 16, no. 6: 529-550.

Park, S., J. Hartley, and D. Wilson. 2001. Quality managementpractices and their relationship to buyer’s supplier rating: A studyin the Korean automotive industry. Journal of OperationsManagement 19, no. 3: 695-712.

Phillips, L. W., D. R. Chang, and R. D. Buzzell. 1983. Productquality, cost position, and business performance: A test of somekey hypothesis. Journal of Marketing 47 (Spring) 26-43.

Powell, T. C. 1995. Total quality management as competitiveadvantage: A review and empirical study. Strategic ManagementJournal 16, no. 1: 15-37.

Pyzdek, T. 2003. The Six Sigma handbook: The complete guidefor greenbelts, blackbelts, and managers at all levels. New York:McGraw-Hill.

Rajan, M., and N. Tamimi. 1999. Baldrige award winners: Thepayoff to quality. Journal of Investing 8, no. 4: 39-43.

Samson, D., and M. Terziovski. 1999. The relationship betweentotal quality management practices and operational perform-ance. Journal of Operations Management 17: 393-409.

Sousa, R., and C. Voss. 2001. Quality management: Universal orcontext dependent? Production and Operations Management 10,no. 4: 383-404.

Treichler, D., R. Carmichael, A. Kusmanoff, J. Lewis, and G.Berthiez. 2002. Design for Six Sigma: 15 lessons learned.Quality Progress 35, no. 1: 33-43.

United States General Accounting Office. 1991. Managementpractices: U.S. companies improve performance through qualityefforts. Washington, D.C.: GAO Report to the Honorable DonRitter, House of Representatives.

Ward, P., K. Leong, and K. Boyer. 1994. Manufacturing proac-tiveness and performance. Decision Sciences Journal 25, no. 3:337-355.

Wilson, D. D., and D. C. Collier. 2000. An empirical investiga-tion of the Malcolm Baldrige National Quality Award causalmodel. Decision Sciences Journal 31, no. 2: 361-383.

York, K., and C. Miree. 2004. Causation and covariation: Anempirical re-examination of the link between TQM and financialperformance. Journal of Operations Management 22, no. 3:291-311.

BIOGRAPHY

S. Thomas Foster is area leader of global supply chain atBrigham Young University. He has published more than 50 quality-re la ted ar t ic les in journals such as Decis ion Sciences,International Journal of Production Research, the Journal ofOperations Management, the International Journal of Qualityand Reliability Management, the Quality Management Journal,and Quality Progress magazine. His book is titled, QualityManagement: Integrating the Supply Chain (Prentice Hall).Foster is founder of http://www.freequality.org, was awardedthe ASBSU 2000 Outstanding Faculty Award, and received theprestigious 2002 Decision Sciences Insti tute InstructionalInnovation Award. He is currently serving as guest editor for aspecial issue on supply chain qual i ty for the Journal ofOperations Management. He served twice as an examiner forthe Malcolm Baldrige National Quality Award and is a memberof ASQ. He can be reached by e-mail at tom_foster@byu.edu.

Foster v14 i4 9/20/07 10:10 AM Page 20

www.asq.org 21

The Long-Term Impact of ISO 9000Certification on Business Performance:A Longitudinal Study Using Turkish

Stock Market ReturnsMUSA PINAR AND CEYHUN OZGUR

VALPARAISO UNIVERSITY

© 2007, ASQ

The goal of this research was to examine the longitudinalimpact of ISO 9000 certification on business perform-ance. The study compared the monthly stock returns andvariability of the returns of ISO 9000 certified versusnon-ISO 9000 certified firms traded on the IstanbulStock Exchange (ISE) from January 1997 to September2005. The study also compared the stock returns and thevariability of the returns for the firms certified byTurkish agency vs. foreign certifying agency. Usingannual two-year, three-year, and four-year averages ofthe monthly stock returns, the results indicated thatISO 9000 certified firms generally had higher returnsand lower variability of returns than non-ISO 9000firms. Moreover, the comparisons of the longitudinalmeans covering successive years showed that ISO 9000certified firms consistently had higher means and lowervariances, some of which were statistically significant.Finally, the certifying agency (Turkish or foreign) andthe length of time with ISO 9000 certification had littleeffect on the stock performance and variability of thereturns.

Key words: business performance, ISO 9000 certification,quality, Turkey

INTRODUCTIONOver the last two decades, the implementation of qualitysystems has become a major initiative of companiesthroughout the world. The concepts of quality gurussuch as W. Edwards Deming, J. M. Juran, and Philip B.Crosby became highly publicized. Japan’s successfulimplementation of continuous improvement, Kaizen,and statistical quality tools, concepts, and practicesstarted the worldwide quality revolution. Eventually,Japan became the world leader in quality for a widerange of products. The emergence of Japan as a leaderin quality products eroded markets of U.S. companies.In response, U.S. companies began implementing totalquality management (TQM) in their operations in thelate 1970s and early 1980s. By improving the quality oftheir products, U.S. automobile makers recapturedmuch of their lost market share. For example, Chryslermoved from the brink of bankruptcy to being a highlyprofitable company. In summary, the quality awardwinners experienced increased income and sales afterthey implemented TQM. Effective TQM implementationresults in improved customer satisfaction, employeeinvolvement, and continuous improvement.

In response to the global quality movements, theEuropean Community (EC) countries initiated theirquality standards known as ISO 9000 certification. ISO9000 is an international standard for quality manage-ment systems. The International Organization forStandardization (ISO) formed Technical Committee 176(TC 176) to develop a universally accepted set of quality

Pinar v14 i4 9/20/07 10:10 AM Page 21

22 QMJ VOL. 14, NO. 4/© 2007, ASQ

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

standards. ISO adopted a series of quality standards in1987, known as the ISO 9000 series of standards. Thesestandards have been applied to firms of all sizes andfrom all areas of business. The worldwide impact of ISO9000 could be illustrated by the fact that the number ofcountries adopting the standards has continuallyincreased from 51 prior to 1992 (Bureau of BusinessPractices 1992) to 161 at the end of 2005 (ISO 2006a).Worldwide the number of ISO 9000 certificates grewfrom only a few in 1987 to 776,608 by the end of 2005(ISO 2006a). Large international industrial companiessuch as Corning, Exxon, Du Pont, Volkswagen, Sandoz,and Kodak initiated vigorous company programs toimplement standards at their operation sites (Rao,Ragu-Nathan, and Solis 1997).

ISO 9000 is not a quality standard per se, but a seriesof requirements for a quality management system. Asindicated in ISO 9001:2000 (Clause 1.1), the scope ofISO 9000 is to specify requirements for a quality man-agement system that: 1) consistently provides productsand services that meet the requirements of its cus-tomers; and 2) enhances customer satisfaction throughthe effective application of the system, includingprocesses for continual improvement and the assuranceof conformity to customer requirements. The 2000 ver-sion of ISO 9000 comprises three documents: ISO 9000,ISO 9001, and ISO 9004. The latest document providesinformation and guidelines to help organizations estab-lish and improve their quality management systems togo beyond the minimum requirements (Peyrat 2001).

Many have examined the motives for implementingISO 9000 standards and the benefits the companies havegained from ISO 9000 certification (Al-Ghamdi 1998;Barnes 1998; Bhuian 1998; Brumm 1995; Chittenden,Poutziouris, and Mukhtar 1998; Ebrahimpour, Whithers,and Hikmet 1997; Eddy 1995; LRQA 1996; MORI 1996;Morita 1996; O’Brien 1996; Peach 1997; Rabbit andBergh 1993; Rao, Ragu-Nathan, and Solis 1997; Raynerand Porter 1991; SGS Yarsley 1996; Struebing 1996;Vloeberghs and Bellens 1996; Weston 1995). These stud-ies found that the most common motives for implement-ing ISO 9000 fell into one of two broad categories ofinternal or external company motives or activities. Theinternal motives were improved customer service/reducedcustomer complaints, increased efficiency, and the desireto embed a quality culture or management control. The

external motives were customer requirements andresponding to competitors’ demands. In addition,achieving quality assurance certification is required bylegislation in several countries, and is implemented as amarketing and promotional tool.

Several reasons for seeking ISO 9000 certificationhave been suggested and/or documented empirically.Anderson, Daley, and Johnson (1999) found, in a sampleof both certified and noncertified U.S. firms between 1990and 1996, that contrary to the view held by many criticsof ISO 9000, managers obtained certification as a signalof effective quality management practices rather than inresponse to regulatory requirements. Two other surveysrevealed the most common reasons U.S. companies seekISO 9000 certification are: 1) to increase market share,meet customer requirements, and improve process effi-ciency (Ebrahimpour, Whithers, and Hikmet 1997); and2) to meet customer requiremetns and the belief that ISOcertification would increase operational efficiency andreduce costs (Struebing 1996). Compliance with variousregulatory and customer requirements were significantexplanatory variables for firms dealing with customers inthe EU (Anderson, Daley, and Johnson 1999). Brown, vander Wiele, and Loughton (1998) found that, in a study offirms in Western Australia, the main reasons for seekingISO 9000 certification included customer service, market-related factors, and potential quality improvement.

With increasing globalization of trade, the ISO 9000standards have rapidly become a force to be reckonedwith for any company doing business in internationalmarkets, either directly or indirectly. As a result, a consid-erable research effort has been devoted to investigatingthe awareness, attitudes, concerns, barriers, and pre-paredness of companies, especially manufacturing firms,in regard to ISO 9000 in different countries. In fact, it hasbecome a competitive necessity for firms doing businessglobally to acquire ISO 9000 certification. Several studiesprovided support for the strategic role of ISO 9000 certifi-cation for enhancing and facilitating international tradeand marketing, especially in Europe (Brumm 1995;Ferguson 1996; Morita 1996; Peach 1997; Rao, Ragu-Nathan, and Solis 1997; Reimann and Hertz 1996;Simmons and White 1999). As a series of standards, ISO9000 lays out the requirements for quality managementsystems, which ensures a company’s buyers that it at leasthas a fundamental quality system (Erel and Ghosh

Pinar v14 i4 9/20/07 10:10 AM Page 22

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

1997). In different countries standards provide a com-mon reference point of quality. For more informationabout ISO 9000 readers may refer to Clements (1993) orRothery (1993).

Quality Initiatives and ISO 9000 in Turkey In recent years, the quality of Turkish products exportedto European countries has been steadily improving.Evidence of this improvement is the number ofEuropean Quality Awards and prizes won by Turkishmanufacturing companies in the last few years. In 1998a consumer durable manufacturing company, BekoTicaret, a subsidiary of Turkish conglomerate Koc Group,and Netas Inc., a Turkish telecommunications company,won the small and medium enterprises (SME) and thelarge business sector European Quality Awards, respec-tively. In 2000, Eczacibasi Vitra, a major pharmaceuticalcompany, and Arcelik A.S., another Turkish durable con-sumer goods manufacturer, were both named as two ofthe “Large Business European Quality Prize” winners. Asthe prospect of European Union (EU) membershipimproves Turkish firms’ abilities to export their goodsto EU members, TQM programs, especially ISO 9000certification, has gained more significance for thecompetitiveness of the companies in Turkey.

In a study of 140 large Turkish manufacturing firms,Ozgur, Meek, and Toker (2002) found differences inusage and awareness of quality tools and conceptsbetween ISO 9000 certified and noncertified companies.ISO 9000 certified companies constituted 78 percent ofthe companies surveyed that used basic quality toolsmore than noncertified companies. A study of ISO 9000experiences of Turkish firms (Pinar, Guder, and Yucel2000) showed that: 1) improving organizational per-formance and product quality were the most importantmotives for seeking ISO 9000 certification; and 2) ISO9000 had the highest impact on management andhuman resource factors, followed by organizational effec-tiveness and customer factors. ISO 9000 certification hadthe least impact on financial market factors. Anotherstudy of ISO 9000 certified Turkish companies (Pinar andCrouch 2001) found that 38.1 percent of firms have“totally or somewhat achieved” their expectations from

ISO 9000 certification, while 61.9 percent said it was “tooearly to decide or have not achieved” their expectations.The results also showed a significant difference between“achievers and nonachievers” of ISO 9000 expectationswith respect to firm size and export volume, and whetherthe firm is classified as an exporter or a nonexporter,where the achievers outperformed the nonachievers in allperformance areas.

While ISO 9000 is arguably the most influential setof standards of its kind in the world, there is no com-pelling evidence that the standards are ultimately goodor bad (Uzumeri 1997). Studies have shown that ISO9000 companies expect that their quality systems willlead to improved product design, process design, prod-uct quality, public image, and supplier relationship(Ebrahimpour, Whithers, and Hikmet 1997). Simmonsand White (1999), however, claim that the specificempirical relationship between ISO 9000 registrationand business performance has not been established,especially in the long term, based on objective measures.They state that while there has been much publishedwork associated with ISO 9000, the empirical researchon this question is lacking systematic measurement andmultiorganizational comparisons.

Research Objectives The purpose of this article is to examine the longitudinaleffect of ISO 9000 certification on business performanceby comparing the stock performances of ISO 9000 certi-fied firms versus non-ISO 9000 certified firms traded onTurkey’s Istanbul Stock Exchange (ISE). It is believedthat the impact of ISO 9000 success should be reflectedin their stock prices and returns. The main premise ofthe study is that ISO 9000 certified firms are expected tooutperform the non-ISO 9000 firms in their respectedindustries and, ultimately, on the stock market. Thespecific objectives of the article are:

1. To provide information about the characteristics ofISO 9000 and non-ISO 9000 firms traded on the ISE

2. To analyze the impact of ISO 9000 certification onstock returns and stock return variability

3. To determine if the stock returns and the variabilityof returns of ISO 9000 certified firms are influencedby certifying organization (Turkish vs. non-Turkish)

www.asq.org 23

Pinar v14 i4 9/20/07 10:10 AM Page 23

24 QMJ VOL. 14, NO. 4/© 2007, ASQ

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

To accomplish these objectives, the study will use theaverage stock market returns and variances obtainedfrom monthly stock returns covering a time period fromJanuary 1997 through September 2005. The study alsoexamines the difference in certifying agency (Turkishvs. foreign). It is expected that the impact of ISO 9000success should be reflected in their stock price returnsand variances. In other words, the main premise of thestudy is that ISO 9000 firms should experience higherreturns and lower variances than non-ISO 9000 firms.Since this study uses historical stock returns of thecompanies, it is possible to examine the longitudinalimpact of ISO 9000 certification on business perform-ance over a 105-month time period from January 1997to September 2005.

BACKGROUND AND PRIORRESEARCH ON ISO 9000One of the primary means of providing quality verifi-cation is through ISO 9000 certification. ISO 9000certification is based on quality standards written in1987 by ISO and was adopted by the EC in 1989(Wilson 1999). The only requirement of the originalISO 9000 standards was that the organization shouldhave documentation and a verifiable process in placeto ensure that it consistently produced what it said itwould produce. Thus, a company could comply withthe standards and still produce a poor-quality product.As a result, these standards have been criticized in theliterature (Erel and Ghosh 1997; Jackson and Ashton1995; Reimann and Hertz 1996; Zuckerman 1997) foromitting clauses that address customer satisfaction,continuous improvement, and top management sup-port. In a response to the dissatisfaction with the oldstandards, these standards were revised in 1994 andagain in 2000. The most recent version is called theISO 9000:2000 family of standards (ISO 2006b). Thecurrent standards represent a fundamental change inapproach over the earlier versions by focusing on thefollowing elements: top management commitment,customer satisfaction, emphasis on processes, and con-tinual improvement. These ISO 9000 standards areclosely aligned with the philosophy and objectives ofmost quality programs such as TQM and the Malcolm

Baldrige National Quality Award. Flynn, Schroeder, andSakakibara (1995) demonstrate the importance ofTQM in implementing various quality programs. Theirpaper establishes that measures of quality performancewere related to competitive advantage. Many ISO 9000firms in Turkey got their certification believing that itwould enable them to gain a competitive advantage.Because of this, ISO 9000 appears to be related to TQMand other quality initiatives.

It is claimed that ISO 9001:2000 is now firmly estab-lished as the globally accepted standard for providingassurance of the quality of goods and services in supplier-customer relations (ISO 2006a). ISO 9000 is the world’slargest developer of voluntary international standardsfor business, government, and society. At the beginningof June 2006 its portfolio comprised more than 15,900standards that provide practical solutions and achievebenefits for almost every sector of economic activity andtechnology (ISO 2006a). The best-known standards areISO 9001:2000 for quality management systems, ISO14001 (both 1996 and 2004 versions) for environmentalmanagement systems, ISO/TS 16949:2002 for the auto-motive sector, and ISO 13485:2003 for medical devices(ISO 2005). The ISO 9000 2006 survey (ISO 2006a)reports the recent status of each of these standards.Table 1 presents the growth of the number of certifi-cates issued in different countries from December 2001to December 2005 for 608 ISO 9001:2000, ISO 14001,ISO/TS 16949:2000, and ISO 13485:2003. These num-bers in Table 1 signify the global importance of ISO9000 certification for each of the standards.

The revised ISO 9000:2000 series includes eightquality management principles as the basis of qualitymanagement system standards (ISO 2006b). Theseprinciples are derived from the collective experienceand knowledge of the international experts. They are:1) customer focus; 2) leadership; 3) involvement ofpeople; 4) process approach; 5) system approach; 6) continuous improvement; 7) factual approach todecision making; and 8) mutually beneficial supplierrelationships. Peyrat (2001) states that the primaryobjective of this new version of standards is customerorientation: the ISO 9000:2000 series focuses on satisfy-ing the customers of the organization, in the case ofISO 9001, and on satisfying all interested parties, in thecase of ISO 9004 and ISO 14001. With the publication

Pinar v14 i4 9/20/07 10:10 AM Page 24

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

of the 1994 versionsand after several yearsof additional experi-ence, Peyrat (2001)indicated that the ISO9000 standards gradu-ally acquired a reputa-tion of being fairlyeasily transportablemanagement systemstandards. In fact, theyhave been imple-mented by all types oforganizations, frommultinational corpora-tions to SMEs and fromhumanitarian NGOs( n o n g o v e r n m e n t -organizations) to pub-lic services. This makesthe 2000 version of theISO 9000 series trulygeneric (Peyrat 2001).

There have been several recent studies that investi-gated the relationship between ISO 9000 certificationand business performance. There are no agreementsamong results of the studies. Some support a positiverelationship between the two, while others indicate nodirect relationship. Haversjo (2000) reported that inDenmark, ISO 9000 certified companies seem to havehigher returns than non-ISO 9000 certified companies.He believes, however, that ISO 9000 registration did notcause the change in superior financial performance ofthe firm; rather, it could be attributed to other factorssuch as innovative management. The study by Brown,van der Wiele, and Loughton (1998) found that themost significant benefits of ISO 9000 certification arein the area of raising quality awareness and in settinga foundation to improve quality (internal factors).Gustafson et al. (2001) analyzed the implementation ofISO 9000 in small business enterprises in Sweden. Theirresults showed that the higher the level of involvement ofthe CEO and the employees, the higher the satisfactionlevel of the company with ISO 9000 certification andthe higher the continuation of quality improvementsafter obtaining it. Singels, Ruel, and van de Water

(2001), based on their results, concluded that ISO 9000certification serves organizations that are internallymotivated better than organizations that are externallymotivated (as a result of external pressures such asbuyer demands). While these papers indicate and discusspossible benefits of ISO 9000 certification, none shows adirect relationship between ISO 9000 certification andbottom-line performance.

The findings of several studies concerning a rela-tionship between ISO 9000 certification and businessperformance were contrary to those findings cited pre-viously. For example, Jeng (1998) found that at least30 percent of Taiwanese managers don’t believe thathaving ISO 9000 approval or pursuing certification isthe best way to improve organizational performance,and 70 percent of ISO 9000 certified Taiwanese compa-nies surveyed exhibited performance results similar tothose prior to certification (Jeng 1998). However, theresults of Jeng’s study showed that ISO 9000 implemen-tation works well for the entire organization in improv-ing overall performance. Terziovski, Samson, and Dow(1997) found that, for manufacturing companies inAustralia and New Zealand, ISO 9000 certification is

www.asq.org 25

Table 1 Distributions of ISO 9000 standards in the world.

ISO 9001:2000 Dec. 2001 Dec. 2002 Dec. 2003 Dec. 2004 Dec. 2005

World total 44,388 167,919 497,919 660,132 776,608

No. of countries & economies 97 133 149 154 161

ISO 14001:2000 Dec. 2001 Dec. 2002 Dec. 2003 Dec. 2004 Dec. 2005

World total 36,464 49,440 64,996 89,937 111,164

No. of countries & economies 112 116 113 127 138

ISO /TS 16949:2002 Dec. 2001 Dec. 2002 Dec. 2003 Dec. 2004 Dec. 2005

World total 0 0 0 10,019 17,047

No. of countries & economies 0 0 0 62 80

IOS 13485:2003 Dec. 2001 Dec. 2002 Dec. 2003 Dec. 2004 Dec. 2005

World total 0 0 0 2,403 5,065

No. of countries & economies 0 0 0 55 67

Total 80,852 217,359 562,915 762,491 909,884

ISO 9001:2000 for quality management systems, ISO 14001 (both 1996 and 2004 versions) for environmentalmanagement systems, ISO/TS 16949:2002 for the automotive sector, and ISO 13485:2003 for medical devicesSource: The ISO Survey—2005 ©

2007

, ASQ

Pinar v14 i4 9/20/07 10:10 AM Page 25

26 QMJ VOL. 14, NO. 4/© 2007, ASQ

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

not significantly related to organizational performancemeasures.

Despite the expected benefits of ISO 9000 certification,the standards have several criticisms (Erel and Ghosh1997; Jackson and Ashton 1995: Reimann and Hertz1996; Zuckerman 1997). Some of the ISO 9000 criticismsare: 1) the standards are static, inflexible, and cannot beadapted to certain situations; 2) they are not customerdriven/market oriented; 3) they mainly concentrate onproduction firms; and 4) concentration on ISO 9000takes the attention away from continuous improve-ment. Despite these concerns, adoption of ISO 9000 israpidly growing.

Quality Movements in Turkey TQM had to be sponsored by a number of nationalorganizations in Turkey. Data indicated that TurkStandard Enstitusu (TSE) (or Turkish StandardsInstitute), National Council for Quality Accreditation(KAMK), and quality association (KALDER) had signif-icant roles in convincing managers through trainingand publications that TQM was a good initiative and, ifimplemented, companies would experience rewardingbenefits. The mission statement of Turkish companiesreflected the commitment to TQM; however, on a globalbasis all companies must be able to demonstrate thatthey have the systems in place to provide a qualityproduct if they wish to be competitive. The demand forquality in products and services brought about thedevelopment of the ISO 9000 series of quality systemrequirements and made ISO certification an importantquality factor in the international marketplace. ISOquality certification has almost become a prerequisitefor doing business in theglobal arena. In Turkey onemotivation for quality camefrom Turkey’s desire tobecome a member of the EU.Companies that desire tomarket to countries in the EUare expected to be ISO 9000certified. This is especiallytrue for companies that aredoing or desire to do businessin the EU.

In Turkey, TSE has been approved to consider andcertify firms with ISO 9000. TSE is the only organizationin Turkey that can approve ISO 9000 certification.KALDER also promotes quality by offering training pro-grams and publishing a quarterly practice journal, butcannot offer any quality certification. In addition to TSE,there are a number of European certification agenciesoperating in Turkey that also issue ISO 9000 certification.It has been argued that the source of certification makesa difference in the reputation of the certified company,and it may affect them getting certain export contracts.A study performed by Erel and Gosh (1997) indicatesthat 57 percent of companies were certified by foreignagencies and approximately 30 percent had both foreigncertification and TSE certification.

The popularity of ISO 9000 certification as a partof a quality program among Turkish companies hasgrown in recent years. More companies in Turkey arelooking to become ISO 9000 certified. As of the end ofDecember 2005, there were 12,133 Turkish firms withsome form of ISO 9000 certification (ISO 2006a).Table 2 shows the distributions of ISO 9000 certifica-tion by type of certification (ISO 2006a). Based on thesurvey results, at the end of December 2005, 10,929Turkish firms had received ISO 9000:2000 certification,918 firms had received ISO 14001:2000 certification,265 firms had ISO/TS 16949:2002 certification, and 21firms had received ISO 13485:2003 certification.

It seems that only a small number of ISO certifiedTurkish firms, however, applied for the National QualityAward. Even though Turkish firms appear to be interestedin improving quality, as evidenced by a large numberof ISO 9000 certifications, only a few of them have trulyadopted TQM (Yildirim 1999). The duration of getting

Table 2 Distributions of ISO 9000 standards in Turkey.

Type of certification Dec. 2001 Dec. 2002 Dec. 2003 Dec. 2004 Dec. 2005

ISO 9001:2000 72 911 3,248 5,009 10,929

ISO 14001:2000 91 135 240 338 918

ISO/TS 16949:2002 0 0 0 98 265

ISO 13485:2003 0 0 0 12 21

Total 163 1,046 3,488 5,457 12,133

Source: The ISO Survey—2005 ©20

07, A

SQ

Pinar v14 i4 9/20/07 10:10 AM Page 26

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

to be ISO 9000 certified takes one to one-and-a-halfyears, and the certification time is affected among oth-ers by employee involvement, top management support,existence of relevant procedures, professional consultantparticipation, and effective communication within theorganization (Halls and Oztas 2002). Difficultiesencountered in ISO 9000 implementation in Turkeyinclude: changing the department’s traditional policiesand procedures, using statistical techniques/tools, andmotivating personnel (Calisir, Bayraktar, and Beskese2001). A 1996 study by Erel and Ghosh (1997), whichdocumented the state of ISO 9000 implementation inTurkey, showed that 68 percent of the 73 firms surveyedhad an ISO 9000 certification.

In a study by Calisir, Bayraktar, and Beskese (2001),the following factors were considered important in theimplementation of ISO 9000 in Turkey: 1) enhancingthe quality reputation, and 2) increasing market share.They point out that, using ISO 9000, the importance ofachieving the expected improvements included reducingthe defect rate, improving product/service quality,enhancing organization’s quality reputation, andimproving standardization. Moreover, ISO certificationimplies that a bona fide commitment to quality and toobtaining ISO 9000 certification can be: 1) a strategicmarketing tool; 2) a means for improving productivity;3) a strategy for gaining competitive advantage; and 4) away to enhance position in the global marketplace.

Impact of ISO 9000 Certificationon Business PerformanceDespite the expected benefits of ISO 9000 certification,only a few empirical studies have examined whetherISO 9000 certified companies actually perform betterthan non-ISO 9000 companies. Past studies havedetailed perceived relationships between quality andbusiness success in Deming (1982); Garvin (1987);Kannan et al. (1999); Lam (1996); Lofgren (1991); andPfau (1989). There are also a few studies that empiri-cally compared the performance of ISO 9000 certifiedfirms to non-ISO 9000 firms. Some of these studiesinclude LRQA (1996); Lee (2003); Rajan and Tamimi(2003); Simmons and White (1999); and Terziovski,Power, and Sohal (2003).

A survey by Lloyd’s Register Quality Assurance (LRQA1996) compared the financial and sales performances of asample of 222 ISO 9000 certified British companies withthe industry average. The survey found that ISO 9000registered companies: 1) were two or three times moreprofitable than their nonregistered competitors and sig-nificantly outperformed the industry average on fiveother key financial and sales measures of profit margin,return on capital employed, sales per employee, capitalemployed per employee, and asset turnover; 2) experi-enced improved efficiency and productivity, reduced wasteand costs, and had improved market share; and 3) werefinancially stronger than their noncertified competitors.

Simmons and White (1999) conducted a systematic,multiorganizational analysis of the relationship betweenISO 9000 registration and objective measures of businessperformance, using the ratio sales/stockholder equity,profitability (ROA), and foreign sales (as a percentage oftotal sales) as indicators of business outcome. Based on126 U.S. companies (63 ISO 9000 and 63 non-ISO 9000firms) with cross-sectional data from seven SIC codedindustries, Simmons and White (1999) found that ISO9000 companies had significantly higher average prof-itability than non-ISO 9000 certified companies, but ISO9000 certified companies did not have higher opera-tional performance or higher foreign sales than non-ISO9000 companies. Their findings did not support theclaim that ISO 9000 companies realize advantages inoperational performance and foreign sales over non-ISO9000 companies; thus, they did not offer any evidencethat ISO 9000 may ultimately be good for organizations.Given that their sample data only covered 1995, they rec-ommend longitudinal analysis to better examine thelong-term impact of ISO 9000 on business performance.

Lee (2003), based on a survey of ISO 9000 and non-ISO 9000 firms in the Korean electronics industry,found that manufacturing strategies differed signifi-cantly between the two groups, and these differencessignificantly impacted firms’ business practices andperformance. Specifically, his results showed that theISO 9000 certified firms are more likely to do businessglobally and be affiliated with foreign firms, tend to bemore quality-conscious and more concerned aboutproduct identity, are more likely to have just-in-time(JIT) or Kanban-based production planning and controlsystems, tend to have lower material cost with less labor

www.asq.org 27

Pinar v14 i4 9/20/07 10:10 AM Page 27

28 QMJ VOL. 14, NO. 4/© 2007, ASQ

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

intensive operation, are likely to have much less safetymargin of inventory, and tend to put more priority onresearch and development and the introduction of newproducts with more focus on quality.

In investigating the link between attaining ISO 9000certification and the benefits to the shareholders, Rajanand Tamimi (2003) compared the performance of aportfolio consisting of the stocks of ISO 9000 certifiedcompanies to the performance of the S&P 500 index fordifferent investment holding periods. They found thatwhile the ISO 9000 portfolio appears to be considerablymore volatile than the S&P, it has significantly outper-formed the market index. Also, Terziovski et al. (2003)showed a significant and positive relationship betweenthe manager’s motives for adopting ISO 9000 certifica-tion and business performance.

In order to determine whether ISO 9000 certificationpays off, Corbett et al. (2002) used “event study” to com-pare the financial performance of a group of companiesthat have gone through ISO 9000 certification with thefinancial performance of a comparable group of compa-nies that did not have ISO 9000 certification. The authorsof the paper claimed that superior performance of ISO9000 firms may not be related to ISO 9000 certification,but may simply reflect that these firms already had supe-rior performance prior to their certification. Their resultsshowed that the companies demonstrated better perform-ance after deciding to seek ISO 9000 certification thannon-ISO firms from the same industry with similar per-formance prior to certification. Corbett et al. (2002) statethat the decision to get certified led to superior cost con-trol and higher sales; therefore, one could argue that thedecision to seek ISO 9000 certification is positively relatedto other good management practices that result inimprovements in sales and ROA, instead of certificationitself. Also, since the control group consists of companieswith the similar ROA prior to certification, this indicatesthat certification is helpful, where the companies withISO 9000 certification experienced significant improve-ments in their ROAs. Their results suggest that companiesthat are not seeking ISO 9000 certification have graduallydecreasing performance in terms of their ROA and sales.

Some studies alluded that ISO 9000 improved thesmoothness of the business operations, which mightimply that ISO 9000 certification could lower the vari-ability of the returns. The prior studies, however, did not

empirically examine the impact of ISO 9000 certificationon the variability (or volatility) of business performance.As variability of returns represents a measure of risk, it isimportant to examine how ISO 9000 impacts the risk (orvariability) of the returns. Brigham and Houston (2007)claim that risk is a difficult concept to grasp. They indi-cate that the tighter the probability distribution ofexpected future returns, the lower the variability of thereturns, thus the smaller the risk. Therefore, it is custom-ary to assess risk by using standard deviation (sigma).The lower the value of sigma, the lower the risk(Brigham and Houston (2007: 251). Following the samelogic, the lower the variability (variance) of the returns,the lower the risk of those returns.

Many of the ISO 9000 studies used surveys to examinethe performances of ISO 9000 firms based on financialand sales measurements. The literature on ISO 9000 isoverwhelmingly either prescriptive or descriptive, withstudies that are speculative, anecdotal, and often basedon the observation of a single company (Ebrahimpour,Whithers, and Hikmet 1997). In addition, Simmons andWhite (1999) also claim that the specific empirical rela-tionship between ISO 9000 registration and businessperformance has not been established. They state thatwhile there has been much published work associatedwith ISO 9000, the empirical research on this questionis lacking systematic measurement and multiorganiza-tional comparisons. Unlike most of the other studieswith self-reported (subjective) performance measuresinvolving ISO 9000 certification or cross-sectional stud-ies, this paper used objective measures of monthly stockreturns that include 105 months or almost nine years ofdata. This is a significantly longer time than the studyby Pinar, Guder, and Yucel (2000), which included onlyfour years of data.

Research Hypotheses As stated before, the main goal of this study is to investi-gate the performance of ISO 9000 and non-ISO 9000firms traded on Turkey’s ISE market of Turkey by usingtheir average stock market returns and the variability ofthe stock returns from January 1997 through September2005. Both theoretical research (Fine 1986; Lederer andRhee 1995) and empirical studies (Chowdhury andMenon 1995; Hendricks and Singhal 1997; Jacobson

Pinar v14 i4 9/20/07 10:10 AM Page 28

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

and Aaker 1987; Powell 1995) have established linksbetween quality and business performance, which isconsistent with the conceptual and empirical founda-tions provided by the authorities on quality (that is,Deming 1986; Juran 1974; Ishikawa 1985). There havenot been enough studies on the effectiveness of the ISO9000 standard on business performance. Based on theclaims of expected higher business performance,increased profitability, and marketing and global com-petitive advantages of ISO 9000 certified companiespresented in the previous sections, the followingresearch hypotheses were developed:

• H01: ISO 9000 certified firms don’t have higher stockreturns than non-ISO 9000 firms.

• Ha1: ISO 9000 certified firms have higher stockreturns than non-ISO 9000 firms.

• H02: ISO 9000 firms don’t have lower stock variabilitythan non-ISO 9000 firms.

• Ha2: ISO 9000 firms have lower stock variability thannon-ISO 9000 firms.

• H03: ISO 9000 firms certified by non-Turkish certifyingorganization’s returns do not differ from returns ofISO 9000 firms certified by TSE.

• Ha3: ISO 9000 firms certified by non-Turkish certifyingorganizations have higher returns than ISO 9000firms certified by TSE.

• H04: ISO 9000 firms certified by non-Turkish certify-ing organizations’ variability of returns do not differfrom variability of returns of ISO 9000 firms certifiedby TSE.

• Ha4: ISO 9000 firms certified by non-Turkish certifyingorganizations have lower variability than ISO 9000firms certified by TSE.

Expressed in mathematical notation the aforemen-tioned hypotheses are as follows:

H01: α1 <– α2 = 0 where 1 = ISO certified and 2 = non-ISO,…,n

Ha1: α1 > α2 ≠ 0 where 1 = ISO certified and 2 = non-ISO,…,n

H02: σ12 >– σ2

2 = 0 where 1 = ISO certified and 2 = non-ISO,…,n

Ha2: σ12 < σ2

2 = 0 where 1 = ISO certified and 2 = non-ISO,…,n

H03: γ3 <– γ4 = 0 where 3 = foreign certified and4 = TSE certified

Ha31: γ3 > γ4 = 0 where 3 = foreign certified and4 = TSE certified

H04: σ32 >– σ4

2 = 0 where 3 = foreign certified and4 = TSE certified

Ha4: σ32 < σ4

2 = 0 where 3 = foreign certified and4 = TSE certified

DATA AND METHODOLOGYFor this study, the authors used the monthly stock returns(the percent change in their prices from one month tothe next month) of all the manufacturing firms (with orwithout ISO 9000 certification) listed on the ISE marketof Turkey from January of 1997 to September of 2005 (afull year of data was not available for 2005). This studyexcludes the service firms because of their differentincome statement structures. The monthly stock returnswere obtained from the ISE (ISE 2006). Previous studiesused monthly return data to examine the ISE of Turkey(that is, Akdeniz, Altay-Salih, and Aydogan 2000). A listof all ISO 9000 certified companies traded on the ISEwere obtained from KALDER (KALDER 2006). The listincluded the names of the companies, their ISO 9000certification dates, the certifying organization, and theprimary industries they were operating in.

The stock market and ISO 9000 certification arerecent phenomena in Turkey. As a part of the economicstabilization and liberalization programs in the early1980s, Turkey’s ISE market was established in 1986.According to ISE (2001), there were only 80 companieslisted at the beginning with market capitalization of$938 million. The number of firms has steadilyincreased and reached 387 in 2001 (ISE 2001), butbecause of recent economic crises, the number of firmslisted on ISE declined to 319 in late 2005 (ISE 2006).Market capitalization has also experienced dramaticgrowth over the years, reaching more than $93.1 billionat the end of 2001 (ISE 2001) and $201.8 billion at theend of 2005 (ISE 2006). ISE officials classified all thefirms on the stock market into 17 distinct industrygroups (ISE 2001; 2006).

Turkish companies began receiving ISO 9000 cer-tification in 1992. Throughout 1992, there were six

www.asq.org 29

Pinar v14 i4 9/20/07 10:10 AM Page 29

30 QMJ VOL. 14, NO. 4/© 2007, ASQ

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

companies on the ISE that were ISO 9000 certified. At thebeginning of the study period, there were about 220 firmsthat traded on the ISE market (ISE 2001). One hundredand three firms had ISO 9000 certification and 117 ofthem did not have ISO 9000 certification. Sixty percent ofthe firms on the ISE belong to seven industry groups asclassified by the ISE board (food and beverage, textile andleather, wood, paper and printing, chemical, petroleumand plastics, non-metal minerals, basic metals, andmachinery and metal products) and 40 percent belong toother industry groups. The majority (87.4 percent) of the103 ISO 9000 certified firms on the ISE belongs to theseven industry groups (manufacturing industries) andthe remaining 12.6 percent belong to other industrygroups. It is just the opposite for non-ISO 9000 firms,where only 35.9 percent belong to the seven industrygroups and 64.1 percent belong to other groups. Of theISO 9000 certified companies, 62.1 percent receivedtheir ISO 9000 certification from the TSE, the govern-ment quality-certifying agency, and 37.8 percent ofthem from certifying agencies from other countries.

Selection of Sample andResearch DesignThe objectives of the study and time period covered setthe criteria for the firms to be included in the study. Thetwo criteria for ISO 9000 certified firms to be included inthe analysis were that they should: 1) have received theirISO 9000 certification before January 1997; and 2) havebeen listed on the ISE during the entire study period.For non-ISO 9000 firms, the requirement was that theywere traded on the ISE for the entire time period. Thiswas important to examine the longitudinal impact ofISO 9000 certification on business performance. Inaddition, the selected companies operate in seven man-ufacturing industries, as was stated earlier. Companiesfrom other industries were not included in the study, toprevent any potential inconsistency in the data.

Given the aforementioned criteria and requirements,a total of 95 firms were qualified for the study. These werethe only firms that were traded on ISE for the entire timeperiod of nine years. In fact, the authors do not have asample, but the entire population. Later, more firmsstarted being traded on the ISE, but the authors could not

include them because of the requirement of keeping thesame firms on their panel. Of these companies, 62 wereISO 9000 certified and 33 were non-ISO 9000 firms thatmet the two requirements to be included in the study foranalysis. While these requirements limited the number offirms in the study, it made the results and interpretationsmore consistent for both ISO 9000 and non-ISO 9000firms. Given that this was a longitudinal study, it wasimportant to keep the same companies over the studyperiod so that the results would not be distorted with theaddition and/or deletion of new companies. This processof studying the same companies over time is similar toconsumer panel studies, where a group of same con-sumers is observed over time to examine various pur-chase behaviors and activities (Churchill and Iacobucci2005; Hair, Bush, and Ortinua 2006). Also, it is desirableto have the same number of ISO 9000 and non-ISO com-panies from the same industries; however, this was notpossible because of the limited number of companiestraded on the ISE and the authors’ selection criteria.

The study examined the yearly returns, various com-binations of two-, three-, and four-year returns, andreturns over the nine-year period. The yearly returnsallowed the authors to compare the stock performance ofISO 9000 vs. non-ISO 9000 firms on a year-by-year basis.They also used the two-, three-, and four-year averages tosmooth out the effects of yearly fluctuations or changesin returns of ISO 9000 or non-ISO 9000. For example,1999 and 2000 seem to be unique years for stockreturns. By having averages for different time periods(that is, two-, three-, four-year), the authors were ableto smooth out these extreme situations of 1999 or 2000and eliminate these random changes. This is similar tothe time-series procedure recommended for economicdata by Makridakis, Wheelwright, and McGee (1978).Therefore, the averages of the stock returns for differenttime periods were essential to capture the long-termeffects of ISO 9000 certification vs. non-ISO 9000 certifi-cation. The authors also obtained the variances of ISO9000 and non-ISO 9000 firms for the yearly returns aswell as other time periods used for means.

The aforementioned mean and variance values wereused to compare the performances of ISO and non-ISO9000 firms with the two-sample t-test procedure.Specifically, the authors compare means and variances ofISO 9000 certified and non-ISO 9000 certified companies

Pinar v14 i4 9/20/07 10:10 AM Page 30

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

for yearly data as well as for other time periods. The two-sample t-test procedure allows them to determine if themeans and variances of ISO 9000 and non-ISO 9000firms were significantly different. Even though p < .05and p < .01 are most commonly used for testing sig-nificance, the studies have also used p < .10 in theirstatistical significance testing (Jain and Tabak 2002).Thus, they consider p-values less than .10 (p < .10) assignificant for their mean comparisons.

Selected Characteristics ofthe FirmsTable 3 presents a distribution of the firms by industry.The distributions of ISO 9000 firms by industry showthat five firms were in the food and beverage industry;seven firms were in the textile and leather industry; sixfirms were in wood, paper, and printing; 13 firms werein chemical, petroleum, and plastics; 14 firms were innon-metal minerals; six were in basic metals; and 11firms were in machinery and metal products. The distri-butions of non-ISO certified firms were six firms in thefood and beverage industry; 13 firms in textile andleather industry; three firms in wood, paper, and print-ing; five firms in chemical, petroleum, and plastics; onein nonmetal minerals; one in basic metals; and four inmachinery and metal products. A large number of firmswith ISO 9000 certification in certain industries indicate

that they are more quality oriented than the firms inother industries, and certain industries seem to be morereceptive than others to quality improvements.

RESULTS

Comparison of ISO 9000 vs.Non-ISO 9000 FirmsThe main purpose of this study was to investigate theimpact of ISO 9000 certification on stock returns andstock variability. Based on this intent, two hypotheseswere developed.

• H01: ISO 9000 certified firms don’t have higher stockreturns than non-ISO 9000 firms vs.

• Ha1: ISO 9000 certified firms have higher stockreturns than non-ISO 9000 firms and,

• H02: ISO 9000 firms don’t have lower stock variabilitythan non-ISO 9000 firms vs.

• Ha2: ISO 9000 firms have lower stock variability thannon-ISO 9000 firms.

In order to test these two hypotheses, the authorsused two-sample independent sample t-test to comparethe means and variances of returns from January 1997to September 2005 for ISO 9000 certified and non-ISO9000 certified firms. As shown in Table 4, these compar-isons covered: 1) yearly returns for nine years; 2) thetwo-year moving average returns; 3) three-year andfour-year simple average returns; 4) nine-year averagereturn. All of these resulted in a total of 23 time-periodscenarios. These combinations provided an opportunityto compare the performances of ISO 9000 certified firmvs. those of non-ISO 9000 certified firms over multipletime periods. As stated before, having different timeperiod averages, it was possible to smooth out the effectsof fluctuations and changes in any year in order to betteridentify the impact of ISO 9000 certification.

The comparisons of the mean returns for both ISO9000 vs. non-ISO 9000 firms in Table 4 and Figure 1show that ISO 9000 firms had significantly higherreturns than non-ISO 9000 firms for the years of 1999(means of 16.41 percent for ISO 9000 vs.13.87 percent fornon-ISO 9000, p < .05) and 2000 (means of .05 percentfor ISO 9000 vs. -2.05 percent for non-ISO 9000, p < .05).

www.asq.org 31

Table 3 Distribution of the firms by industries.

ISO 9000 Non-ISO 9000n Percent n Percent

Food and beverage 5 8.06 6 18.18

Textile and leather 7 11.29 13 39.39

Wood, paper, and printing 6 9.68 3 9.09

Chemical, petroleum, and plastics 13 20.97 5 15.15

Non-metal minerals 14 22.58 1 3.03

Basic metals 6 9.68 1 3.03

Machinery and metal products 11 17.74 4 12.12

Total 62 100.00 33 100.00 ©20

07, A

SQ

Pinar v14 i4 9/20/07 10:10 AM Page 31

32 QMJ VOL. 14, NO. 4/© 2007, ASQ

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

The mean annual returns forother years are not statisti-cally significant. This indi-cates that both ISO 9000 andnon-ISO 9000 firms seem tohave similar performances.During the nine-year period,the increase and decrease ofthe annual returns for ISO9000 and non-ISO 9000firms appear to follow simi-lar patterns. In 1999 bothISO 9000 and non-ISO90000 firms had very goodreturns, but ISO 9000 firmshad significantly higherreturns than non-ISO 9000firms. The year 2000 was theworst year for non-ISO 9000firms (mean of -2.04 percentreturns) and not a very goodyear for ISO 9000 certifiedfirms either (mean of .05percent return), but ISO9000 firms still had a signifi-cantly higher average returnthan non-ISO 9000 firms.

Similar comparisons ofISO 9000 versus non-ISO9000 were conducted forother time period combina-tions. As presented in Table 4and Figure 1, the mean com-parisons for two-year movingaverage returns show thatthe mean returns for ISO9000 are significantly higher than those for non-ISO9000 firms for the time periods of 1998 to 1999 (meanof 8.40 percent for ISO 9000 vs. 7.13 percent for non-ISO 9000, p < .10) and 1999 to 2000 (8.23 percent forISO 9000 vs. 6.25 percent for non-ISO 9000, p < .05).The study found no significant difference between thereturns for ISO 9000 and non-ISO 9000 firms duringother two-year periods. Also, the mean comparisons ofthree-year and four-year simple mean returns show thatISO 9000 firms had significantly higher returns than

non-ISO 900 firms for 1997-1999 period (mean of8.78 percent for ISO 9000 vs. 7.62 percent for non-ISO9000, p < .10) and for 1997-2000 period (mean of.6.60 percent for ISO 9000 vs. 5.37 percent fro non-ISO9000, p < .05). During other time periods, there wasno significant difference between the returns for ISO9000 and non-ISO 9000 firms. Finally, the study com-pared the mean returns over a nine-year time periodfrom 1997 to 2005 and found that ISO 9000 firms(mean of 5.16 percent) had significantly higher

Table 4 Comparisons of means and variability of the returns for ISO 9000vs. non-ISO firms on the Istanbul Stock Exchange, 1997-2005.

Means VariancesYears ISO 9000 Non-ISO 9000 p-sign. ISO 9000 Non-ISO 9000 p-sign.

Avg 97 9.56 8.58 0.389 18.38 45.84 0.108

Avg 98 0.38 0.40 0.985 24.05 21.62 0.691

Avg 99 16.41 13.87 0.015 21.84 21.62 0.879

Avg 00 0.05 -2.05 0.018 16.14 16.42 0.840

Avg 01 7.27 8.15 0.546 14.90 60.40 0.006

Avg 02 0.86 1.58 0.390 10.49 23.71 0.027

Avg 03 4.54 4.89 0.635 10.84 13.64 0.588

Avg 04 3.89 3.39 0.678 52.70 18.92 0.001

Avg 05 2.87 1.31 0.101 18.30 20.28 0.969

97-98 4.97 4.49 0.524 9.77 16.77 0.287

98-99 8.40 7.13 0.059 6.81 14.20 0.287

99-00 8.23 6.25 0.002 8.10 8.30 0.625

00-01 3.66 3.39 0.750 5.25 21.15 0.000

01-02 4.07 4.86 0.213 5.86 14.20 0.116

02-03 2.70 3.24 0.385 4.32 10.14 0.079

03-04 4.21 4.14 0.931 18.64 7.79 0.636

04-05 3.45 2.50 0.282 21.01 8.30 0.584

97-99 8.78 7.62 0.058 6.50 10.74 0.239

00-02 2.73 2.79 0.921 3.27 10.09 0.003

03-05 3.85 3.37 0.465 11.53 4.44 0.356

97-00 6.60 5.37 0.011 2.93 5.65 0.033

01-04 4.14 4.50 0.460 4.74 5.97 0.297

97-05 5.16 4.63 0.082 1.91 1.95 0.567 ©20

07, A

SQ

Pinar v14 i4 9/20/07 10:10 AM Page 32

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

returns than non-ISO 9000 firms (mean of 4.63 percent)at p < .10 level (see Table 4 and Figure 1).

These mean comparisons indicate that ISO 9000firms did not outperform non-ISO firms for all timeperiods or for all scenarios. However, the study foundthat during two years (1999 and 2000) ISO 9000 firmshad significantly higher returns than non-ISO 9000firms. Some of these findings support H01, and otherssupport Ha1. The results, based on various time periodcombinations, suggest that the returns for both ISO9000 and non-ISO 9000 firms seem to follow fairly sim-ilar patterns. The nine-year average stock returns showthat in the long run ISO 9000 firms appear to performbetter than non-ISO 9000 firms, which supports Ha1.

Comparisons of Variances andCumulative Means for ISO 9000vs. non-ISO 9000 firmsThe prior studies did not consider the impact of ISO 9000certification on the variability (or volatility) of businessperformance. As variability of returns represents a meas-ure of risk (Brigham and Houston 2007), it is importantto examine how ISO 9000 impacts the risk (or variabil-ity) of the returns. As indicated before, some studiesalluded that ISO 9000 improved the smoothness of thebusiness operations, which might imply that ISO 9000certification could lower the variability (or volatility) of

the returns. To explore this issue, a part of the secondobjective was to examine the impact of ISO 9000 certifi-cation on stock return variability (or volatility). This ledto the hypothesis that H02: ISO 9000 firms don’t havelower stock variability than non-ISO 9000 firms vs. Ha2:ISO 9000 firms have lower stock variability than non-ISO9000 firms. To test this hypothesis, the variances of ISO9000 and non-ISO 9000 firms were compared over differ-ent time-period scenarios, and the results are presented inTable 4 and Figure 2. The comparisons of the variancesfor annual returns covering from 1997 to 2005 show thatISO 9000 firms had significantly lower variances thannon-ISO 9000 firms for the years 2001 (p < .01), 2002 (p < .05), and marginally for 1997 (p = .108). However,non-ISO 9000 firms had a significantly lower variancethan ISO 9000 firms in 2004 (p < .01). The study foundno significant differences between the return variances ofISO 9000 and non-ISO 9000 firms for other years. Someof these results support H02, while others support Ha2.

Again, the similar comparisons were conducted fortwo-year, three-year, four-year, and nine-year averagereturns. The variances of two-year moving averagereturns for ISO 9000 and non-ISO 9000 did not seemto follow any predictable pattern, and neither grouphad consistently lower or higher variances. The com-parisons of the variances showed that ISO 9000 firmshad a significantly lower variance than non-ISO 9000only during two periods (2000-2001, p < .01 and 2002-2003, p < .10). For the three-year and four-year periods,

www.asq.org 33

17.00

12.00

7.00

2.00

-3.00

ISO 9.56 0.38 16.41 0.05 7.27 0.86 4.54 3.89 2.87 4.97 8.40 8.23 3.66 4.07 2.70 4.21 3.45 8.78 2.73 3.85 6.60 4.14 5.16

Non-ISO

8.58 0.40 13.87 -2.05 8.15 1.58 4.89 3.39 1.31 4.49 7.13 6.25 3.39 4.86 3.24 4.14 2.50 7.62 2.79 3.37 5.37 4.50 4.63

Avg97

Avg98

Avg99**

Avg00**

Avg01

Avg02

Avg03

Avg04

Avg05

97-98

98-99*

99-00**

00-01

01-02

02-03

03-04

04-05

97-99*

00-02

03-05

97-00**

01-04

97-05*

p-sign.: *p < .10, **p < .05, ***p < .01

Figure 1 Mean Comparisons of ISO 9000 vs. Non-ISO 9000 Returns on the Istanbul Stock Exchange, 1997-2005.

©20

07, A

SQ

Pinar v14 i4 9/20/07 10:10 AM Page 33

34 QMJ VOL. 14, NO. 4/© 2007, ASQ

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

ISO 9000 firms had signifi-cantly lower variances during2000-2003 (p < .01) and1997-2000 (p < .05). Finally,comparison of the variancesof ISO 9000 and non-ISO9000 over a nine-year periodfound no significant differ-ence between the two groups.These findings indicate thatISO 9000 certification doesnot seem to lower the vari-ability of the stock returnssignificantly for all time peri-ods and scenarios. Similar toyearly returns, some of theseresults support the hypothesisH02, while others support Ha2.

In addition to yearly return comparisons, the studycompared the mean and variability of returns for ISO9000 vs. non-ISO 9000 over time by combining thereturns for successive years. This allows one to examinethe cumulative impact of ISO 9000 on business per-formance (in this case stock returns) over time. Toaccomplish this, the average returns for successive yearsof one, two, three, and so on were calculated, whichproduced nine scenarios consisting of one-year, two-year, three-year, four-year, five-year, six-year, seven-year,eight-year, and nine-year mean returns. These mean

and variance comparisons are presented in Table 5.The unique feature of these returns is that they make itpossible to measure the cumulative effect of ISO 9000certification over time, rather than year by year. Thecomparisons of mean returns in Figure 3 (also inTable 5) indicate that ISO 9000 companies consis-tently had higher stock returns than non-ISO 9000firms during nine scenarios. However, they were signifi-cantly higher only for 1997-1999 (p < .10), 1997-2000(p < .01), 1997-2001 (p < .01), 1997-2002 (p < .10),and 1997-2005) (p < .10). These findings support H01

18.38 24.05 21.84 16.14 14.90 10.49 10.84 52.70 18.30 9.77 6.81 8.10 5.25 5.86 4.32 18.64 21.01 6.50 3.27 11.53 2.93 4.74 1.91

45.84 21.62 23.85 16.42 60.40 23.71 13.64 18.92 20.28 16.77 14.20 8.30 21.15 14.20 10.14 7.79 8.30 10.74 10.09 4.44 5.65 5.97 1.95

60.00

50.00

40.00

30.00

20.00

10.00

0.00

ISO

Non-ISO

Avg97

Avg98

Avg99

Avg00

Avg01***

Avg02**

Avg03

Avg04***

Avg05

97-98

98-99

99-00

00-01***

01-02

02-03*

03-04

04-05

97-99

00-02***

03-05

97-00**

01-04

97-05

p-sign.: *p < .10, **p < .05, ***p < .01

Figure 2 Variance Comparisons of ISO 9000 vs. Non-ISO 9000 Firms on the Istanbul Stock Exchange, 1997-2005.

©20

07, A

SQ

Table 5 Comparisons of means and variability of the returns for ISO 9000vs. non-ISO firms on the Istanbul Stock Exchange.

Means VariancesYears ISO 9000 Non-ISO 9000 p-sign. ISO 9000 Non-ISO 9000 p-sign.

Avg 97 9.56 8.58 0.389 18.38 45.84 0.108

97-98 4.97 4.49 0.524 9.77 16.77 0.287

97-99 8.78 7.62 0.058 6.50 10.74 0.239

97-00 6.60 5.37 0.011 2.93 5.65 0.033

97-01 6.73 5.37 0.004 1.88 5.65 0.000

97-02 5.76 5.20 0.088 1.70 3.26 0.029

97-03 5.58 5.16 0.197 1.37 2.75 0.034

97-04 5.37 4.94 0.165 1.85 2.36 0.216

97-05 5.16 4.63 0.079 1.91 1.95 0.578 ©20

07, A

SQ

Pinar v14 i4 9/20/07 10:10 AM Page 34

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

for these time periods, and the findings for other peri-ods support Ha1. As stated previously, after nine yearsISO 9000 certified firms still have a significantlyhigher return than non-ISO 9000 certified firms. Thiscould be an indication of the long-term effect of ISO9000 certification on business performance.

Also, the study examined the cumulative impact ofISO 9000 certification on the variability of the stockreturns. Figure 4 (also Table 5) shows the comparisonsof the variability of the stock returns for ISO 9000 vs.non-ISO 9000 firms. The results show that ISO 9000firms had lower variances than non-ISO 9000 firms foreach of the nine scenarios over the nine-year period.However, the differences were significantly lower in1997-2000 (p < .05), 1997-2001 (p < .01), 1997-2002(p < .05), and 1997-2003 (p < .05), which support H02.

The variances were not significant during other periods.These findings support Ha2. One interesting thing is thatthe variances of ISO 9000 and non-ISO 9000 seemed toconverge over time, and after nine years their varianceswere almost identical. This may suggest that the impactof ISO 9000 on the variability of business performancemight disappear over time.

The Effects of ISO 9000Certifying Agency The final objective of the study was to examine whetherthe stock returns of ISO 9000 certified firms were influ-enced by the certifying organization. As there seems to bea preference among the Western European countries for

www.asq.org 35

18.38 9.77 6.50 2.93 1.88 1.70 1.37 1.85 1.9145.84 16.77 10.74 5.65 5.65 3.26 2.75 2.36 1.95

Avg 97 97-98 97-99 97-00** 97-01*** 97-02** 97-03** 97-04 97-05

50.00

40.00

30.00

20.00

10.00

0.00

ISONon-ISO

p-sign.: *p < .10, **p < .05, ***p < .01

Figure 4 Comparisons of Variability of Returns Over Time for ISO vs. Non-ISO 9000 firms on the IstanbulStock Exchange, 1997-2005.

©20

07, A

SQ

10.009.008.007.006.005.004.003.002.001.000.00

ISO 9.56 4.97 8.78 6.60 6.73 5.76 5.58 5.37 5.16Non-ISO 8.58 4.49 7.62 5.37 5.37 5.20 5.16 4.94 4.63

Avg 97 97-98 97-99* 97-00*** 97-01*** 97-02* 97-03 97-04 97-05*

p-sign.: *p < .10, **p < .05, ***p < .01

Figure 3 Comparisons of Mean Returns Over Time for ISO vs. Non-ISO 9000 firms on the Istanbul StockExchange, 1997-2005.

©20

07, A

SQ

Pinar v14 i4 9/20/07 10:10 AM Page 35

36 QMJ VOL. 14, NO. 4/© 2007, ASQ

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

the European certifying registrars (Barnes 1998),the study examined whether such a preference isreflected by the stock returns of the ISO 9000firms certified by foreign non-TSE vs. TSE agen-cies. Based on this notion, the authors developedhypothesis H03: The returns of ISO 9000 firmscertified by non-TSE certifying agencies are notsignificantly higher than those returns of ISO9000 firms certified by TSE vs. Ha3: The returnsof ISO 9000 firms certified by non-TSE certify-ing agencies are significantly higher than thosereturns of ISO 9000 firms certified by TSE. Thishypothesis was tested with data coveringJanuary 1997 to September 2005, again using:1) yearly returns for nine years; 2) the two-yearmoving average returns; 3) three-year and four-year simple average returns; and 4) nine-yearaverage return. The comparisons of means fordifferent time periods in Table 6 found that,with the exception of 2005, there were no signif-icant differences between returns of TSE andnon-TSE certified firms. These results indicatethat hypothesis H03 cannot be rejected in all sce-narios but one. The results for 2005 show that,in contrast to the prediction of Ha3, TSE certifiedfirms had higher returns (3.64 percent) thannon-TSE certified firms (.81 percent), which issignificant at p < .05 level. These findings donot support the implied preferences for theEuropean certifying agencies.

Concerning the variances, the authors testedthe following hypothesis H04: The variances ofthe returns for ISO 9000 firms certified by non-TSE agencies are not significantly differentfrom the variances of returns for ISO 9000firms certified by TSE vs. Ha4: The variances ofthe returns for ISO 9000 firms certified by non-TSE agencies are significantly higher than the variancesof returns for ISO 9000 firms certified by TSE. Theanalysis for variability of the returns in Table 6 showedthat the variances of TSE and non-TSE certified firmswere not significantly different in all time periods buttwo years. During the year of 1997 and the time periodof 2002-2003, non-TSE firms had significantly lowervariances than those of TSE certified firms (p < .10).

Based on these results, hypothesis H04 can be rejected forall scenarios, except two. This indicates that the firmscertified by non-TSE certifying agencies had a lowervariability in their returns. However, these results fromtwo scenarios do not provide strong evidence for rejectingH04. In fact, the comparisons over longer time periodssuggest that the certifying agency does not have anyimpact on the variability of the stock returns.

Table 6 Comparisons of means and variability of returnsfor Turkish vs. foreign ISO 9000 certified firmson the Istanbul Stock Exchange, 1997-2005.

Means VariancesYears Turkish Foreign p-sign. Turkish Foreign p-sign.

Avg 97 9.08 9.74 0.646 28.89 14.85 0.075

Avg 98 1.71 -0.12 0.193 40.17 17.79 0.124

Avg 99 16.03 16.55 0.700 21.06 22.55 0.251

Avg 00 -0.48 0.25 0.524 12.47 17.69 0.677

Avg 01 7.61 7.14 0.677 20.81 13.03 0.128

Avg 02 0.34 1.06 0.442 11.77 10.11 0.579

Avg 03 5.32 4.24 0.252 15.58 9.04 0.427

Avg 04 3.89 3.89 0.999 16.53 67.05 0.930

Avg 05 3.64 0.81 0.019 12.83 18.45 0.129

97-98 5.40 4.81 0.516 11.48 9.28 0.745

98-99 8.87 8.22 0.382 5.59 7.29 0.526

99-00 7.77 8.40 0.443 10.19 7.42 0.914

00-01 3.56 3.70 0.837 4.94 5.48 0.695

01-02 3.97 4.10 0.855 5.84 6.00 0.952

02-03 2.83 2.65 0.796 7.01 3.43 0.056

03-04 4.60 4.06 0.664 10.40 21.98 0.860

04-05 2.57 3.78 0.356 4.21 27.19 0.264

97-99 8.94 8.73 0.769 5.38 7.04 0.592

00-02 2.49 2.82 0.526 2.34 3.66 0.356

03-05 3.57 3.95 0.697 5.66 13.89 0.617

97-00 6.58 6.61 0.963 3.61 2.75 0.376

01-04 4.29 4.08 0.741 2.17 5.77 0.775

97-05 5.04 5.20 0.689 1.95 1.93 0.501 ©20

07, A

SQ

Pinar v14 i4 9/20/07 10:10 AM Page 36

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

CONCLUSIONS AND LIMITATIONSIn contrast to previous studies, which used surveys toexamine the performance of ISO 9000 certified firms, thisstudy examines the performance of ISO 9000 certifiedfirms and non-ISO 9000 firms on the ISE by using theiraverage monthly stock market returns over 105 months.Specifically, the article compared: 1) the monthly stockreturns and their variability of ISO 9000 certified firmswith non-ISO 9000 firms that are traded on Turkey’s ISEmarket; and 2) the returns of ISO 9000 certified firmstraded on the ISE based on ISO 9000 certifying organi-zation (Turkish vs. non-Turkish). The study used aver-age monthly stock market returns of 95 firms listed onthe ISE, 62 of which had ISO 9000 certification and 33of which did not have ISO 9000. The main premise ofthis research was that ISO 9000 firms, because of theirquality orientation (Lee 2003; Ozgur, Meek, and Toker2002), are expected to have higher performance thanthose of non-ISO 9000 certified firms.

To determine if there was a difference between theperformances of ISO 9000 certified firms and non-ISO9000 firms, the study compared the mean stock returnsand average stock variances of ISO 9000 certified andnon-ISO 9000 firms over various time periods of oneyear, two years, three years, and four years during a timeperiod between 1997 and 2005. The results of these sce-narios showed that the ISO 9000 certified firms generallyseem to have higher mean returns than those of non-ISO 9000 firms, but they not are significantly higher forall time periods/scenarios. In addition, comparing theaverage variances of ISO 9000 firms and non-ISO 9000firms, the authors found that ISO 9000 firms had lowervariances than non-ISO 9000 firms over a nine-yearperiod. It appears that, in addition to higher perform-ance given by higher average stock returns, ISO 9000certification seems to reduce the volatility (variability) ofthe stock returns, making these stocks less risky. Oneinteresting observation is that the variability (variances)of stock returns for ISO 9000 and non-ISO 9000 firmsdeclined over time and converged after nine years. Thismay suggest that in the long-run ISO 9000 may not bean important factor in reducing the variability of theirperformance (or variance of the returns).

The authors also made comparisons to see if therewere differences in performance (mean returns and

variability) between Turkish (TSE) certified firms and for-eign certified firms. These findings do not support theimplied preferences for the European certifying agenciesboth in average returns and variability of the stock returns.

These findings are consistent with other empiricalISO 9000 studies (LRQA 1996; Simmons and White 1999;Uzumeri 1997) that say ISO 9000 certification is benefi-cial for the companies. This study also provides supportfor the links between quality and business performancesuggested by theory (Deming 1986; Fine 1986; Juran1974; Lederer and Rhee 1995) and empirical studies(Chowdhury and Menon 1995; Hendricks and Singhal1997; Jacobson and Aaker 1987; Powell 1995). While onlyone prior study examined the impact of variability onbusiness risk, future studies could include the impact ofISO 9000 (or any quality system) on the variability ofthe business performance.

Since the study found no significant differencebetween the mean stock returns and variances of TSEcertified ISO 9000 firms and those of non-TSE certifiedISO 9000 firms, the ISO 9000 certifying organizationdid not have a significant impact on the performanceof the Turkish firms that are traded on the ISE. Thesefindings indicate that the firms do not gain any com-petitive advantage from being certified by foreign(European) agencies. Since international marketingand foreign sales is one of the potential expected bene-fits of ISO 9000 certification, one could expect that theforeign (all European in this case) certified firmsmight be more successful with their exports.

Limitations of the Study There are several limitations to this study. It is impor-tant to state that ISO 9000 registration does not causeor lead to higher stock returns. Following the sugges-tions by Hendricks and Singhal (1997) and Simmonsand White (1999), the longitudinal data covering a105-month time period provided an opportunity toexamine the long-term impact of the ISO 9000 certifi-cation on business performance. While results showedthat ISO 9000 firms had generally higher average stockreturns and lower stock variability than non-ISO 9000certified firms, this study, by design, included only fewvariables (factors) and omitted many others, whichcould have impacted the performance of individual

www.asq.org 37

Pinar v14 i4 9/20/07 10:10 AM Page 37

38 QMJ VOL. 14, NO. 4/© 2007, ASQ

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

firms. Since this study used the stock market returns, itwas not possible to identify additional firm profiles to beincluded in the analysis. To overcome this limitation,future studies may wish to include additional variables/firm characteristics to better understand the impact ofISO 9000 certification on business performance. Stockmarket returns of the firms in this study fluctuatedwidely during the 105 months of this study. A study inthe past analyzed differences in ISE stock returns andstock variability and reached similar conclusions (Pinar,Guder, and Yucel 2000). The contribution of this study isthat it enables one to examine the long-term impact ofISO 9000 on business performance with objective dataand also to analyze the changes in the stock marketreturns and variability over a long time period (nineyears). Unlike the short time period, over nine yearsunexpected positive changes or negative changes willtend to balance out.

The second limitation is that the study did not con-sider other quality programs or initiatives that might beimplemented by the companies, especially by non-ISO9000 companies. Although the research by Lee (2003)and Ozgur, Meek, and Toker (2002) suggests that theISO 9000 firms seem to be more quality oriented thannon-ISO 9000 firms, it was not possible to determine ifnon-ISO firms used any other quality programs fromthe stock market returns. To obtain this information, thesurvey must be conducted. Even so, the survey gives asnapshot of the firms, not long-term perspectives, andthese contributing factors may change year to year. It isbelieved that stock market returns should reflect theimpact of the factors operating in the environment. It isalso assumed that ISO certification should prepare thoseISO certified firms to better cope with these factors. Futureresearch could attempt to overcome this limitation bysupplementing stock market returns with survey data.

The third limitation is that there were manyunidentifiable causal forces at work. Some of themcould have affected ISO 9000 certified and non-ISO9000 firms in different ways. A future study may wishto include some causal forces in the analysis.

The fourth limitation of the study is that the stockmarket and ISO 9000 certification are somewhat recentphenomena in Turkey. There were a limited number offirms that met the criteria to be included. This somewhat

restricted the authors’ analysis, especially with the com-parisons of ISO 9000 vs. non-ISO 9000 firms within thesame industries. Therefore, the differences in industriesand the number of companies from these differentindustries could pose some problems for comparisons.Despite this limitation, this article offers some insightsinto the long-term effect of ISO 9000 certification onbusiness performance. However, the authors recom-mend that similar analysis must be replicated in othercountries where the stock market is more establishedand where many firms are listed.

The final limitation is that this study included onlyTurkish firms. While the results provided some insightsinto the long-term effect of ISO 9000 certification, thefuture studies should include other countries, as well asmulticountry comparisons. Given these limitations, theauthors’ findings must be interpreted with caution.

REFERENCES

Akdeniz, L., A. Altay-Salih, and K. Aydogan. 2000. Cross sectionof expected stock returns in ISE. Working paper, Bilkent University,Ankara, Turkey.

Al-Ghamdi, S. 1998. ISO 9000: Saudi export businesses’ point view.International Journal of Technology Management 16, no. 46: 570-583.

Anderson, S. W., J. D. Daley, and M. F. Johnson. 1999. Why firmsseek ISO 9000 certification: Regulatory compliance or competitiveadvantage. Production and Operations Management 8, no. 1: 28-44.

Barnes, F. C. 1998. ISO 9000 myth and reality: A reasonableapproach to ISO 9000. SAM Advanced Management Journal 63,no. 2: 23-30.

Bhuian, S. N. 1998. Saudi companies’ attitude toward ISO 9000quality standards: An empirical examination. International Journalof Commerce and Management 2, no. 1: 28-42.

Brigham, E. F., and J. F. Houston. 2007. Fundamentals of financialmanagement, fifth edition. Mason, Ohio: Thomson South-WesternPublishing.

Brown, A., T. van der Wiele, and K. Loughton. 1998. Smaller enter-prises’ experiences with ISO 9000. The International Journal ofQuality and Reliability Management 15, no. 3: 273-285.

Brumm, E. K. 1995. Managing records for ISO 9000 compliance.Milwaukee: ASQ Quality Press.

Bureau of Business Practices. 1992. Profile of ISO 9000.Englewood Cliffs, N.J.: Prentice-Hall.

Calisir, F., C. Bayraktar, and B. Beskese. 2001. Implementing the ISO9000 standards in Turkey: A study of large companies’ satisfaction withISO 9000. Total Quality Management Journal 12, no. 4: 429-438.

Chittenden, F., P. Poutziouris, and S. M. Mukhtar. 1998. Small firmsand the ISO 9000 approach to quality management. InternationalSmall Business Journal 17, no. 1: 73-88.

Pinar v14 i4 9/20/07 10:10 AM Page 38

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

Chowdhury, J., and A. Menon. 1995. Multidimensional compo-nents of quality and strategic business unit performance: A PIMStest. Journal of Managerial Issues 7: 449-465.

Churchill, G. A. Jr., and D. Iacobucci. 2005. Marketing research:Methodological foundations, ninth edition. Mason, Ohio: ThomsonSouth-Western.

Clements, R. B. 1993. Quality managers complete guide to ISO9000 registration. Englewood Cliffs, N.J.: Prentice-Hall.

Corbett, C. J., M. J. Montes, D. A. Kirsch, and M. J. Alvarez-Gill.2002. Does ISO 9000 certification pay? ISO Management Systems,Special Report (July-August): 31-40.

Deming, W. E. 1982. Quality, productivity and competitive position.Cambridge, Mass.: MIT Press.

Deming, W. E. 1986. Out of the crisis. Cambridge, Mass.: MITCenter for Advanced Engineering Study.

Ebrahimpour, M., B. E. Whithers, and N. Hikmet. 1997. Experienceof U.S.- and foreign-owned firms: A new perspective on ISO 9000implementation. International Journal of Production Research 35,no. 2: 569-576.

Eddy, T. 1995. Survey provides ‘snapshot’ of ISO 9000 certificationin 10 countries. ISO 9000 News 4, no. 4: 11-13.

Erel, E., and J. B. Ghosh. 1997. ISO 9000 implementation inTurkish industry. International Journal of Operations & ProductionManagement 17, no. 12: 1233-1246.

Ferguson, W. 1996. Impact of the ISO 9000 series standard onindustrial marketing. Industrial Marketing Management 25: 305-310.

Fine, C. H. 1986. Quality improvement and learning in productivesystems. Management Science 32 (October): 1301-1315.

Flynn, B., R. Schroeder, and S. Sakakibara. 1995. Impact of qualitymanagement practices on performance and competitive advantage.Decision Sciences Journal 26, no. 5: 659-691.

Garvin, D. A. 1987. Competing on the eight dimension of quality.Harvard Business Review (Nov.-Dec): 101-109.

Gustafson, R., B. Klefsjo, E. Berggren, and U. Granfors-Wellemets.2001. Experiences from implementing ISO 9000 in small enter-prises—A study of Swedish organizations. The TQM Magazine 13,no. 4: 232-246.

Hair, J. F. Jr., R. P. Bush, and D. J.Ortinau. 2006. Marketingresearch with a changing information environment, third edition.New York: McGraw-Hill/Irwin.

Halls, M., and A. Oztas. 2002. Quality cost analysis in ISO 9000-certified Turkish companies. Managerial Auditing Journal 17, no. 1-2:101-104.

Haversjo, T. 2000. The financial effects of ISO 9000 registration forDanish companies. Marketing Auditing Journal 15, no. 1-2: 47-52.

Hendricks, K. B., and V. R. Singhal. 1997. Does implementing aneffective TQM program actually improve operating performance?Empirical evidence from firms that won quali ty awards.Management Science 43 (September): 1258-1274.

ISE, Is tanbul Stock Exchange of Turkey. 2001. See URL:http://www.ise.org.tr.daily.htm.

ISE, Is tanbul Stock Exchange of Turkey. 2006. See URL:http://www.ise.org.tr.daily.htm.

Ishikawa, K. 1985. What is total quality control? The Japanese way.Englewood Cliffs, N.J.: Prentice-Hall.

ISO. 2005. The ISO survey—2005: ISO and the ISO survey. TheISO Survey of Certifications. Geneva, Switzerland: InternationalOrganization for Standardization.

ISO. 2006a. The ISO 9000 survey sheds light on roles of manage-ment system standards in globalization. Press Release (August 3). SeeURL: http://www.iso.org/iso/en/commcenter/presrelleases/2006/ref1021.html?

ISO. 2006b. Quality management principles: Understand thebasics. See URL: http://www.iso.org/iso/en/iso9000-14000/understand/qmp.html?

Jackson, P., and D. Ashton. 1995. Managing a quality system usingBS/EN? ISO 9000 (formerly BS 5750). London: Kogan Page.

Jacobson, R., and D. A. Aaker. 1987. Strategic role of productquality. Journal of Marketing 51, (October): 31-44.

Jain, B., and F. Tabak. 2002. Organizational quality management inemerging economies. Quality Management Journal 9, no: 2: 10-24.

Jeng, Y. 1998. Performance evaluation of ISO 9000 registeredcompanies in Taiwan. The TQM Magazine 10, no. 2: 132-138.

Juran, J. M. 1974. The quality control handbook, third edition. NewYork: McGraw-Hill.

Kalder. 2006. See URL: http://www.kalder.org.

Kannan,V., T. Keah-Choon, R. Hanfield, and S. Ghosh. 1999. Toolsand techniques of quality management: An empirical investigationof their impact on performance. Quality Management Journal 6,no. 3: 34-49.

Lam, S. K. 1996. Applications of quality improvement tools in HongKong: An empirical analysis. Total Quality Management 7, no. 6:675-680.

Lederer, P. J., and S. K. Rhee. 1995. Economics of total quality man-agement. Journal of Operations Management 12 (June): 353-367.

Lee, C. Y. 2003. A comparative study of manufacturing strategiesand business performance between ISO 9000 and non-ISO 9000firms: A case in the Korean electronics industry. Working paper,Pittsburgh State University, Kansas.

Lofgren, G. Q. 1991. Quality system registration. Quality Progress24, no. 5: 35-37.

LRQA—Lloyd’s Register Quality Assurance. 1996. ISO 9000 regis-tered companies are twice as profitable, says survey. ISO 9000News 5, no. 1: 8-10.

Makridakis, S., S. V. Wheelwright, and V. E. McGee. 1978. Fore-casting: Methods and applications. New York: John Wiley and Sons.

MORI — Market and Opinion Research International. 1996.Attitudes of captains of industry toward ISO 9000. ISO 9000 News5, no. 6: 12-14.

Morita, C. 1996. Japanese survey: ISO 9000 ascending in land ofthe rising sun. ISO 9000 News 5, no. 6: 26-30.

www.asq.org 39

Pinar v14 i4 9/20/07 10:10 AM Page 39

40 QMJ VOL. 14, NO. 4/© 2007, ASQ

The Long-Term Impact of ISO 9000 Certification on Business Performance: A Longitudinal Study Using Turkish Stock Market Returns

O’Brien, T. 1996. Australian ISO 9000 survey reveals significantgains for certified organizations. ISO 9000 News 4, no. 6: 5-12.

Ozgur, C., G. E. Meek, and A. Toker. 2002. The impact of ISO cer-tification on the levels of awareness and usage of quality tools andconcepts: A survey of Turkish manufacturing companies. QualityManagement Journal 9, no. 2: 57-69.

Peach, R.W. 1997. The ISO 9000 handbook, third edition.Homewood, Ill.: Irwin.

Peyrat, O. 2001. New ISO 9000 standards: Generic, attractive, butnot as simple as they look. ISO 900 Management System (October):46-49.

Pfau, L. D. 1989. Total quality management gives companies a wayto enhance posit ion in the global marketplace. IndustrialEngineering 21: 17.

Pinar, M., F. Guder, and T. Yucel. 2000. ISO 9000 standards imple-mentation in Turkey: Reasons, difficulties, and results. The Academyof Information and Management Sciences 3, no. 1: 1-12.

Pinar, M., and H. L. Crouch. 2001. Do all companies achieve theirexpectations from ISO 9000 certification? An empirical study inTurkey. Journal of Global Awareness, no. 2: 62-71.

Powell, T. C. 1995. Total quality management as competitive advan-tage: A review and Empirical study. Strategic Management Journal16 (January): 15-37.

Rabbit, T. J., and P. A. Bergh. 1993. The ISO 9000 book: A globalcompetitor’s guide to compliance & certification. New York: QualityResources.

Rajan, M., and N. Tamimi. 2003. Payoff to ISO 9000 registration.Journal of Investing 12, no. 1.

Rao, S. S., T. S. Ragu-Nathan, and L. E. Solis. 1997. Does ISO 9000have an effect on quality management practices? An internationalempirical study. Total Quality Management 8, no. 6: 335-346.

Rayner, P., and L. J. Porter. 1991. BS5750/ISO 9000—The experi-ence of small and medium-sized firms. International Journal ofQuality and Reliability Management 8, no. 6: 16-28.

Reimann, C. W., and H. S. Hertz. 1996. The Baldrige Award andISO 9000 registration compared. Journal for Quality andProduction 19 (January-February): 12-19.

Rothery, B. 1993. ISO 9000, second edition. Brookfield, Vt: Gower.

SGS Yarsley, International Certification Services. 1996. ISO9000—Does it work? ISO 9000 News 5, no. 6: 4-11.

Simmons, B. L., and M. A. White. 1999. The relationship betweenISO 9000 and business performance: Does registration really matter?Journal of Managerial Issues 11, no. 3: 330-343.

Singels, J., G. Ruel, and H. van de Water. 2001. ISO 9000series—Certification and performance. The International Journal ofQuality and Reliability Management 18, no. 1: 62-75.

Struebing, L. 1996. Survey finds ISO registration is market driven.Quality Progress 29, (March): 23.

Terziovski, M., D. Samson, and D. Dow. 1997. The business valueof quality management systems certification: Evidence fromAustralia and New Zealand. Journal of Operations Management15, no. 1: 1-18.

Terziovski, M., D. Power, and A. S. Sohal. 2003. The longitudinaleffects of the ISO 9000 certification process on business perform-ance. European Journal of Operational Research 146: 580-595.

Uzumeri, M. V. 1997. ISO 9000 and other metastandards:Principles for management practice? The Academy of ManagementExecutive 11 (February): 21-36.

Vloeberghs, D., and J. Bellens. 1996. Implementing the ISO 9000standards in Belgium. Quality Progress (June): 43-48.

Weston, F. C. Jr. 1995. What do managers really think of the ISO9000 registration process? Quality Progress (October): 67-73.

Wilson, S. R. 1999. The impact of standards on industrial develop-ment and trade. Quality Progress 32, no. 7: 71-75.

Yildirim, E. 1999. Modern management techniques in the develop-ing world: The case of TQM and its impact on workers in Turkey.Work, Employment & Society 13, no. 4: 693-709.

Zuckerman, A. 1997. ISO/QS-9000 registration issues heating upworldwide. The Quality Observer (June): 21-23.

BIOGRAPHIES

Musa Pinar is associate professor of marketing in the College ofBusiness Administration at Valparaiso University in Indiana. Hereceived his undergraduate degree from Ege University in Turkey andhis MBA and doctorate from Mississippi State University. He haspublished numerous articles in proceedings and refereed journalssuch as Journal of Euromarketing, Services Marketing Quarterly,Journal of Global Awareness, Tourism Analysis, Journal of Businessand Economics, The Journal of American Academy ofBusiness –Cambridge, The International Journal of Business andPublic Administration, Journal of Business and Behavioral Sciences,Sex Roles, and Journal of International Food and AgribusinessMarketing, and also presented papers at regional, national, and inter-national conferences. Pinar teaches various undergraduate and MBAmarketing classes. His research interests are in the areas of marketorientation and marketing strategy, tourism marketing, service quality,and services marketing in banking, ISO 9000, and gender effect onsales recruiting, selling performance, and sales management. Hehas conducted various marketing seminars in Turkey.

Ceyhun Ozgur is a professor of information and decision sciencesin the College of Business Administration at Valparaiso University.He earned a bachelor’s degree in industrial management and amaster’s degree in management from the University of Akron and adoctorate in business (operations management/operationsresearch) from Kent State University. He coauthored a textbook byMcGraw-Hill Irwin titled Introduction to Management Science withSpreadsheets, first edition. Among others, Ozgur has published inthe Journal of Modern Applied Statistical Methods, Journal of DataScience, Decision Sciences Journal of Innovative Education,Interfaces, OMEGA, Quality Management Journal, Mid-AmericanJournal of Business, Production Planning and Control, InternationalJournal of Operations and Quantitative Management, InternationalJournal of Quality and Reliability Management, InternationalJournal of Business Disciplines, Bogazici Journal, and IndustrialMathematics and Teaching Statistics. He is a member of theDecision Sciences Institute and APICS. He can be reached by e-mailat Ceyhun.ozgur@valpo.edu.

Pinar v14 i4 9/20/07 10:10 AM Page 40

www.asq.org 41

Applying the Toyota ProductionSystem to a Healthcare Organization:A Case Study on a Rural Community

Healthcare ProviderKEVIN F. COLLINS, HEALTHCARE CONSULTANT

SENTHIL K. MUTHUSAMY, BOWLING GREEN STATE UNIVERSITY

© 2007, ASQ

Over the last five decades the Toyota ProductionSystem (TPS) has evolved from an advanced socio-technical concept in manufacturing to a participativedesign for large-scale change management. Toyotahas been able to sustain a strategic competitiveadvantage by applying TPS as a process innovationand intervention, as measured by quality, reliability,productivity, cost reduction, sales and market sharegrowth, and market capitalization. Many organiza-tions are trying to replicate Toyota’s success with TPSin their respective business/industry environments. Itcould be argued that the correlation between theapplication of TPS as part of organizational strategyand Toyota’s documented success in achieving theaforementioned outcomes creates an “industrialengineering paradigm” or “social change interven-tion” that crosses multiple industries. In this light, TPScan be a powerful intervention technique, even inindustries unaccustomed to advanced productiontechniques such as the healthcare industry. Becausethe healthcare industry is under enormous pressure toreduce costs, increase reliability and quality, andenhance organizational effectiveness, TPS-like inter-ventions are significant to healthcare organizations.This article captures the process of applying TPS to ahealthcare organization. It analyzes the challenges,problems, and outcomes, and addresses remedies forenhancing the success of TPS implementation.

Key words: healthcare, patient safety, quality, ToyotaProduction System

INTRODUCTIONSmall rural community healthcare providers, in gen-eral, have limited access to labor and capital resourcescompared to their larger counterparts; however, theyface the same regulatory pressures from accreditationorganizations with regard to quality of care and patientsafety. A paradoxical relationship exists between theamount (and changing nature) of regulation thatexists within the healthcare environment, and theability of community healthcare providers to absorbthe impact of frequent policy and market changes.This is because such providers generally serve smallerdedicated markets and, because of aging populations,are dependent on federal Medicare program reim-bursement. Because that reimbursement is establishedprospectively (not based on cost), the requirement forproviders to effectively manage their cost per patientcase is paramount to ensure viable margins. In addi-tion to the challenge of managing costs, communityhealthcare providers also must be able to offer broad-based healthcare services to the populations theyserve. That puts them in a position to manage qualitycare as well as cost and do so in an environment ofintense patient safety compliance concerns.

Community Health Systems (CHS) is one networkthat is facing these challenges and trying to improvethe performance of its rural hospitals throughadvanced management techniques. Rural communityhealthcare systems constitute 41 percent of the com-munity healthcare service provided in the United

Collins v14 i4 9/20/07 10:09 AM Page 41

42 QMJ VOL. 14, NO. 4/© 2007, ASQ

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

States (AHA 2006). Heartland Regional Medical Centerin Marion, IL, is one of the hospitals in the CHS net-work that is addressing quality, cost management,and patient safety goals together. This hospital isMedicare dependent (qualifying also as a Medicaredisproportionate share hospital) and designated as a“small/rural” provider, according to the IllinoisHospital Association Health Information Department.In preparation for a 2005 accreditation survey, thishospital applied a business-level strategy that not onlycreated a compliant environment of quality care andsafety, but also established the foundation forimproved margins based on cost control. In this arti-cle, the authors capture the application of the ToyotaProduction System (TPS) as a process innovation/intervention technique in this hospital to improveperformance in quality, safety, and cost reduction, andsurvive accreditation audits.

This article reviews various historical and modernconceptualizations of TPS; traces the significance ofTPS in strategy, structural, and cultural interventions;and highlights the role of TPS in establishing controlsystems and standards. Through this case study onHeartland Regional Medical Center, the authors assessthe prerequisites for successful intervention and eval-uate the preparedness of rural community healthcareproviders to implement TPS. The authors also analyzethe regulatory environment within the healthcareindustry, and highlight the need for TPS to improvepatient safety outcomes. Emphasis is placed on TPS asa participative design (PD) intervention within thehealthcare setting. Projects undertaken by the QualityAssurance/Regulatory Compliance Department atHeartland Regional Medical Center also are identified.

TOYOTA PRODUCTION SYSTEM(TPS) AND ITS APPLICATIONS:A REVIEWThe TPS is a manufacturing philosophy and method-ology created by Toyota with the main objective ofreducing wastes, defects, transportation costs, inven-tory, waiting time, and the extent of processing (Liker2003; Monden 1998; Ohno 1995). A modern viewpoint

describes TPS as striving toward an ideal whole system(Emiliani et al. 2003; Spear and Bowen 1999) or onethat has the following output characteristics:

• Is defect free (elimination of “Jidoka” or “Andon”events causing the warning system to “flash”)

• Can be delivered one request at a time

• Can be supplied on demand in the version requested

• Can be delivered immediately

• Can be produced without wasting materials, labor,energy, or other resources

• Can be produced in a work environment that is safephysically, emotionally, and professionally for everyemployee

This definition combines employee involvement inwork design output objectives, and is possibly moresuitable for organizations of all kinds to directly relateto, regardless of technology employed.

The Need for TPS inHealthcare OrganizationsBased on the aforementioned conceptualizations ofTPS, one can visualize how TPS can result in outcomesthat can be considered building blocks of competitiveadvantage, for example, efficiency (elimination ofwaste), customer responsiveness (time and cus-tomization elements), quality (zero defect output),and innovation (creation of a culture that involves andempowers all employees to enhance creativity). Toyotahas been able to sustain strategic competitive advan-tage by applying this process intervention/innovationover five decades, as measured by quality, reliability,productivity, cost reduction, sales and market sharegrowth, and market capitalization. It could be arguedthat the correlation between the application of TPS aspart of organizational strategy and its documentedsuccess in achieving the aforementioned outcomescreates an “industrial engineering paradigm” thatcrosses multiple industries. Further, from a strategicperspective, it could be argued that the application ofTPS plays to cost leadership advantage in industriesunaccustomed to advanced production techniques,

Collins v14 i4 9/20/07 10:09 AM Page 42

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

such as healthcare services (Spear 2004). Healthcareproviders in general and small rural healthcareproviders in particular are in need of a process inter-vention such as TPS and are characteristically similarto that of a manufacturing environment.

Moreover, the external industry environment is alsoforcing the healthcare industry to seek performanceenhancements through the application of techniquessuch as TPS (Spear 2004; 2005). There are severalenvironmental forces that are demanding a changewithin the healthcare industry (Futurescan 2005).These forces include:

1. The level and nature of demand for acute care servicesis changing to include dramatic increases in theneed for specialized services, due to the aging popu-lation and the increasing uninsured population.

2. Hospitals are facing increasing competition indelivering acute care services. Physician-based clin-ics and specialty hospitals are taking business away.An emphasis on quality also is making consumersproactive.

3. New technology offers advantages and challenges.Increased patient safety value comes at a high price.

4. Costs grow while available capital may be shrinking.Increasing costs for wages and pharmaceuticals arechallenging profits.

5. An adequate work force is not guaranteed. Laborshortages are expected to grow.

6. Improved quality and safety standards are a must.Higher performing hospitals will be rewarded.

These trends are more painstakingly realized byrural community hospitals, as access to capital, tech-nology, and resources is somewhat limited because oftheir locations and size.

Like manufacturing organizations, healthcareproviders like those within the CHS network are facingchallenges from rising labor and material costs,intense competition, scarce human resources, cus-tomer demand for impeccable quality, and stringentsafety and performance standards. Table 1 representsapplicability of TPS attributes to hospital operations.

CRITERIA FOR SUCCESSFULAPPLICATION OF TPSTo achieve a competitive advantage using TPSrequires adherence to the fundamental technicalrules, developing suitable organizational/structuraldesign, and fostering cultural norms and values thatfacilitate the TPS processes. Several scholars havetraced the important criteria for successful applica-tion of TPS in terms of task, structural, cultural, and

www.asq.org 43

Table 1 Applicability of TPS attributes to hospital operations.

TPS attribute Hospital application

• Product delivered defect free • Defects correlate to patient safety standards in healthcare. Standards that are establishedby JCAHO require adherence in order for hospitals to be licensed, participate in the federal Medicare program, and differentiate themselves on the basis of quality, in order to effectively deal with increasing demand for specialized acute services. Higher qualityhospitals will be rewarded.

• Product delivered one request at a time • Each patient represents a unique treatment complexity and demands a customized • Product can be supplied on demand treatment plan. When presenting for treatment, care is requested, but may not be available

in the version requested immediately; however, providers can differentiate themselves based on the speed with • Product can be delivered immediately which they can deliver care.

• Product can be produced without • Faced with increasing costs for drugs, technology, and labor, hospital expenses must bewasting materials carefully contained within the context of net revenue produced. Sophisticated material

• Product can be produced without and labor management systems assist hospitals with controlling and budgeting costs.wasting labor

• Product can be produced in a work • Labor shortages in healthcare are expected to grow. With this in mind, hospitals areenvironment that is physically, required to ensure adequate employee training and development in order to mitigateemotionally, and professionally safe turnover, improve morale, and ensure a safe working environment. ©

2007

, ASQ

Collins v14 i4 9/20/07 10:09 AM Page 43

44 QMJ VOL. 14, NO. 4/© 2007, ASQ

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

control systems prerequisites (Emiliani et al. 2003;Spear and Bowen 1999). For example, Spear andBowen (1999) have identified four rules that form thebasis of TPS as a suitable application within theorganization. To be applied successfully, TPS demandsthat an organization’s task structure comply withthese rules:

1. All work must be highly specified as to content,sequence, timing, and outcome.

2. Every customer-supplier connection must be direct,and there must be an unambiguous yes-or-no wayto send requests and receive responses.

3. The pathway for every product and service must besimple and direct.

4. Any improvement must be made in accordancewith the scientific method, under the guidance of a teacher, at the lowest possible level of theorganization.

These rules specify how the work is performed,how knowledge is transferred between workers andwithin the system, how production is coordinatedbetween tasks and services, and how the process iscontrolled, measured, and sustained. These rules alsoimplicitly prescribe the organizational structure andcontrol systems that are required to be in place toimplement TPS gainfully. Additionally, organizationalcultural norms that need to be adhered to are evidentin the knowledge exchanges demanded in the TPSmodel. For example, TPS emphasizes knowledgetransfers, knowledge creation at the lowest level of theorganization, and the contributions required ofemployees at all levels as they interact with their workenvironment. Structure, control systems, and culturalspecifications are integrally built into the TPS model,and in order for this system to work appropriately,organizations need to evaluate these areas and designthem to match the demands of TPS.

The TPS method is scientific in that it clearlydelineates the responsibilities at the operational levelto a very high degree of specification, and it does notpromote vertical command and controls (Spear andBowen 1999). As a result one can see the structure ina TPS environment as decentralized, flat, and having

a minimum chain of command. Additionally, controlsystems are established to raise the level of both inno-vation and quantitative performance. TPS facilitatescreative behaviors and stimulates workers and man-agers to engage in “experimentation” that is widelyrecognized as the cornerstone of the learning organi-zation. Successful companies adapting versions ofTPS use the collective talents of multiple individualsto invent or reinvent work processes. TPS alsoemphasizes certain leadership qualities in managersimplementing TPS. For example, it demands anoperations-oriented team leader who is passionateabout ideas and innovations (Moore 2004; Spear2004). So, it is evident that TPS as a process innova-tion demands sponsorship and commitment from alllevels of employees, and emphasizes organizationalculture that is driven by knowledge creation andinnovation. While this may be a hurdle for somecompanies, it is a prerequisite to the implementationof TPS.

ACHIEVING QUALITY CARE ANDPATIENT SAFETY THROUGH TPS Under the audit methodology established by the Joint Commission on Accreditation of HealthcareOrganizations (JCAHO), hospitals must explicitlydemonstrate a compliant environment of care ineach of the six patient safety goal categories (seeTable 4). They must do so through the process intelli-gence of patient caregivers, and appropriate actionsteps along the clinical pathway. The clinical pathwayor “production line” is audited using two techniques:the patient tracer methodology and the system tracermethodology. JCAHO defines the patient tracermethodology as “taking the patient’s experience andtracing it through the organization and determiningin a variety of different areas—it could be informa-tion management, assessment, treatment, patienteducation — for that individual patient how theorganization is performing, how it’s executing on itsplans related to standards compliance, and best qualitysafest care” (JCAHO 2005). System tracer methodol-ogy is defined as “tracing a process through the

Collins v14 i4 9/20/07 10:09 AM Page 44

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

organization and focusing on the process or the systemitself” (JCAHO 2005). JCAHO uses these tools to reviewthe confluence of individual activities and organiza-tionwide systems and processes to gain a completepicture of effectiveness and performance, much likethe TPS analysis technique. In Table 2, the authorsshow the correlation between TPS elements and theaccomplishment of quality and patient safety.

Hospitals, in general, are in need of a high degreeof integration, because often they are organizedaround functions or disparate departments. As such,they inherently lack the reliable mechanisms for inte-grating the individual departments into a coherentwhole (Spear 2004). One such example of this needfor integration is that in hospital environments, therevenue generation process is started at the time ofpatient registration, with the input of a “raw mate-rial,” that is, patient demographic, insurance, andchief complaint information. In order for the hospitalto appropriately bill for patient care, there must besignificant coordination between the registrationprocess, the throughput clinical and ancillary diag-nostic areas that charge for services, and thebilling/patient accounting department that sends aninvoice to the insurance companies to get paid.Without coordination, lapses in information andprocess standards result in delays, defects, and extracost burdens. TPS allows for effective coordination to occur.

ORGANIZATIONAL ASSESSMENTOF RURAL COMMUNITYHEALTHCARE SYSTEMS A small rural healthcare provider is defined as one thathas fewer than 150 acute care beds and/or is locatedoutside a metropolitan statistical area. HeartlandRegional Medical Center represents a prototype of asmall rural healthcare provider in the CHS network.Located in Marion, IL, it is the only full-service hospitalin Williamson County, IL, which has a population of63,094 (U.S. Census Bureau 2004). Its businessmodel is defined as follows (CHS 2006):

• Increase the share of healthcare dollars spent bylocal residents and limit inpatient and outpatientmigration to larger urban facilities. This is donethrough the recruitment of primary care physi-cians and specialists; targeted capital expendituresdesigned to expand the breadth of services; andproviding capital to invest in facilities, primarilythe emergency room and ancillary services.

• Growth through selective acquisitions (approxi-mately three to four per year).

• Reduce costs through standardizing and centraliz-ing operations; optimize resource allocation byoptimizing the company devised case and resourcemanagement program; capitalize on purchasing

www.asq.org 45

Table 2 The correlation between TPS elements and the accomplishment of quality and patient safety.

TPS rule Enhancement to patient safety

All work shall be highly specified as to content, • All patient treatment is highly specified through documentation of actions,sequence, timing, and outcome. located in patient’s medical record.

• Events, specification of treatment intervention times, and patient outcomes management must be noted.

Every customer-supplier connection must be direct • The patient-nurse interaction is direct, as is the nurse-alternate caregiver.and there must be an unambiguous yes-or-no way • Ambiguity can result in compromises to patient safety; therefore, communicationto send requests and receive responses. must be precise and clear.

The pathway for every product and service must • Clinical pathways established for each treatment initiative, ensuring that a be simple and direct. “best practice” methodology is employed.

Any improvement must be made in accordance with • Quality improvements initiated under the direction of an internal independentthe scientific method, under the guidance of a teacher, body (Quality and Regulatory Compliance), which documents, experiments, at the lowest possible level of the organization. guides, and coordinates patient safety initiatives to be compliant with patient

safety goals, at the caregiver level within the organization. ©20

07, A

SQ

Collins v14 i4 9/20/07 10:09 AM Page 45

46 QMJ VOL. 14, NO. 4/© 2007, ASQ

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

efficiencies; install standardized managementinformation systems; and manage staffing levels.

• Improve quality through training, best practicesharing, assistance in complying with regulatoryrequirements, standardized accreditation documenta-tion, and patient/ physician/staff satisfaction surveys.

As part of the strategy to improve quality, the CHSnetwork has established a standard structure that sup-ports regulatory compliance objectives and internaloperating performance. Table 3 identifies the qualityand regulatory compliance team case problems.

DEMONSTRATION OF TPSACTION STEPS WITHINHEARTLAND REGIONAL MEDICAL CENTER

ObservationIn action, TPS insists on learning from the defects orfailures in operations. Observing failures or defectswithin the system that cause that system to be inefficientis a key principle of TPS. An example of inefficiencies inhospital operations is observing patients being routed toa nursing unit after arrival for inpatient admission. Theadmission check-in procedures require all patients tohave identification banding performed when they arrive

at the facility. This banding, by JCAHO standards, is tohave two patient identifiers. As part of the audit processat Heartland, it was found that expectant mothersarriving for delivery were told by their physicians toreport directly to the obstetrical unit and bypass admis-sions. By doing so, the admission requirements werenot being met and, therefore, a breach in the standardwas identified. This created ambiguity in the nursingunit and resulted in the patient being sent backupstream in the process. In this case, treatment com-menced on the patient prior to proper identificationand treatment consent, which is a JCAHO violation ofpatient safety goal no. 1.

To deal with this issue, a team comprising first-shift nursing unit staff and two admitting clerks fromthe dayshift hypothesized that coordinating patientregistration closer to the point of presentation/carewould alleviate incorrect routing of patients. Theteam devised a countermeasure plan to create amini-registration unit in the obstetrical unit. Theystocked supplies and forms, and requested a labelprinter from information services to be used for identi-fication banding, so they could band patients arrivingdirectly from the physician’s office. Admitting clerkstrained nurses on the proper procedures, and themethod was tested with staff members acting aspatients. The result was improved customer respon-siveness from the patient, as well as physicians, whopreviously received negative feedback from theirpatients. And although complaint issues had beendocumented, directly observing the breakdown in thesystem prompted the change to occur.

Testing of Hypotheses Through ExperimentationTPS also involves fine tuning or constant “experi-mentation.” This allows contingencies to be exploredin greater depth, and the interaction among the staffcreates more unity as alternative options areexplored, problem scenarios are presented, andresults are debated. In the previous example, twooriginal alternative hypotheses were agreed upon to move to experimentation. The first one has been

Table 3 Quality and regulatory compliance team case problems.

Case problems Accountable areas

Obstetrical admission/ Patient registration/check-in procedures Obstetrical nursing unit

Medication management Pharmacy/and storage procedures Emergency nursing unit

Use of standard abbreviations Health information management/Nursing administration

Reconciling patient medications Pharmacy/Nursing units

Patient risk prevention program Nursing units ©20

07, A

SQ

Collins v14 i4 9/20/07 10:09 AM Page 46

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

discussed; the second one involved a paging system,whereby an admitting clerk would receive notifica-tion that a breakdown had occurred on the unit, so asto then be summoned to the unit to complete theadmitting and banding procedure. A pilot of thishypothesis failed for the following reasons: respon-siveness from the admitting staff could not always bepredicted. When a page was sent, it was found thatadmitting clerks may not have been available at thetime because they were performing admitting proce-dures on other patients. Their ability to respond tothe page was compromised, but for good reason. Also,because of staffing constraints on the second shift,particularly during the 3:00 p.m. to 6:00 p.m period,admitting clerks were unable to leave the work area.On the nursing side, the staff members were primarilyinterested in ensuring that the patient received initialtreatment and, therefore, became ambivalent to goingthrough the paging procedure. This hypothesis wasdiscarded, but only after experimentation among thesame group of workers occurred and findings wereconfirmed.

Coordinated ConstantExperimentationAs customization increases, demand increases, andservice or product categories are enhanced, existingsociotechnical systems cannot remain static. As aresult, existing processes must undergo coordinated,constant experimentation. Documentation of thecoordinated experimentation efforts through a log-ging system allows for reflection and progression. Thisprocess permits the adaptation of design as the prod-uct or service changes, and it does so in shorter cycletimes, allowing for quicker changes to occur. As such,any redesign process does not become burdensome,overcomplicated, or drawn out over long periods oftime. Members of the original obstetrical department“mini-registration” initiative continue to evaluatethe process as new physicians are recruited, a changein key personnel takes place, or other departmentssuch as case management require information fromthe task. Constant experimentation also battles the

constraint of worker turnover. It acts as a learning labfor new employees, and shortens the learning curve onrequired processes.

It’s important, therefore, to engage in constantexperimentation to solicit better ideas and ensurecontinuous training, as Heartland and hospitals ingeneral compete to recruit and retain valuablehuman resources.

Manage Through EnablingLower-Level WorkersTPS is effective only when lower-level workers areinvolved in solving problems. Spear (2004) points outthat the higher the level of management, the lesser thelevel of employee involvement in problem solving.Within the hospital sphere, the Quality Assurance andRegulatory Compliance (QRC) department managesthe boundary. They are observed acting as enablers inthe system, using the JCAHO patient safety goals as aguideline for staff and managers to produce their ownresults. The result is measured against the patientsafety goal and the QRC department processesdescriptive statistics to allow for team evaluation.The QRC team is also there for interpretation — toevaluate experimentation against the context of thepatient safety goal, as intended by JCAHO. Also, func-tional managers are not formal managers in thechange management process within Heartland’squality initiatives. They are members of the team,and content expertise is pushed down the formalchain of command. Another example solidifies therules of engagement and TPS action steps describedup to this point.

Establishing Countermeasures toBreaching Quality/Safety GoalsThe QRC observation practices usually result in evi-dence of a gap in the process of delivering patientcare. Something within the clinical pathway can bepinpointed as askew, and in the case of medicationmanagement within three clinical departments, that

www.asq.org 47

Collins v14 i4 9/20/07 10:09 AM Page 47

48 QMJ VOL. 14, NO. 4/© 2007, ASQ

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

gap was discovered. A drowning victim arrived in theemergency room at Heartland. At the same time, aQRC team member was stationed in the emergencyroom setting, observing practices being followed on anumber of patient encounters. Drowning victimshave a hypothermic body temperature of less than 95degrees F. Part of the treatment protocol is introduc-ing warm intravenous (IV) fluid to the victim in anattempt to revive the organs. This technique alsoincreases body temperature slowly, without inducingshock.

The QRC team member observed the case nurseproceed to an incubation storage unit, which looksmuch like a refrigerator except that it is designed towarm the contents rather than cool them. Each IV fluidset has a manufacturer’s recommended temperature-level threshold, so as not to disable the effectivenessof the medication when introduced to a patient. Thesequence of steps involved in obtaining the IV fluidfrom the incubation unit includes verification of datestored, temperature setting, and manufacturer’s rec-ommended threshold. The discovery by the case nurseof an excessive temperature setting within the incu-bation unit caused her to question the reliability ofthe IV solution she was obtaining for the patientunder her care. To her surprise, the heat thresholdhad been exceeded to a level greater than what wasrecommended by the manufacturer and, of course,caused an immediate patient safety concern.

Because of the ambiguous situation, the nurse cre-ated a work-around, which meant a “stat” page topharmacy for retrieval of the same IV solution fromtheir storage unit. Luckily for everyone involved, theclinical pathway continued with a favorable outcomefor the patient, notwithstanding the delay involved inprocuring the IV solution. This is precisely the value ofTPS. The observation of a work-around, a noticeabledefect in the process, and an ambiguity of next steps allrequired intervention toward standardization so as toconfirm patient safety rather than guess at it. Whenthe “red light” went on, a “Jidoka” or “Andon” eventoccurred, and the wheels were put into motion tobreak down the sequence of events. Within minutes, ateam comprising first-shift nurses and two emergencyroom clerks huddled with the QRC team member to

analyze this situation. The first point made was thatthis represented a breach in patient safety goal no. 3, asprescribed by the JCAHO, and if the JCAHO was on siteduring this event it would be considered a “type 1,” orsevere, violation causing the hospital to potentiallylose accreditation. This was the “burning platform”required for the next step, and that was the solicita-tion of open ideas to mitigate this from occurring inthe future. Focus was on accountability to ensure anadequate record was kept of the three variablesinvolved in the storage of IV fluid medication, that is,existing temperature of the incubator, date of originalstorage, and manufacturer’s recommended storagethreshold. The team members chose a lead from eachshift to comply with completion of a mediation man-agement log on the incubator, which included thesethree items as well as the name of the team memberchecking them.

For a period of one month, the same person isrequired to perform the evaluation, and a second teammember verifies that it was performed. This project isin the “experimentation” stage at Heartland. Theimportant lessons are: 1) the huddle took placeimmediately after the observation; 2) the QRC teammember acted only to create the “burning platform”instead of acting as the problem solver; and 3) theteam created a countermeasure within the context ofparticipative design, that is, the form and accountabil-ity for completing it was generated by a collaborationof workers. This exercise clearly follows the spirit ofTPS. However, for it to be useful to the hospital as awhole, it must be spread throughout the institution,and the process of observation and experimentationmust be everyone’s task.

Additional ObservationsIt was found that the QRC team employed the sameTPS “whole system” methodology when performingadditional patient safety audits. These findings and theimplemented remedies are explored here. In order toimprove communication among caregivers and ulti-mately ensure patient safety, a common theme withinhospital facilities is to implement a standardized list ofabbreviations and also identify a “dangerous” list of

Collins v14 i4 9/20/07 10:09 AM Page 48

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

abbreviations to be avoided. The QRC team atHeartland, through chart auditing, discerned thatseveral abbreviations and symbols were potentially inuse that could result in the intended meaning beingmisinterpreted and patient safety being compromised(a potential breach of patient safety goal no. 2).

Two examples of this are illustrated here to draw thelink between interpretation and potential patient safety

compromise. The abbreviation “IU” was documentedwith the intended meaning of “international unit;”however, it could have been misinterpreted as “IV”—“intravenous,” or the number 10. To abate this prob-lem, the recommendation was to require that thephrase “international unit” be appropriately spelledout, instead of abbreviated. Another “dangerous”abbreviation identified was the symbol “Q.D,” which

www.asq.org 49

©20

07, A

SQ

Table 4 Goals related to improving quality of care and patient safety and the associated TPS strategy.

Goals Affect on healthcare operations Application of TPS philosophy

Goal 1 Wrong-patient errors occur in virtually all aspects of Creating a treatment plan that is unique to the Improve the accuracy diagnosis and treatment. The intent for this goal is individual fulfills the TPS requirement that all workof patient identification twofold; first, to reliably identify the individual as is highly specified as to content. While clinical

the person for whom the service or treatment is pathways can be created for “categories” of care,intended; second, to match the service or treatment each represents a customized version of the careto that individual. template. Pathways must be documented and act

as controls during delivery of care.

Goal 2 Ineffective communication is the most frequently Reduction in errors is the direct result of creatingImprove the effectiveness cited category of root causes of sentinel events. an unambiguous customer-supplier connection. of communication Effective communication, which is timely, accurate, Role-playing and experimentation allows caregiversamong caregivers complete, unambiguous, and understood by the to identify weaknesses in that connection.

recipient, reduces error and results in improved patient safety.

Goal 3 When medications are part of the patient treatment A specific pathway must be created and followedImprove the safety of plan, appropriate management is critical to consistently in order to analyze variations fromusing medications ensuring patient safety. The development of the standard.

standardized and redundant systems has been shown to decrease errors and improve outcomes.

Goal 4 Compliance with the CDC hand-hygiene guidelines Work environments must be physically safe.Reduce the risk of will reduce the transmission of infectious agents by Training and development programs that addresshealthcare-associated staff to patients, thereby decreasing the incidence the environment of care satisfies the TPS attribute infections of healthcare-associated infections. of workplace safety.

Goal 5 Patients are most at risk during transitions in care Implementation of a product delivery plan Accurately and (handoffs) across settings, services, providers, or (reconciling medications at the point of productioncompletely reconcile levels of care. The development, reconciliation, and dispensing) ensures safe customized delivery. medications and communication of an accurate medication list

throughout the continuum of care is essential in the reduction of transition-related adverse drug events.

Goal 6 Falls account for a significant portion of injuries in Identify risks that create a potential for failure, Reduce the risk of hospitalized patients, long-term care residents, and through repeat assessment of patient care areaspatient harm resulting home care recipients. In the context of the population from shift to shift. Identify potential risks as “defects”from falls it serves, the services it provides, and its environment and mitigate them accordingly.

of care, the organization should evaluate its patients’risk for falls and take action to reduce the risk offalling and to reduce the risk of injury, should a fall occur. The evaluation could include fall history,medications and alcohol consumption review, gaitand balance screening walking aids, assistivetechnologies and protective devices assessment, andenvironmental assessments.

Collins v14 i4 9/20/07 10:09 AM Page 49

50 QMJ VOL. 14, NO. 4/© 2007, ASQ

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

was intended to mean “everyday.” This, however, wasroutinely misinterpreted as “Q.I.D” or “every otherday.” Clearly, this misinterpretation could result inmedication errors and a patient safety concern. Thisabbreviation was discontinued in favor of the phrase“daily,” or “every day.”

To reduce the risk of healthcare-acquired infec-tions (patient safety goal no. 4), the QRC departmentat Heartland not only installed reminder signs to“wash hands for a minimum of 10 to 15 seconds”within each clinical area, but also came up with arhetorical/symbolic method for helping employees

Table 5 Specific case problems categorized by patient safety goal.

Goals Process rules to implement TPS Problem assessment Remedies

Goal 1 • Use at least two patient identifiers (neither to be the • Direct admissions to • Implementation ofImprove the accuracy of patient’s room number) whenever administering obstetrical unit precluded “mini-registration”patient identification medications or blood products, taking blood proper identification of process on the

samples and other specimens for clinical testing, patients. obstetrical unit.or providing any other treatments or procedures.

Goal 2 • For verbal or telephone orders, verify the complete • Potential misinterpretation • Eliminate commonImprove the effectiveness order or test result by having the person receiving of abbreviations and misinterpretations inof communication the order or test result “read back” the complete symbols. favor of mandatedamong caregivers order or test result. documentation.

• Standardize a list of abbreviations, acronyms, andsymbols that are not to be used throughout theorganization.

• Measure, assess, and take action to improve the timeliness of reporting of critical test results.

• Implement a standardized approach to “hands off” communications including an opportunity to ask and respond to questions.

Goal 3 • Standardize and limit the number of drug • Excessive temperature • Implementation Improve the safety of concentrations available within the organization. setting within IV storage of medication using medications • Identify and annually review a list of look alike/ incubation unit. management log,

sound alike drugs used in the organization, and accountability fortake action to prevent errors involving the completion of loginterchange of these drugs. assigned to team lead.

• Label all medications, medication containers, orother solutions in preoperative and sterile settingsand ensure proper storage conditions.

Goal 4 • Comply with the CDC hand-hygiene guidelines. • Potential noncompliance • Install signage andReduce the risk of • Manage as sentinel events all identified cases with CDC handwashing provide emblematichealthcare-associated of unanticipated death or major loss of function guidelines. method as a reminder.infections associated with a healthcare infection.

Goal 5 • Implement a process for obtaining and documenting • Potential avoidance of • Mandate use of Accurately and a complete list of the patient’s current medications medication management medication completely reconcile upon admission. assessment. management medications • A complete list of the patient’s medications is documentation as

communicated to the next provider of service essential componentwhen a patient is referred or transferred to of patient’s chart.another setting. Subject to QRC audit.

Goal 6 • Implement a fall reduction program and evaluate • Potential improper lifting • Implement employeeReduce the risk of the effectiveness of the program. and transportation retraining and buildpatient harm resulting techniques. training into newfrom falls employee orientation.

Subject to QRC audit. ©20

07, A

SQ

Collins v14 i4 9/20/07 10:09 AM Page 50

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

understand this. During orientation for new employees,and reorientation of existing employees, the QRC teamdemonstrated the use of the “Happy Birthday” song,that is, singing the song while washing one’s hands toensure hygiene guidelines are met. This was not onlyengaging, but also emblematic and pragmatic.

Patient safety goal no. 5 requires hospitals toaccurately and completely reconcile patient medica-tions. As a patient proceeds through the admissionprocess and is referred to a clinical unit, that unit isrequired to document existing medication use. TheQRC team facilitates this requirement by ensuringthat the medication management documentation is amandated component of the patient’s chart and sub-ject to review by the QRC audit team. Finally, patientgoal no. 6 requires hospitals to engage in a riskreduction program to manage patient harm resultingfrom falls. In many cases, it was determined thatpatient falls resulted from inappropriate physicalassistance provided to the patient by the caregivers.This was observed by QRC members, and a retrainingprogram was implemented to ensure proper liftingand transportation techniques. In addition, the pro-gram was instilled in the new employee orientationprogram, and is also subject to audit through QRCobservation. A summary of the goals related toimproving quality of care and patient safety, the effecton healthcare operations, and the associated TPSstrategy are presented in Table 4.

The specific case problems are categorized accord-ing to patient safety goal, with remedies identified inTable 5.

LESSONS LEARNEDAs previously mentioned, small rural communityproviders have a magnified challenge in dealing withcurrent trends, particularly access to capital, technol-ogy, and resources, and dealing with cost burdensimposed by pharmaceuticals and technically skilledhuman resources. CHS has devised a business modelthat provides guidance to its small rural communityproviders in the area of quality and patient outcomemanagement, specifically requiring standardizationof accreditation documentation. This is a model that

when enforced can produce positive outcomes, andwhen implemented with the help of methods like TPScan result in building competitive advantages.

In returning to Emery’s (1995) rationale for devel-oping participative design, one can find constraints inintroducing TPS to a community hospital. In essence,the rationale describes universal organizational con-straints that must be overcome. Traditional hierarchiesare unacceptable in a TPS environment, where partici-pation of employees creates collective consciousnessand a shared mental model, rather than fragmentedtop-down solutions. This requires rethinking the roleof management, one that precludes the higher levelsof the hierarchy exclusively acting in roles of problemsolvers at the expense of the ideas of the work group.Additional studies that account for variations inmanagement style are recommended.

Systemwide, expansion of TPS from a work unitconcept to a whole system change is limited by “silo”affects, or the divisional nature of hospital operations.Breaking down these silos through challengingprocesses within each of them increases the interactionamong divisional employees and recognizes theirinterdependence. Additional TPS intervention studiesthat take into account the diversity of multidiscipli-nary operations is necessary to ascertain the level ofsustainability of the TPS intervention.

Successful projects that are not sustainable throughtime because of worker turnover can be mitigated bystrategic reward systems such as pay for performanceor best practice rewards tied to patient safety and costreduction outcomes. Incentives built into a controlsystem help the employees to realize a sense of owner-ship in their actions. Additional studies are needed todetermine the strength of correlation between com-pensation/reward systems and TPS sustainability.Finally, hiring practices and skill-building mecha-nisms associated with a learning organization cancreate an expert-driven process that possesses theenergy for change rather than the complacency ofresistance. This would include creating observationskills associated with TPS, training in the scientificmethod, and establishing interdisciplinary processeducation sessions. This will build the perception thatredesign is part of the job description rather than a

www.asq.org 51

Collins v14 i4 9/20/07 10:09 AM Page 51

52 QMJ VOL. 14, NO. 4/© 2007, ASQ

Applying the Toyota Production System to a Healthcare Organization: A Case Study on a Rural Community Healthcare Provider

complicated process, and move the right people intothe right “experimentalist” positions, such that theorganization will start to achieve documented strate-gic results, much like Toyota. The structure, controlsystems, and culture will make the difference.

REFERENCES

AHA. 2006. American Hospital Association (AHA) statistics.Washington D.C.: American Hospital Association.

CHS. 2006. See URL: http://www.chs.net/pdf/Investor-FactSheet.pdf.

Emery, F. 1995. Participative design: Effective, flexible and successfulnow! The Journal for Quality and Participation 18, no. 1: 6-10.

Emiliani, M. L., D. Stec, L. Grasso, and J. Stodder. 2003. Betterthinking, better results: Using the power of lean as a total businesssolution. Kensington, Conn: The CLBM.

Futurescan. 2005. Futurescan: Healthcare trends and implications2005-2010. Chicago: Society for Healthcare Strategy and MarketDevelopment.

JCAHO. 2005. Facts about hospital accreditation. See URL:http://www.JCAHO.org.

Liker, J. 2003. The Toyota way: 14 management principles fromthe world’s greatest manufacturer. New York: McGraw-Hill.

Monden, Y. 1998. Toyota production system: An integratedapproach to JIT, third edition. Norcross, Ga.: Engineering andManagement Press.

Moore, G. 2004. Innovating within established enterprises, top-line growth. Harvard Business Review (July-August): 87-92.

Ohno, T. 1995. Toyota Production system: Beyond large scaleproduction. University Park, Ill.: Productivity Press.

Spear, S. 2005. Fixing healthcare from the inside today. HarvardBusiness Review (September): 78-91.

Spear, S. 2004. Learning to lead at Toyota. Harvard BusinessReview (May): 79-86.

Spear, S., and H. K. Bowen. 1999. Decoding the DNA of the ToyotaProduction System. Harvard Business Review (Sept.-Oct): 97-106.

U.S. Census Bureau. 2004. Estimate. Washington D.C.: U.S.Census Bureau, Population Division.

BIOGRAPHIES

Kevin F. Collins is a management consultant to a diverse range ofhealthcare organizations. He received his master’s degree inorganization development from Bowling Green State University.He specializes in revenue optimization and organizationalprocess improvement strategies. Collins is a member of theAmerican Health Information Management Association, theHealthcare Financial Management Association, and the Academyof Management. His research interests include healthcare strat-egy, organizational behavior, clinical quality outcomes analysis,and leadership development. He can be reached by e-mail atKevincollins3@lycos.com.

Senthil Kumar Muthusamy is an associate professor of strategicmanagement in the College of Business Administration atBowling Green State University. He received his doctoral degreein management from Oklahoma State University. He is a memberof the Academy of Management, the Strategic ManagementSociety, and the Organization Development Network. Hisresearch interests include social exchanges in interfirm alliances,cooperative strategies, knowledge management, and organiza-tional cognition. His articles have appeared in OrganizationStudies, Journal of Business Research, Organization DevelopmentJournal, and Journal of Information & Knowledge Management.

Collins v14 i4 9/20/07 10:09 AM Page 52

www.asq.org 53

Ranking Customer Requirementsin QFD by Factoring in Their

Interrelationship ValuesJ. R. SHARMA, INSTITUTE OF MANAGEMENT TECHNOLOGY, NAGPUR

A. M. RAWANI, NATIONAL INSTITUTE OF TECHNOLOGY, RAIPUR

© 2007, ASQ

Quality function deployment (QFD) is a methodologyused to achieve higher customer satisfaction. The engineer-ing characteristics (ECs) affecting product performanceare designed to match the customer requirements (CRs).Computing raw weights of CRs and priority scores of ECsfrom various input variables is one of the most criticalphases in QFD application. This study deals with the mostneglected and often-omitted part of the QFD process—theinterrelationship matrix among customer requirements.This article presents a mathematical solution to the cor-relation triangle problem that incorporates both CRs andECs. An attempt is made to address the issue by consider-ing the impact of these interrelationships on the overallcalculations, using weighted average method for the CRinterrelationship value operator in order to prioritizecustomers’ requirements in QFD. The outcome of thestudy is a mathematical solution to the perennial corre-lation triangle problem in the form of a framework thatfactors in the correlation triangle values in conjunctionwith computed raw weights for customer requirements.The model fine tunes and adds precision to an otherwisequalitative strategic decision process. The applicabilityof the authors’ proposed model is demonstrated with areal-life example.

Key words: customer requirements, interrelationshipvalues, prioritization, QFD

INTRODUCTIONIn any kind of project, the design team needs to create,improvise, innovate, or improve a product by incorporat-ing the customer requirements (CRs) and engineeringcharacteristics (ECs) recognized in the quality functiondeployment (QFD) planning process into a product. Adesign team needs to make tradeoffs while selecting theCRs and ECs based on the order of their relative rawweights and priority scores and their ranking to achievetotal customer satisfaction without violating the time,money, and effort constraints. The technical priorityscore and their rankings are the key results of QFD, sincethey guide the design team in decision making, resourceallocation, and the subsequent product analyses.Therefore, deriving the final importance rankings of CRsand ECs from input variables is a crucial step towardsuccessful QFD. The inherent abstraction and theinexactness of the result, however, presents a specialchallenge in effectively calculating the customer rawweights and technical priority scores and, hence, theimportance rankings of CRs and ECs. The aim of thisarticle is to advance the theory and practice of incor-porating and factoring in the interrelationship valuesof the house of quality (HoQ). Implementation ofthis improvised approach to comprehensive matrixanalysis is discussed from the perspective of a productby conducting a hypothetical analysis.

QFD: ORIGIN AND PROGRESSIONQFD is the systematic translation of the voice of thecustomer (VOC) to the actions of the manufacturerrequired to meet the customers’ desires, based on a

Sharma v14 i4 9/20/07 10:11 AM Page 53

54 QMJ VOL. 14, NO. 4/© 2007, ASQ

Ranking Customer Requirements in QFD by Factoring in Their Interrelationship Values

matrix that compares what the customer wants andhow the supplier plans to provide it. It is said that QFDwas proposed in Japan in 1966 by Yoji Akao. However,it did not emerge as a viable and formalized approachto quality control in planning until 1972 when Akaodeveloped a quality control chart previously introducedat the Kobe shipyard of Mitsubishi Heavy Industriesand instituted the QFD quality tables (Griffin andHauser 1996). QFD is an innovative approach tobringing quality—as demanded by the customers—upstream into the product development process (Akao1990). QFD’s first industrial applications in the UnitedStates originated primarily in the automotive industry.With some modifications, QFD could be applied to thecontinuous improvement of any product, process, ortechnique (Sharma and Rawani 2006). Early users ofQFD included Ford Motor Company, Procter & Gamble,and 3M Corp., but many other U.S. companies have alsoadopted it. In Europe, the first symposium on QFD tookplace in 1992 in the United Kingdom, but companieslike Philips Corporation have reportedly worked withQFD since 1986 (Govers 1996). QFD has been in use inthe manufacturing industry since 1987. Meanwhile,many authors have advocated it as a planning tool tohelp manage product and process development.

QFD FRAMEWORK In the QFD approach, the matrix to be built is theproduct planning matrix. It also is called the “house”of quality (HoQ) because of its house-like shape. Itspurpose is to translate important customer require-ments regarding product quality into key end-productcontrol characteristics. The HoQ comprises severaldifferent parts or rooms, which are sequentially filledin order to achieve an actionable translation fromrequirements into characteristics (see Figure 1).

Voice of Customer The HoQ’s first room concerns the product and dealswith the definition of the selected product for which QFDis constructed. The second room concerns the VOC, butbefore this there is a need to identify the customers.There are different ways to identify the customers of a

product or service. The most common approach is toask: Who must be satisfied with the product for it to beconsidered successful? Identifying the CRs and theirimportance remains one of the most serious obstacles toa successful QFD application in product development.

CR Triangle The CR correlation triangle is designed to show theinterrelationship between the CRs of the matrix.Tradeoffs between the different CRs are identified andcaptured as strong positive, positive, negative, and strongnegative. If two CRs help each other move in theirdesired direction, they are considered to be positivelyrelated. Similarly, if moving in the desired direction onone CR has a detrimental effect on another, they areconsidered to be negatively related. Since they are gener-ally not used in calculation, the interrelationship dataare defined with symbols for associated relationships.

Strategic Planning RoomOnce the CRs and their priority rankings are established,the next step is to understand where the company andits competitors stand in terms of satisfying customers inthe marketplace. This is achieved through an index ofthe customer perceptions as to how well the productone uses meets his or her needs. This section provides alink between the QFD project and the company’s strate-gic vision by uncovering market opportunities and

Roof

Product

Relationshiproom

Voice of engineer

Strategicplanning

room

Technicalpriorities

room

Voic

e of

cust

omer

s

CR

trian

gle

Figure 1 Basic building blocks of QFD(Sharma, Sharma, and Rawani 2006).

©20

07, A

SQ

Sharma v14 i4 9/20/07 10:11 AM Page 54

Ranking Customer Requirements in QFD by Factoring in Their Interrelationship Values

identifying priorities for product development orimprovement. Moreover, it helps build competitiveadvantage by focusing on the requirements that cus-tomers would like to see better addressed by the market.

Voice of the Engineer and RoofAt this stage, the QFD team should have a clear pictureof what the customer requires from the concept orproduct and how this can be related to the company’sstrategy. The team now has to decide how these require-ments can be incorporated into the final product, so thatthe customer is satisfied. This is achieved by buildingthe Voice of Engineer Room. The goal is to develop aset of engineering specifications from the CRs.

Relationship RoomThe Relationship Room is the core of the HoQ—wherethe relationship between each CR (VOC) and the ECs(voice of the engineer), as well as their intensity, isdepicted. The relationship between CRs and ECs mightbe achieved by asking, “To what degree does this engi-neering characteristic predict the customer’s satisfactionwith this requirement?” Based on in-house expertise,surveys, data from statistical studies, and controlledexperiments, the team seeks consensus as to how mucheach EC affects each CR. This task is widely recognizedas highly complex and represents another critical stagein the HoQ building process.

Technical Priorities Room The Technical Priorities Room starts with technicaldescriptors. These descriptors pertain to the ECs andthe units of measurement in each column. They addthe details necessary to bring the product definitionfrom the abstraction of words to the concrete reality ofproduct and process engineering. With technical com-petitive assessment of the ECs (voice of the company)in the currently marketed products, the QFD team canview their competitors’ and their own technical per-formance level regarding ECs directly affecting CRs. Arow indicating the level of organizational difficultyrelated with realizing each EC can also be added.

Based on the information depicted in the HoQ, theQFD team then must select the end-product character-istics to be deployed through the remainder of theproduct development and market introduction activi-ties. ECs showing a technical importance rating abovea predetermined threshold, thus indicating a strongrelative importance in achieving the CRs, are selectedfor further deployment. The same happens to ECsrelated with CRs that have strong sales points or poorcompetitive performance satisfaction rating.

RANKING METHODS IN QFD: A REVIEW The study of the available literature on various modelsand frameworks for QFD suggests that most of themethodologies suffer from one major weakness—alack of concrete, clear guidelines as to how one couldadequately conceptualize, integrate, and implementits correlation triangle phase. Several methodologieshave been worked out, but their validity and applica-bility to prioritize CRs remains inconclusive and theimplementation results and findings are scarce.

In QFD studies various CRs are always stated andincluded, but interrelationship with one another ishardly ever incorporated and, even if depicted, isnever brought into the tabulations of the CR finalimportance ratings. One reason for this is the diffi-culty in obtaining the relevant data, although itcould be expressed in a symbolic manner usingappropriate synergist ic or detrimental scales .Interrelationships among the CRs require the QFDpersonnel to make a lot of pair-wise comparisonsabout the degree of association and the direction towhich these CRs are interrelated. Also, there are few,if any, acceptable methods to incorporate the interre-lationship into the calculation of the final impor-tance ratings of the CRs. Most of these methods,however, adopt calculation procedures using theweighted product of the relationship measures in theimportance rating column, sales factor, andimprovement ratio column, without considering themagnitude, direction, and degree of associationamong all CRs.

www.asq.org 55

Sharma v14 i4 9/20/07 10:11 AM Page 55

56 QMJ VOL. 14, NO. 4/© 2007, ASQ

Ranking Customer Requirements in QFD by Factoring in Their Interrelationship Values

A few approaches have been introduced for deter-mining CRs and ECs importance rating in the QFDprocess. The noticeable ones found in the QFD litera-ture dealing with prioritization are mentioned here:The prioritization that is consistent with CRs impor-tance ratings was advocated by Lyman (1990)through the concept of deployment normalization.Armacost et al. (1994) used the analytic hierarchicalprocess to prioritize CRs and ECs in an industrializedhousing application. Khoo and Ho (1996) proposed

an approach based on the application of possibilitytheory and fuzzy arithmetic to address the vaguenessin QFD operation. Fung, Popplewell, and Xie (1998)developed a hybrid system to incorporate the principlesof QFD, analytical hierarchy progress, and fuzzy settheory for customer requirements analysis and productattribute targets determination. Shen, Tan, and Xie(2001) proposed a fuzzy procedure to examine thesensitivity of ranking different parameters to thedefuzzification strategy and degree of fuzziness offuzzy numbers. Vangeas and Labib (2001) proposed amodel for deriving optimum targets of CRs and ECsthrough the implementation of the fuzzy weightedaverage. On the similar lines, new methods for ratingCRs and ECs in fuzzy QFD is proposed by integratingfuzzy weighted average method (Kao and Liu 2001)and the fuzzy expected value operator (Liu and Liu2002). Karsak, Sozer, and Alptekin (2002) combinedthe AHP and goal programming approach for deter-mining the priority levels of CRs. Han, Kim, and Choi(2004) suggested a linear partial ordering approachfor prioritizing ECs in QFD with incomplete informa-tion. Karsak (2004) used the fuzzy multiple objectiveprogramming framework for prioritization in QFD.Chen, Fung, and Tang (2005) proposed a rankingmodel using the weighted average method in the fuzzyexpected value operator in fuzzy QFD.

INTERRELATIONSHIPINTEGRATED MATHEMATICALMODELTo overcome the aforementioned problem, a newmethod for prioritizing CRs in QFD is proposed byintegrating the CR correlation triangle values withpreliminary (initial) raw weights. The proposedmethod can weigh the CRs more efficiently, as not onlythe relative importance ratings, sales points, andimprovement ratios of CRs are considered, but the val-ues and their degree of association in the correlationtriangle are also factored in. Through the methodproposed in this study, the CR correlation trianglesymbols are translated into numerical values andwith this a discreet but exact solution is obtained.

IRWi = NPRWi

OUTPUT: IRW VALUES

i = 1

j = 1

START

STOP

If j ≠ i

j = j + 1

If j ≤ n

i = i + 1

If i ≤ n

Yes

No

Yes

No

IRWi = NPRWi + Σ [NPRWi * (1 + NPRWj) * Xij]n

j = 1

Figure 2 Flowchart for the proposed method.

©20

07, A

SQ

Sharma v14 i4 9/20/07 10:11 AM Page 56

Ranking Customer Requirements in QFD by Factoring in Their Interrelationship Values

The prioritization weights are thencomputed by using the prelimi-nary raw weights. To factor in theCR correlation triangle values—which usually are in the form ofsymbols — the model starts onlyafter the prioritization of the CRsby rating them through their rawweights. This article considersQFD done on a healthcare prod-uct. It should be noted, however,that the whole QFD matrix prepa-ration and its calculation for theselected product are beyond thescope of this article. The legendsused are:

• n—Number of customer requirements

• PRW—Preliminary raw weight

• SPRW—Sum of all preliminary raw weights

• NPRW—Normalized preliminary raw weight

• IRW—Intermediate raw weight

• SIRW—Sum of all intermediate raw weights

• NIRW —Normalized intermediate raw weight

• FRW—New raw weight

• Xij—Interrelationship values among CRs, that is,between row i and column j

The interrelationship values are taken as 1.5 for“+ +”, 1.25 for “+”, 0.5 for “–”, 0.25 for “– –” and0.00 for “blanks.” The value conversion has been doneconsidering the synergistic or detrimental effect of theserelationships on various CRs under consideration. Thecalculations for ith (i = 1 to n) CR is given next:

NPRWi = PRWi / SPRW

IRWi = NPRWi +n

Σj=1

[NPRWi * (1 + NPRWj) * Xij], forall j ≠ i

These intermediate raw weights are then normal-ized, and these normalized values are distributedaccording to preliminary raw weights to give finalraw weights, as given next.

NIRWi = IRWi / SIRW

FRWi = NIRWi * SPRW

The rationale behind the devised formulae is therelative direction and the magnitude of change thatone CR causes on to the other, depicted with the help ofsynergistic or detrimental symbols. If there is a syner-gistic (or detrimental) relationship between two CRs,the effect is positive (or negative). This causes CRs tomove in the right (or opposite) direction, the intensityof which is decided by the nature of relationship, thatis, strong or weak. The net effect is to increase (that is,multiplying factor > 1.00) or decrease (that is, multi-plying factor < 1.00) the considered CR’s weight bythe related percentage value of other related CRs, thustaking into account the assigned degree and directionof association. A Visual Basic Code has been writtenfor the proposed method. A flowchart for the same isshown in Figure 2.

AN ILLUSTRATIVE EXAMPLEAfter going through all QFD steps depicted previously inthe basic building blocks for the selected healthcareproduct “syringe and needle,” the calculated rawweights for each CR are shown next. The matrix ignoresredundant steps and considers only the relevant ones,that is, CR, interrelationship among CRs, raw weights,normalized raw weights, and their ranks, as shown inFigure 3. The sample calculation (of CR2) shown nexthelps in deciphering how changes in the correlationtriangle value get reflected in IRW and finally in FRW,thus affecting the resultant rankings of the CRs.

www.asq.org 57

Customer Requirements PRW NPRW Rank

Cleanliness and purity 69 0.1076 IV

Safe, reliable, and efficient 72 0.1123 III

Ease of handling and use 40 0.0624 X

Cost of the product 86 0.1342 II

Leakage factor (air/liquid) 59 0.0920 VII

Right size/Correct volume 63 0.0983 VI

Proper visible markings 42 0.0655 IX

Precise movement 47 0.0733 VIII

Safe convenient packaging 67 0.1045 V

Tamper proof (no reuse) 96 0.1498 I

+

++

++

+

+

+

+++++

– –

– –

– ––

––

–

– ++

++

+

Figure 3 Customer requirements and their interrelationships.

©20

07, A

SQ

Sharma v14 i4 9/20/07 10:11 AM Page 57

58 QMJ VOL. 14, NO. 4/© 2007, ASQ

Ranking Customer Requirements in QFD by Factoring in Their Interrelationship Values

Considering row no. 2 (CR2: safe, reliable, andefficient)

PRW2 = 72.00 (Old rank is 3)

NPRW2 = 72.00 / 641.00 = 0.1123

X2j = {0.00, – , 0.00, 0.25, 1.25, 1.25, 1.25, 0.00,1.50, 1.50}

SPRW = 641.00

IRW2 = NPRW2 + {n

Σj=1

[NPRW2 * (1 + NPRWj) * X2j]} =

0.1123 + {0 + 0 + 0.1123 * (1 + 0.1342) *0.25 + 0.1123 * (1 + 0.0920) * 1.25 + 0.1123 * (1 + 0.0983) * 1.25 + 0.1123 * (1 + 0.0655) *1.25 + 0 + 0.1123 * (1 + 0.1045) * 1.50 +0.1123 * (1 + 0.1498) * 1.50} = 1.1367

Similarly, values of IRW for all other CRs arecomputed.

SIRW = 5.9455

NIRW2 = IRW2 / SIRW = 1.1367 / 5.9455 = 0.1912

FRW2 = NIRW2 * SPRW = 0.1912 * 641.00 = 122.5449(New rank is 1)

Similarly, values of FRW for all other CRs arecomputed (see Table 1).

RESULTS Results can be deciphered from the comparativeanalysis of CRs raw weights and their ranking (seeTable 2). The old traditional method and the resulting

Table 1 CR with correlation values, preliminary and final raw weights.

C R PRW NPRW IRW NIRW FRW

Cleanliness & purity – 1.25 0 0.50 0 0 0 0 1.25 1.25 69 0.1076 0.6217 0.1046 67.03

Safe, reliable,& efficient 1.25 – 0 0.25 1.25 1.25 1.25 0 1.50 1.50 72 0.1123 1.1367 0.1912 122.54

Ease of handling& use 0 0 – 0 0 0 1.25 1.25 0 0 40 0.0624 0.2292 0.0386 24.71

Cost of the product 0.50 0.25 0 – 0.50 0.50 0.50 0.25 0.50 0.25 86 0.1342 0.6128 0.1031 66.07

No leakages (Air/Fluid) 0 1.25 0 0.50 – 1.25 0 1.25 0 0 59 0.0920 0.5221 0.0878 56.29

Right size/Correct vol. 0 1.25 0 0.50 1.25 – 1.50 0 0 0 63 0.0983 0.5819 0.0979 62.74

Proper markings 0 1.25 1.25 0.50 0 1.50 – 0 0 0 42 0.0655 0.3887 0.0654 41.91

Precise movements 0 0 1.25 0.25 1.25 0 0 – 0 0 47 0.0733 0.2916 0.0490 31.44

Safe packaging 1.25 1.50 0 0.50 0 0 0 0 – 1.50 67 0.1045 0.6632 0.1115 71.50

Tamper proof (No reuse) 1.25 1.50 0 0.25 0 0 0 0 1.50 – 96 0.1498 0.8976 0.1510 96.77

SUM 641 1.000 5.9455 1.000 641.00

Clean

lines

s&

Purity

Safe,

Relia

ble,

& Eff

icien

tEa

se of

Han

dling

& Us

eCo

st of

thePr

oduc

tNo

Leak

ages

(Air/

Fluid)

Right

Size/

Corre

ct Vo

lume

Prop

erMa

rking

sPr

ecise

Move

ments

Safe

& Co

nven

ient

Pack

aging

Tamp

er Pr

oof

(No r

euse

possi

bility

)

©20

07, A

SQ

Sharma v14 i4 9/20/07 10:11 AM Page 58

Ranking Customer Requirements in QFD by Factoring in Their Interrelationship Values

rankings leave a lot to be desired. However, thislacuna gets solved with the computation of the finalraw weight obtained through CR correlation trianglevalues. The final raw weights, their rankings, andtheir order are much more precise and accurate,leading to better and informative decision making byproduct designers.

CONCLUSIONSThe prioritization and the judicious selection of CRsfrom the market’s perspective is a complex task andneeds special attention. Over the past few years, therehas been an huge effort in establishing a decisionframework that quantifies the imprecise and subjectivecustomer information inherent in the product plan-ning process. This research has proposed a novel QFDapproach that solves some methodological problemsencountered in the conventional QFD. While the con-ventional approach uses only symbols to express thestrength or weakness of the relationships among CRs,albeit only in few cases, the new approach addressesrelationships in a different and improved numericalmethod.

The article shows the effectiveness and precisenessof the proposed model. Numbers are used to representthe imprecise nature of the judgments and to define

more appropriately the relationships among CRs. Themodel is shown to be computationally feasible forrealistic problems and outperforms traditionalapproaches on the basis of being relatively straightforward and simplified. A systematic algorithm forcalculating the critical targets of CRs is followed. Theframework presented in this article for prioritizingCRs extends the single objective viewpoint of maxi-mizing customer satisfaction by considering the com-pany’s other market-related objectives, and thus,precludes an unreasonable QFD planning in practice.The proposed methodology provides product develop-ers, designers, and engineers with an objectivemethod for making a judicious selection of criticalCRs satisfying the overall CRs. The illustrated exam-ple clearly shows that the proposed approach canproduce more informative results, adding credibilityto the outcome and its analysis, for the benefit of theproduct designers and developers.

REFERENCES

Akao, Y. 1990. Quality function deployment: Integrating cus-tomer requirements into product design, Cambridge, Mass.:Productivity Press.

Armacost, R., P. Componation, M. Mullens, and W. Swart. 1994.AHP framework for prioritizing customer requirement in QFD: Anindustrialized housing application. IIE Transactions 26, no. 4: 72-79.

www.asq.org 59

Table 2 A comparative analysis of CRs preliminary and final raw weights and rankings.

C R (Conventional Method) PRW NPRW Rank C R (Proposed Method) FRW NIRW Rank

Tamper proof (no reuse) 96 0.1498 I Safe, reliable, and efficient 122.5449 0.1912 I

Cost of the product 86 0.1342 II Tamper proof (no reuse) 96.7727 0.1510 II

Safe, reliable, and efficient 72 0.1123 III Safe & convenient packaging 71.4992 0.1115 III

Cleanliness and purity 69 0.1076 IV Cleanliness and purity 67.0252 0.1046 IV

Safe convenient packaging 67 0.1045 V Cost of the product 66.0727 0.1031 V

Right size/Correct volume 63 0.0983 VI Right size/Correct volume 62.7384 0.0979 VI

No leakages (Air/Fluid) 59 0.0920 VII No leakages (Air/Fluid) 56.2857 0.0878 VII

Precise movements 47 0.0733 VIII Proper markings 41.9110 0.0654 VIII

Proper markings 42 0.0655 IX Precise movements 31.4355 0.0490 IX

Ease of handling and use 40 0.0624 X Ease of handling and use 24.7148 0.0386 X ©20

07, A

SQ

Sharma v14 i4 9/20/07 10:11 AM Page 59

60 QMJ VOL. 14, NO. 4/© 2007, ASQ

Ranking Customer Requirements in QFD by Factoring in Their Interrelationship Values

Chen Yizeng, R. Fung, and J. Tang. 2005. Rating technical attrib-utes in fuzzy QFD by integrating fuzzy weighted average methodand fuzzy expected value operator. European Journal ofOperational Research (Production, Manufacturing and Logistics).

Fung, R. Y. K., K. Popplewell, and J. Xie. 1998. An intelligenthybrid system for customer requirements analysis and productat tr ibute targets determinat ion. Internat ional Journal ofProduction Research 36, no. 1: 13-34.

Govers, C. 1996. What and how about quality function deploy-ment (QFD). International Journal of Production Economics:575-585.

Griffin, A., and J. Hauser. 1996. Integrating R & D and marketing:A review and analysis of the literature. Journal of ProductInnovation Management: 191-215.

Han Chang, Kim Jae, and Choi Sang Hyun. 2004. Prioritizingengineering characteristics in QFD with incomplete information:A linear partial ordering approach. International JournalProduction Economics 91: 235-249.

Kao, C., and S. Liu. 2001. Fractional programming approach tofuzzy weighted average. Fuzzy Sets and Systems 120: 435-444.

Karsak, E. 2004. Fuzzy multiple objective programming frame-work to prioritize design requirements in QFD. Computers &Industrial Engineering 47: 149-163.

Karsak, E., S. Sozer, and E. Alptekin. 2002. Product planning inQFD using a combined analytic network process and goal pro-gramming approach. Computers and Industrial Engineering 44:171-190.

Khoo, L., and N. Ho. 1996. Framework of a fuzzy quality functiondeployment system. International Journal of Production Research34, no. 2: 299-311.

Liu, B., and K. Liu. 2002. Expected value of fuzzy variable andfuzzy expected value models. IEEE Transactions on Fuzzy Systems10, no. 4: 445-450.

Lyman, D. 1990. Deployment normalization. In Transactions fromthe 2nd Symposium on Quality Function Deployment: 307-315.

Sharma, J., and A. Rawani. 2006. Customer driven productdevelopment through quality function deployment (QFD). AsiaPacific Business Review — A Journal of Asia-Pacific Institute ofManagement 2, no. 1: 45-51.

Sharma, J., D. Sharma, and A. Rawani. 2006. Quality driven prod-uct development. Manufacturing Engineer (IET). (June-July): 38-41.

Shen, X., K. Tan, and M. Xie. 2001. The implementation of qual-ity function deployment based on linguistic data. Journal ofIntelligent Manufacturing 12: 65-75.

Vangeas, L., and A. Labib. 2001. A fuzzy quality functiondeployment (FQFD) model for deriving optimum targets.International Journal of Production Research 39, no. 1: 99-120.

BIOGRAPHIES

J. R. Sharma is assistant professor at the Institute of ManagementTechnology, Nagpur. Before joining this organization, he servedas senior faculty in the Department of Production and MechanicalEngineering, Manoharbhai Patel Institute of Engineering andTechnology (MIET), Gondia, India. He is pursuing his doctoralresearch at the National Institute of Technology, Raipur. Hisresearch interests are in the areas of quality management andproduct development. Sharma has more than 20 research paperspublished in various national and international conferences, tech-nical periodicals, and journals. He has the honor of reviewingpapers for several international and national journals includingElsevier’s European Journal of Operations Research. He can bereached by e-mail at sjiten1@gmail.com.

A. M. Rawani is professor and head of the Department ofMechanical Engineering, National Institute of Technology, Raipur,India. He has more than 23 years of experience in academics.He acquired his doctorate from IIT-Delhi. He has pursued hisresearch in the area of strategic management. Rawani has donecommendable work in research and has more than 35 publicationsin national and international journals to his credit. He has been anominated member of the Confederation of Indian Industry(CII) — HR & Training Program Panel. He can be reached by e-mail at amrawani@rediffmail.com.

Sharma v14 i4 9/20/07 10:11 AM Page 60

ReviewsBook

China Now: Doing Businessin the World’s Most DynamicMarket 2007. N. Mark Lam andJohn L. Graham. New York: McGraw-Hill. 385 pages.

Globalization requires thatmanagers are aware of the nuancesof the business customs and char-acteristics of major internationalmarkets. China Now provides anexcellent primer on the emergingChinese business culture with aspecific focus on the techniques ofnegotiating in various areas of thecountry. Lam and Graham haveproduced a suburb overview of thebusiness culture that exists inChina today and how to effectivelycope with it in negotiations.

As an illustration of how nego-tiations work in Shanghai, thebook begins with an anecdotal taleof the attempts of the Americanautomotive industry to break intothe Chinese market. It describeswhy the Ford Taurus did not getthe Chinese checkered flag whileGeneral Motors’ (GM) Buicks did.GM’s negotiator, Shirley Young, issaid to have “not only brought amarketing imagination to the GMteam, but she also brought greatgaunxi (connections). (p.5) Sheunderstood that for Chinese, face-to-face meetings and relationshipsbetween people at all levels are theessential elements of successfulnegotiations.

Part 1, “The Necessary Back-ground for Negotiations withChinese Business People,” providesinformative tidbits of informationabout China, beginning with achapter on the history and cultureof the Chinese people. Lam andGraham have done an excellentjob of providing salient pointsfrom the 5,000 years of Chinesehistory and enormous diversity ofits immense population. Certainly,the 18 things they have selected askey things one needs to know con-cerning key people, turning points,and prominent themes are worthyof investigation.

The next area explored con-cerns the economic development ofChina. The authors enthusiasm forChina is found in their statement:“There is no more important sin-gle market than China…China’sdual economic system, embracingsocialism along with many tenetsof capitalism, has produced aneconomic boom with expandedopportunity for foreign investmentthat has resulted in annual grossnational product (GNP) averag-ing nearly 10 percent since 1970.”(p. 43) China is said to be “neitheran economic paradise nor an eco-nomic wasteland. It is a relativelypoor nation going though apainfully awkward transformationfrom a socialist market system to ahybrid socialist/free system, not yet

complete and with rules of thegame still being written.” (p. 56)

This is especially evident in thebook’s concluding chapter onChinese legal and business envi-ronment. The Chinese legal systemis not as well defined as theWestern system nor is the legalprofession. In China, businessesare driven by technically trainedindividuals rather than lawyers. Infact, today China has about115,000 lawyers, and being alawyer is not a prerequisite forserving as a judge. There is a con-stitution, civil, and criminal code,as well as laws and regulationscovering commerce in general anda variety of specific business con-cerns. Because of the developingnature of Chinese law most dis-putes are settled by negotiationrather than adjudication becausethe outcome is unpredictableunder the best of circumstances.

Part 2 explores what happenswhen Americans meet Chineseacross the negotiation table.Americans are characterized ashaving a free-enterprise, cowboystyle of negotiations. On the otherhand, the elements of the Chinesestyle of business negotiation are:

• Guanxi (personal connections)

• Mianzi (face or social capital)

• Shehui Dengji (social hierarchy)

• Renji Hexie (interpersonal harmony)

www.asq.org 61

Book Reviews v14 i4 9/20/07 10:11 AM Page 61

Reviews

• Qundai Guanxi (nepotism)

• Zhengti Guannian (holisticthinking)

• Chiku Nailao (endurance)

• Jiejian (thrift)

• Linghe Tanpan (zero-sumnegotiations)

• Jiao Ta Liangshi Chuan (threat-ening to do business elsewhere)

The remaining chaptersdescribe the “how to” of preparing,conducting, and following up afternegotiations. Each provides help-ful tips on what to do and not to doto assure success in the negotiationprocess.

Every region is not the same inChina, so each of the generic point-ers has to be modified for the specificoffice in which the negotiations takeplace. Part 3 illustrates each of thegeneral areas of China and theirrespective cultural differences.

The book’s final two parts pre-sent brief discussions of the issues ofnegotiating and enforcing intellec-tual property rights and speculatingabout the future of Chinese andAmerican commercial relations. Atsome point a quality issue willdevelop that requires skillful negoti-ating to resolve a disagreement witha supplier, outsource producer, orcustomer. The insights presented inChina Now by Lam and Grahamwill provide an excellent road mapto a successful resolution of thedifferences, perceived or real.

Reviewed byDr. James KohnenSt. Mary’s College of California

Made to Stick: Why SomeIdeas Survive and Others Die2007. Chip Heath and Dan Heath.New York: Random House. 291pages.

Made to Stick is a wonderfulbook designed to entertain, inform,and enlighten anyone interested ineffectively communicating an idea.Chip Heath and Dan Heath havesynthesized their perception of howto do this for organizations, groups,and individuals into six succinctprinciples (SUCCES):

1. Simplicity

2. Unexpectedness

3. Concreteness

4. Credibility

5. Emotions

6. Stories

Each point is thoroughly pre-sented in a chapter of the bookthat is effectively explained withsupporting data, anecdotal informa-tion, and experiential findings. TheIntroduction and Epilogue provideadded material that contributes tothe book’s readability. A sectioncalled “The Easy Reference Guide”at the end of the book further sup-ports the material by providing acomprehensive outline that jogsthe reader’s memory of the topicspresented.

Stories seem to be the commondenominator for making someideas survive and others die,because they provide the context inwhich fact and fiction can beweaved to illustrate a point. A goodstory will be remembered long

after the specifics are lost, modified,or completely replaced. The authorssay, “A story is powerful because itprovides the context missing fromactive prose.” (p. 214)

The point of developing ideasthat incorporate the SUCCES prin-ciples is said to be to “beat thecurse of knowledge” which inhibitsembracing new ideas. (p. 20) Thewell-developed notes at the end ofthe book document the researchthat supports this theory, whichsays it “increases as people gainmore expertise.” So the more a per-son knows about something, themore difficult it is for him or her toaccept a new idea that challengesor changes what he or she knowsabout a topic. (p. 260-261)

The Heath brothers presenttheir profound concepts of makingideas survive in almost a playfulfashion. The solid academicresearch that supports their theoryarticulated as SUCCES principles ishidden in each of the tales theyspin to make point after point topresent a creditable case for it. Theurban legends, business anecdotes,and findings from their own inde-pendent research provide a richbackground that draws the readerinto their world. Using the meta-phor developed in each story theyare able to build a bond of believ-ability with the reader, which is notoften seen in management texts.Reading this book is truly adelightful experience.

Because the information pre-sented is universal, the book has a

62 QMJ VOL. 14, NO. 4/© 2007, ASQ

Book Reviews v14 i4 9/20/07 10:11 AM Page 62

uniquely wide appeal to organiza-t ions of al l types and to theirmembers at all levels. The Heathbrothers keep their word when theysay: “We can promise you this:Regardless of you level of ‘naturalcreativity,’ we will show you how ali t t le focused ef fort can makealmost any idea stickier, and asticky idea is an idea that is morelikely to make a dif ference.” (p. 24)

Reviewed byDr. James KohnenSt. Mary’s College of California

Managing Without Walls:Maximize Success withVirtual, Global, and Cross-Cultural Teams Colleen Gartonand Kevin Wegryn. 2006. Lewisville,Texas: McPress. 409 pages.

As virtual offices become morecommon for quality professionals,the concepts presented by Gartonand Wegryn in Managing WithoutWalls become more and more ger-mane. The idea of virtual officesremoves the collective place forthe value-added activi ty thatdefines the work being done indi-vidually or collectively by a team.When meaningful work supportingthe common goals of an organiza-tion can be performed anyplace, atany t ime, management of theindividuals engaged in it entersinto a new dimension of time andspace.

Employees engaged in thiskind of working relationship arenot traveling when they are away

from the center of mass of theorganization, indicated by i tsbusiness address; rather, they arein an organization that clearlyconsiders their actual global loca-tion exactly the same as if it werelocated at its stated businessaddress. In this virtual environmentaudio, visual, and printed commu-nication is possible universally andinstantaneously regardless of theindividual’s actual location at thegiven moment.

This emerging phenomenoncreates a huge paradigm shift in theway workers and work is managed.Organizational norms of behaviorare no longer transferred by indi-vidual and group interaction at theworkplace. The informal commu-nication chain takes on greatersignificance when it is transmittedthrough cyberspace, mainly becauseit is done without body languageor other visual indicators to clarifyits total meaning.

After justifying the need for vir-tual management, they describethe desired virtual manager’s idealskill set that includes: generalmanagement, people skills, com-munications, technical knowl-edge, decision making, problemsolving, team orientation, andadministration, which are alsoheld by tradit ional managers.They emphatically state, however,that not all traditional managerscan adapt to the virtual arena.They cite some obvious things avirtual manager must cope with todevelop effective virtual teams such

as time, language, and culturaldifferences that must be consideredto be effective. Then they move toan informative discussion relatedto time management and the needfor team processes and proceduresthat standardize the efforts of themembers of the team. The insight-ful reviews of vir tual poli t ics ,managing conflict, and virtualmanagement of high-risk and cata-strophic events are very relevant tothe management challenges foundin organizations today. The bookconcludes by presenting sectionsdescribing virtual managementand communication tools andleadership. One of the specific lead-ership points stressed by theauthors relates to controlling one’semotions. They say, “Staying calmand unemotional are the keys tosuccessfully managing conflict andhighly emotional situations. Wheneveryone else is stressed out, angry,and frustrated it takes just one per-son remaining calm and being thevoice of reason to bring everyoneelse back to resembling normal,healthy human beings! Strive tobe that person.” (p. 354)

Managing Without Walls is anexcellent introduction to an emerg-ing form of organizational structureand behavior found in global busi-nesses today. Garton and Wegrynhave done a commendable jobidentifying the idiosyncrasies ofmanaging in this environment.

Reviewed byDr. James KohnenSt. Mary’s College of California

www.asq.org 63

Book Reviews v14 i4 9/20/07 10:11 AM Page 63

Reviews

Project Sponsorship:Achieving Management Com-mitment for Project SuccessRandall L. Englund and AlfonsoBucero. 2006. San Francisco: Jossey-Bass/Business & ManagementSeries. 201 pages.

Most organizations employ someform of project management toaccomplish special tasks outside thescope of their normal operations orto serve the specific needs of aunique customer base. Traditionally,a project team led by a projectmanager is assigned to the task toassure that the elements of cost,schedule, and quality are fulfilledto meet the organization’s andcustomer’s expectations. Englundand Bucero make a convincingargument for having a projectsponsor to champion the projectwith the organization. Their con-cern is that many executives areassigned as project sponsor with-out a full understanding of theirexpected roles and responsibilitiesduring project life cycles.

The authors offer an intriguing10-step model that outlines theprogress of sponsor development,from responsibilities to knowledgemanagement. Bucero providesinformative experiences he has hadin the field as antidotal evidence

for each step. These generally beginwith the phrase “I (Bucero)” andcontinue as first person accounts ofsome event that he has either par-ticipated in or witnessed during hiscareer.

The authors make full use ofthe Project Management Body ofKnowledge published by theProject Management Institute(PMI) throughout the book. InChapter 2, “Obtaining a ProjectSponsor,” they cite the “Top 10Attributes of a Great ProjectSponsor” presented by LonniePacelli at the 2005 PMI GlobalCongress. (pp. 42-43)

• Clearly understand the problemto be solved.

• Ensure that the solution fixesthe problem.

• Know where “good enough” is.

• Build the right team to solvethe problem.

• Hold the team accountable forresults.

• Know the big issues and what isneeded to resolve them.

• Be the advocate, coach, influ-encer, and battering ram.

• Make the thoughtful, tough,decisions.

• Ensure that the project finishesstrong.

• Know when to pull the plug.

In the final chapter of the book,“Knowledge Management,” theauthors stress the importance ofreflecting periodically on thelessons learned. Periodic projectreviews are essential for the pro-ject sponsor to capture not onlythe status of the project but alsoways to improve its execution. Thefour steps in the process include:preparing the session, conductingthe session, collecting learning,and sharing the lessons learned.According to the authors, theimportance of the project sponsorsharing the outcomes of thesemeetings with others cannot beoverstated in teams of benefits tothe organization.

Englund and Bucero have donea creditable job of defining thevalue of having a project sponsorto achieve management commit-ment for project success. The roleof the project sponsor and theimportance of formal developmentprograms are clearly articulated ina very succinct method that is wor-thy of anyone’s attention involvedin project management.

Reviewed byDr. James KohnenSt. Mary’s College of California

64 QMJ VOL. 14, NO. 4/© 2007, ASQ

Book Reviews v14 i4 9/20/07 10:11 AM Page 64

A

Al-Marsumi, Mujbil, “Total QualityManagement in a Chosen Sectionof the Hospitals in Amman, Jordan”(vol. 14, no. 1): 45–57.

C

Cheng, Jung-Lang, “Six Sigma and TQMin Taiwan: An Empirical Study”(vol. 14, no. 2): 7–18.

Collins, Kevin F., and Senthil K.Muthusamy, “Applying the ToyotaProduction System to a HealthcareOrganization: A Case Study on aRural Community HealthcareProvider” (vol. 14, no. 4): 41–52.

Costa, Micaela Martínez, and ÁngelRafael Martínez Lorente, “ISO9000:2000: The Key to Quality? AnExploratory Study” (vol. 14, no. 1):7–18.

D

de Koning, Henk, and Jeroen de Mast,“The CTQ Flowdown as a Concep-tual Model of Project Objectives”(vol. 14, no. 2): 19–28.

de Mast, Jeroen. See de Koning, Henk

F

Foster, S. Thomas, “Does Six SigmaImprove Performance?” (vol. 14,no. 4): 7–20.

Frazier, Gregory V. See Masters, Brad

G

Guimaraes, Tor, D. Sandy Staples, andJames McKeen, “Assessing theImpact From Information SystemsQuality” (vol. 14, no. 1): 30–44.

H

Hillmer, Steve, and Canan Kocabasoglu,“Improving Quality and Productivityvia Stratification: A Call CenterExample for Forming Homoge-neous Employee Groups” (vol. 14,no. 1): 19–44.

L

Lorente, Ángel Rafael Martínez. SeeCosta, Micaela Martínez

M

Masters, Brad, and Gregory V. Frazier,“Project Quality Activities and GoalSetting in Project PerformanceAssessment” (vol. 14, no. 3): 25–35.

McKeen, James. See Guimaraes, Tor

Muthusamy, Senthil K. See Collins,Kevin F.

O

Ozgur, Ceyhun. See Pinar, Musa

P

Pasupathy, Kalyan Sunder, and Konstantinos P. Triantis, “A Frame-work to Evaluate Service Operations:Dynamic Service-Profit Chain”(vol. 14, no. 3): 36–49.

Pinar, Musa, and Ceyhun Ozgur, “TheLong-Term Impact of ISO 9000Certification on Business Perfor-mance: A Longitudinal Study UsingTurkish Stock Market Returns”(vol. 14, no. 4): 21–40.

R

Rao, K. S. P. See Saravanan, R.

Rawani, A. M. See Sharma, J. R.

Roelofse, Frederick, “An ExploratoryStudy into the MultidimensionalNature of Quality in Analytical Lab-oratories: Managerial Implications”(vol. 14, no. 3): 7–14.

S

Saravanan, R., and K. S. P. Rao, “ServiceQuality From the Customer’s Per-spective: An Empirical Investigation”(vol. 14, no. 3): 15–24.

Senol, Sanslı, “Poisson Process Approachto Determine the Occurence Degreein Failure Mode and Effect ReliabilityAnalysis” (vol. 14, no. 2): 29–40.

Sharma, J. R., and A. M. Rawani,“Ranking Customer’s Requirementsin QFD by Factoring in Their Inter-relationship Values” (vol. 14, no. 4):53–60.

Staples, D. Sandy. See Guimaraes, Tor

T

Triantis, Konstantinos P. See Pasupathy,Kalyan Sunder

www.asq.org 1

IndexQMJ Volume 14 Author

Annual Index v14 i4 9/20/07 10:20 AM Page 1

Automobile SectorSaravanan, R., and K. S. P. Rao, “Service

Quality From the Customer’s Per-spective: An Empirical Investigation”(vol. 14, no. 3): 15–24.

Critical to Qualityde Koning, Henk, and Jeroen de Mast,

“The CTQ Flowdown as a Concep-tual Model of Project Objectives”(vol. 14, no. 2): 19–28.

Customer SatisfactionPasupathy, Kalyan Sunder, and

Konstantinos P. Triantis, “A Frame-work to Evaluate Service Operations:Dynamic Service-Profit Chain”(vol. 14, no. 3): 36–49.

Sharma, J. R., and A. M. Rawani,“Ranking Customer’s Requirementsin QFD by Factoring in Their Inter-relationship Values” (vol. 14, no. 4):53–60.

Financial OutcomesFoster, S. Thomas, “Does Six Sigma

Improve Performance?” (vol. 14,no. 4): 7–20.

FMEASenol, Sanslı, “Poisson Process Approach

to Determine the Occurence Degreein Failure Mode and Effect ReliabilityAnalysis” (vol. 14, no. 2): 29–40.

Functional QualityRoelofse, Frederick, “An Exploratory

Study into the MultidimensionalNature of Quality in Analytical Lab-oratories: Managerial Implications”(vol. 14, no. 3): 7–14.

HealthcareAl-Marsumi, Mujbil, “Total Quality

Management in a Chosen Sectionof the Hospitals in Amman, Jordan”(vol. 14, no. 1): 45–57.

Collins, Kevin F., and Senthil K.Muthusamy, “Applying the ToyotaProduction System to a HealthcareOrganization: A Case Study on aRural Community HealthcareProvider” (vol. 14, no. 4): 41–52.

Information SystemsGuimaraes, Tor, D. Sandy Staples, and

James McKeen, “Assessing theImpact From Information SystemsQuality” (vol. 14, no. 1): 30–44.

ISO 9000Costa, Micaela Martínez, and Ángel

Rafael Martínez Lorente, “ISO9000:2000: The Key to Quality? AnExploratory Study” (vol. 14, no. 1):7–18.

Pinar, Musa, and Ceyhun Ozgur, “TheLong-Term Impact of ISO 9000Certification on Business Perfor-mance: A Longitudinal Study UsingTurkish Stock Market Returns”(vol. 14, no. 4): 21–40.

Measurementde Koning, Henk, and Jeroen de Mast,

“The CTQ Flowdown as a Concep-tual Model of Project Objectives”(vol. 14, no. 2): 19–28.

Guimaraes, Tor, D. Sandy Staples, andJames McKeen, “Assessing theImpact From Information SystemsQuality” (vol. 14, no. 1): 30–44.

Performance AssessmentCheng, Jung-Lang, “Six Sigma and TQM

in Taiwan: An Empirical Study”(vol. 14, no. 2): 7–18.

Masters, Brad, and Gregory V. Frazier,“Project Quality Activities and GoalSetting in Project PerformanceAssessment” (vol. 14, no. 3): 25–35.

Performance ImprovementFoster, S. Thomas, “Does Six Sigma

Improve Performance?” (vol. 14,no. 4): 7–20.

Hillmer, Steve, and Canan Kocabasoglu,“Improving Quality and Productivityvia Stratification: A Call CenterExample for Forming Homoge-neous Employee Groups” (vol. 14,no. 1): 19–44.

Pinar, Musa, and Ceyhun Ozgur, “TheLong-Term Impact of ISO 9000Certification on Business Perfor-mance: A Longitudinal Study UsingTurkish Stock Market Returns”(vol. 14, no. 4): 21–40.

2 QMJ VOL. 14, NO. 4/© 2007, ASQ

IndexQMJ Volume 14 Subject

Annual Index v14 i4 9/20/07 10:20 AM Page 2

Project ManagementMasters, Brad, and Gregory V. Frazier,

“Project Quality Activities and GoalSetting in Project PerformanceAssessment” (vol. 14, no. 3): 25–35.

Quality Function DeploymentSharma, J. R., and A. M. Rawani,

“Ranking Customer’s Requirementsin QFD by Factoring in Their Inter-relationship Values” (vol. 14, no. 4):53–60.

Reliability AnalysisSenol, Sanslı, “Poisson Process Approach

to Determine the Occurence Degreein Failure Mode and Effect ReliabilityAnalysis” (vol. 14, no. 2): 29–40.

Service QualityPasupathy, Kalyan Sunder, and

Konstantinos P. Triantis, “A Frame-work to Evaluate Service Operations:Dynamic Service-Profit Chain”(vol. 14, no. 3): 36–49.

Roelofse, Frederick, “An ExploratoryStudy into the MultidimensionalNature of Quality in Analytical Lab-oratories: Managerial Implications”(vol. 14, no. 3): 7–14.

Saravanan, R., and K. S. P. Rao, “ServiceQuality From the Customer’s Per-spective: An Empirical Investigation”(vol. 14, no. 3): 15–24.

Six SigmaCheng, Jung-Lang, “Six Sigma and TQM

in Taiwan: An Empirical Study”(vol. 14, no. 2): 7–18.

de Koning, Henk, and Jeroen de Mast,“The CTQ Flowdown as a Concep-tual Model of Project Objectives”(vol. 14, no. 2): 19–28.

Foster, S. Thomas, “Does Six SigmaImprove Performance?” (vol. 14,no. 4): 7–20.

StratificationHillmer, Steve, and Canan Kocabasoglu,

“Improving Quality and Productivityvia Stratification: A Call CenterExample for Forming Homoge-neous Employee Groups” (vol. 14,no. 1): 19–44.

Technical QualityRoelofse, Frederick, “An Exploratory

Study into the MultidimensionalNature of Quality in Analytical Lab-oratories: Managerial Implications”(vol. 14, no. 3): 7–14.

Total Quality ManagementAl-Marsumi, Mujbil, “Total Quality

Management in a Chosen Sectionof the Hospitals in Amman, Jordan”(vol. 14, no. 1): 45–57.

Cheng, Jung-Lang, “Six Sigma and TQMin Taiwan: An Empirical Study”(vol. 14, no. 2): 7–18.

Costa, Micaela Martínez, and ÁngelRafael Martínez Lorente, “ISO9000:2000: The Key to Quality? AnExploratory Study” (vol. 14, no. 1):7–18.

Masters, Brad, and Gregory V. Frazier,“Project Quality Activities and GoalSetting in Project PerformanceAssessment” (vol. 14, no. 3): 25–35.

Toyota Production SystemCollins, Kevin F., and Senthil K.

Muthusamy, “Applying the ToyotaProduction System to a HealthcareOrganization: A Case Study on aRural Community HealthcareProvider” (vol. 14, no. 4): 41–52.

www.asq.org 3

Subject Index

Annual Index v14 i4 9/20/07 10:20 AM Page 3

Cover spine v14 i4 9/20/07 10:12 AM Page 2