Putting Best Practices to Work www.qualityprogress.

com | June 2008

QUALITY PROGRESS
VOC Key to
Nonprot
Baldrige
Wins p. 28

Clear
View
Bringing customer loyalty
into focus through new Plus:
measurement methods p. 20 ASQs Stance
On Voting
Reform
p. 14

Residual
Analysis:
A Refresher
p. 72
 Its not like
Im saving
lives here. Im
just improving
a process
that does.

QUALITY COMPANION 2 BY MINITAB TM

Around here, they call me the Doctor of Quality Improvement.

So whats my secret? Quality Companion by Minitab. With

Quality Companion I can organize all my Six Sigma and quality

improvement projects. It gives me the tools I need and saves me

time, so I can concentrate on my most important taskquality

improvement. Its like this: Quality Companion makes me

better, I make our staff better, and our staff makes our patients

better. The doctor is in. Visit minitab.com today or call

1.800.448.3555 (US/Canada/Mexico).

MINITAB and all other trademarks and logos for the Companys products and services are the exclusive property of Minitab Inc.
All other marks referenced remain the property of their respective owners. See minitab.com for more information.
Dr. Donald J. Wheeler is, in my mind,
the author to turn to. He has reached
the pinnacle of understanding by careful research.
W. Edwards Deming

Dr. Wheelers books are known

for their combination of unparalleled insight,
understanding, and practical advice,
presented with unsurpassed clarity, simplicity,
and accessibility.

His latest books continue this tradition.

Get your copies today.

For more information about our seminars and books, visit www.spcpress.com.
SPC Press Statistical Process Controls, Inc. Knoxville, Tennessee, USA 800-545-8602
 Putting Best Practices to Work | June 2008 | www.qualityprogress.com

Contents
FEATURES
20 CUSTOMER LOYALTY
20
The True Test of Loyalty
Measuring customer loyalty related to a product or service through advo-
cacy, purchasing and retention indexes can help companies build stronger,
longer relationships with customers.
by Bob E. Hayes

28 BALDRIGE AWARD
Now Hear This
Looking to improve their operations, two nonprot organizationsa U.S.
Army division and a municipal government in Floridaturned to voice of
the customer to drive improvement efforts. Their impressive results were
acknowledged with 2007 Baldrige awards.
by Brett Krzykowski

36 TEAMS
Incredible Journey
Quality methods and tools helped one Boeing team pinpoint risk and
safety concerns at a manufacturing site. The team implemented a safety
management system and helped prevent injuries and reduce workers
compensation costs.
by Nicole Adrian

42 SARBANES-OXLEY ACT
Two Are Better Than One ONLY @
www.qualityprogress.com
To comply with SOX, organizations neednt
reinvent the wheel. ISO 9001 offers lever-
age in helping organizations achieve SOX Exclusive Content
compliance.
Additional gures for Expert
by Vivek Vic Nanda Answers and Whos
Keeping Score? Sidebar of
personal narratives for
PROCESS IMPROVEMENT
48 Whos Keeping Score?
Incredible Journey.

A division of Sears developed its own React and Rate

scorecard to keep close tabs on how the Comment on this months
retailers regions and managers were
columns and articles.
complying with its quality management
system.
Get Out the Vote
28 by Rebecca M. Neenan
Take the monthly Quick Poll.

Tools and Resources

Basic quality, including a glossary.
 DEPARTMENTS

12
8 Inbox
Media misses mark on Juran.
Improving audits with a process
approach.

Expert Answers
Calculating DPMO.
14 QP
Mail
Quality Progress/ASQ
QUALITY PROGRESS
QUALITY PROGRESS

600 N. Plankinton Ave.

Milwaukee, WI 53203
Origins of sampling plan. Telephone Fax
800-248-1946 414-272-1734
414-272-8575
14 Keeping Current
E-mail
Votes for election reform. Follow protocol of rst initial and full last
Olympic-sized concerns over food. name followed by @asq.org (for example,
vfunk@asq.org).

Mr. Pareto Head Article Submissions

19 Quality Progress is a peer-reviewed publica-
tion with 85% of its feature articles written
78 QP Toolbox by quality professionals. For information
about submitting an article, call Valerie Funk
at 800-248-1946 x7373, or e-mail
80 QP Reviews manuscripts@asq.org.

82 QP Calendar Free QP Live

Subscribe to our free electronic newsletter,
QP Live, for a summary of each issues

COLUMNS contents. Visit www.asq.org/keepintouch.

html, or contact ASQ customer care at help@
asq.org.
6 Upfront 67 Standards Outlook Photocopying Authorization
A fresh perspective. Seven things to remember when
Authorization to photocopy items for
initiating an audit.
internal or personal use or the internal or
3.4 per Million personal use of specic clients is granted by
61 Career Corner Quality Progress, provided the fee of $1 per
Is your data dependable? 70 Ready your rsum.
copy is paid to ASQ or the Copyright Clear-
ance Center, 222 Rosewood Drive, Danvers,
MA 01923, 978-750-8400. Copying for other

76 72 Statistics Roundtable
A refresher on residual analysis.
purposes requires the express permission
of Quality Progress. For permission, write
Alice Haley, PO Box 3005, Milwaukee, WI
53201-3005, call 414-272-8575 x7406, fax
Quality in the First Person 414-272-1734 or e-mail ahaley@asq.org.
76
Navy experience leads to one mans
career in quality. Photocopies, Reprints
And Microform
Article photocopies are available from
88 One Good Idea ASQ at 800-248-1946. To purchase bulk
Build a probability plot with Excel. reprints (more than 100), contact Barbara
Mitrovic at ASQ, 800-248-1946. For micro-
form, contact ProQuest Information and
Learning, 300 N. Zeeb Road, Ann Arbor, MI
53 Special Advertising 48106, 800-521-0600 x2888, international
Section: ASQs Software 734-761-4700, www.il.proquest.com.
Directory and Buyers Guide
Membership and Subscriptions
For more than 60 years, ASQ has been the
worldwide provider of information and learn-
ing opportunities related to quality. In addi-
NEXT MONTH tion, ASQ membership offers information,
networking, certication and educational
- THE COST TO COMPETE - 10 RULES OF AUDITING opportunities to help quality profession-
Feigenbaum on quality in the global Guidelines to improve the process. als obtain practical solutions to the many
marketplace. problems they face each day. Subscriptions
to Quality Progress are one of the many
benets of ASQ membership. To join, call
800-248-1946 or see information and an
ASQs Vision: By making quality a global priority, an organizational imperative and application on p. 66 of this issue.
a personal ethic, the American Society for Quality becomes the community for everyone
who seeks quality technology, concepts or tools to improve themselves and their world. List Rentals
ASQs member and nonmember buyer
lists can be purchased by contacting Rose
Quality Progress (ISSN 0033-524X) is published monthly by the American Society for Quality, 600 N. Plankinton Ave., Milwaukee, WI 53203.
Editorial and advertising ofces: 414-272-8575. Periodicals postage paid at Milwaukee, WI, and at additional mailing ofces. Institutional DeLuca at the Walter Karl List Management
subscriptions are held in the name of a company, corporation, government agency or library. Requests for back issues must be prepaid and Co., 845-732-7019 or fax 845-620-1885.
are based on availability: ASQ members $15 per copy; nonmembers $23 per copy. Canadian GST #128717618, Canadian Publications Mail
Agreement #40030175. Canada Post: Return undeliverables to 2835 Kew Drive, Windsor, ON N8T 3B7. Prices are subject to change without
prior notication. 2008 by ASQ. 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 Canadian and international addresses.
Postmaster: Please send address changes to the American Society for Quality, PO Box 3005, Milwaukee, WI 53201-3005. Printed in USA.

4 QP www.qualityprogress.com
The Balancing Act of
Handling Industry Challenges
Complex manufacturing, supplier
and distribution management,
increased regulatory scrutiny,
the need to streamline
processes and much more.
Improving the Quality of your Business
Trust the industry
leader in compliance
and quality management
software solutions.
Streamline complex regulatory
processes, achieve control, improve
quality and reduce risk, while optimizing
your operations, cutting costs associated
with production, and improving delivery
and time to market.
TrackWise is the only industry
enterprise software which provides a
fully integrated solution, implementing
all key quality and compliance areas:
Deviations / Investigations
Corrective and Preventive Actions
Customer Complaints
Regulatory Reporting
Change Control / Management
Audits / Self Assessments
and Supplier Quality
Document Management
Employee Qualification
and Training Management
U.S. International
732.203.0400 +972.3.755.4040
info@sparta-systems.com info-europe@sparta-systems.com
 UPFRONT
A Fresh QP QUALITY PROGRESS

Perspective PUBLISHER
William A. Tony

EDITOR
Seiche Sanders

The value of stepping outside the everyday ASSOCIATE EDITOR

Mark Edmund

ASSISTANT EDITOR
ATTENDING A CONFERENCE geared toward your professional interests is kind of like Brett Krzykowski
getting your teeth cleaned. On both occasions, you walk away feeling refreshed, reinvigo- MANUSCRIPT COORDINATOR
rated, andif youre luckywith a smile on your face. Valerie Funk

Such was the case for me when I returned from ASQs World Conference on Quality EDITOR AT LARGE
Susan E. Daniels
and Improvement, held in May in Houston.
I returned to work with many new ideas and a renewed sense of purpose. Here are CONTRIBUTING EDITOR
Nicole Adrian
just a few highlights from my three days in Houston:
Keynote speakers: This years lineup was phenomenal. I especially enjoyed astronaut COPY EDITORS
Susan Gronemus
T.K. Ken Mattinglys talk. He recounted the awesome adventure in teamwork that helped Kelly Sullivan
him lead the Apollo 13 space missions crew safely back to earth. But, his story had a greater ART DIRECTOR
theme: the importance of having a clear-cut goal and aligning resources with that goal. Mary Uttech

General sessions: I attended a few sessions based on the conference theme, GRAPHIC DESIGNER
Sandy Wyss
Generation Quality. Speakers offered some great tips on motivating and leading teams
comprised of individuals of all ages. Leaders will get better results if they recognize and PRODUCTION
Cathy Schnackenberg
acknowledge the distinct differences among individuals in each age group.
Networking: Best of all, I was able to talk with QP readers from around the world. ADVERTISING PRODUCTION
Barbara Mitrovic
When you rely on readers opinions to steer important decisions related to the magazine,
DIGITAL PRODUCTION SPECIALISTS
face time is priceless. Eric Berna, Laura Franceschi
Feedback: Finally, I led a focus group that allowed readers to critique QP and www.
MEDIA SALES MANAGER
qualityprogress.com. What helpful feedback I received! In the coming months, youll no- Erica Gumieny
tice some of the suggestions incorporated into the magazine and website. This voice-of-
ACCOUNT EXECUTIVES
the-customer feedback will help me continue to make QP what you want it to be: a tool Angela M. Mitchell
Mitchell Pezanoski
to help you improve and excel in your professions.
If youd like to comment on how were doing, or if were missing the mark in some CLASSIFIED/RECRUITMENT
ADVERTISING
way, youre always encouraged to e-mail me at editor@asq.org. Ramona Garcia
For more conference highlights, turn to Conference Roundup, p. 16. MARKETING ADMINISTRATOR
Matt Meinholz

This months cover story, The True Test of Loyalty, p. 20, gives insight into innova- EDITORIAL AND ADVERTISING OFFICES
Phone: 414-272-8575
tive, more accurate ways of measuring customer satisfaction. A clearer view of loyalty
Fax: 414-272-1734
can be found by measuring advocacy, purchasing and retention.
ASQ ADMINISTRATION
The 2007 recipients of the Malcolm Baldrige National Quality Award in the nonprot Executive Director
category sought and studied voice-of-the-customer feedback, too. In Now Hear This, Paul E. Borawski

p. 28, read about how ARDECan Army weapons R&D centerand the municipality of Managing Directors
Christopher D. Bauman
Coral Springs, FL, focused on voice of their customers to make improvements that saved Brian J. Lehouillier
not only money, butmore importantlyhuman lives. Michelle Mason
Laurel Nelson-Rowe

To promote discussion of issues in the eld of quality and

ensure coverage of all responsible points of view, Quality
Progress publishes articles representing conicting and minor-
ity views. Opinions expressed are those of the authors and not
necessarily of ASQ or Quality Progress. Use of the ASQ logo in
Seiche Sanders
advertisements does not necessarily constitute endorsement of
Editor that particular product or service by ASQ.

6 QP www.qualityprogress.com
 INBOX
Remembering Juran
Putting Best Practices to Work
I remember well the day Dr.
W. Edwards Deming died. I re-
member how sad I was that
there was so little said in this
country about the passing of Joseph M. Juran
this great man who had done
so much for so many. The way I
found out Dr. Joseph Juran had
died left me a great deal sad-
der. It was not until I received the April issue
of Quality Progress, a month after Dr. Juran
died, that I came to know he was gone.
I read a variety of online and print
publications for my news, including the
websites of USA Today and CNN, as well as
Time magazine. I also read the local news-
paper and watch the local TV news here in
Boston, as well as the national TV news.
Not one of those media outlets brought
the story to my attention. Maybe I was
having a bad month. Maybe I was asleep at
the switch. So I decided to ask some of my
friends in the quality profession whether
they had heard or seen anything. To a per-
son, they had not.
I have spent almost 30 years of my life
working on the front lines of quality, forever
reminding people how important quality is.
I have listened to and lived through every
avor of quality that has come along in that
time. The importance of quality has been
inescapable, right? So tell me how I am
to understand that, for an entire month, it
was as if Dr. Joseph Juran and all he did for
quality in the world didnt matter at all.
ASQ, have we been fooling ourselves?
There are reasons to believe we have.
With sadness and sympathy for the
Juran family,
Paul Aganski
Field quality engineer
West Newbury, MA
8 QP www.qualityprogress.com
QP QUALITY PROGRESS
Shell game
www.qualityprogress.com | April 2008
Regarding the article The CHAIR OF THE BOARD
Ronald D. Atkinson, General Motors
QUALITY PROGRESS Great Debate (April 2008, p.
26), is using the turtle diagram PRESIDENT
Michael D. Nichols, Nichols Quality Associates
the only way to show objective
proof to a third-party auditor PRESIDENT-ELECT
1904-2008
A tribute to his life and work p. 20
Roberto M. Saco, Aporia Advisors
that you are using the process
approach? If it is, I believe the TREASURER
E. David Spong, Boeing (retired)
conclusions about the process
PARLIAMENTARIAN
approach for the environmen- James J. Rooney Jr., ABS Consulting
tal management system (EMS) are true.
DIRECTORS
Although ISO/TS 16949 might say that Jochen Amelsberg, APC International
Belinda Chavez, United Space Alliance
you have to use a turtle diagram (I dont
Brenda M. Fisk, Software Quality Solutions
know because I do not work for a company Richard A. Gould, RG Management Solutions
Kamla P. Gupta, Continuous Improvement Technology
that is certied to ISO/TS 16949), ISO 9001 Stephen K. Hacker, Transformation Systems
does not prescribe the use of turtle dia- International
Gary L. Johnson, Environmental Protection Agency
grams. Currently, my company is integrating Kay A. Kendall, Sun Microsystems
William H. LaFollette, Humana Inc.
ISO 9000, ISO 14001 and OHSAS 18001, and Lou Ann Lathrop, General Motors
we believe the three management standards David B. Levy, Levy Quality Consulting
Richard A. Litts, Litts Quality Technologies
can be audited using the process approach. I Richard F. McKeever, D2 Quality Associates
Aimee H. Siegler, Benchmark Electronics
think Tables 1 and 2 in the article will help us
Donald C. Singer, GlaxoSmithKline
improve our internal auditing process. Steven E. Wilson, Department of Commerce Seafood
Inspection Program
As for not using checklists or the copy of
the standard, what do we use as proof that QP EDITORIAL REVIEW BOARD
Randy Brull, chair
all requirements of the standard are being
met? The checklist is a basic tool used in Administrative Committee
Roger Berger, Brady Boggs, Randy Brull, Jane
auditing and, like all tools, can be misused. If Campanizzi, Larry Haugh, Jim Jaquess, Gary MacLean,
Christine Robinson, Richard Stump
the auditor spends all of their time in a con-
ference room looking only at the documen- Technical reviewers
I. Elaine Allen, Andy Barnett, David Bonyuet,
tation, the auditee should ask their registrar John Brown, Bernie Carpenter, Ken Cogan, Linda
Cubalchini-Travis, Ahmad Elshennawy, Tim Folkerts,
to send a different auditor next time. Eric Furness, Mark Gavoor, Kunita Gear, Lynne Hare,
There is no value to the auditee or the Ron Kenett,
Ray Klotz, Tom Kubiak, William LaFollette,
auditees customers to an audit that is Shin Ta Liu, Pradip Mehta, Gene Placzkowski, Paul
conducted without spending time studying Plsek, Tony Polito, Peter Pylipow, Philip Ramsey, R.
Dan Reid, Wayne Reynolds, John Richards, James
the processes of the company and talking Rooney, Anil Sengupta, Sunil Thawani, Joe Tunner,
Jeffrey Vaks,
to the people who are doing the work. Manu Vora, Jack Westfall, James Zurn
The checklist and the standard are not the
problempoor training or the competence
of the auditor is the problem.
Beth Clelland
Quality manager,
Amesbury Group Foam-Tite
Amesbury, MA
Authors response describe what was observed and what was As far as conference room audits, they
Clelland makes some great comments and evaluated. Comprehensive notes combined provide little to no value to the organization
asks some very good questions. I probably with a coverage matrix can show that the and should be avoided.
dont have enough space to address them elements have been evaluated. Terry A. Mors
adequately, but here goes.
The turtle diagram is not the only way
to show objective proof to a third-party
auditor that you are using the process Terry Holley Consulting Inc
approach, and neither ISO/TS 16949 nor
ISO 9001 requires its use. But the diagram
317.532.7866
is one method being employed regularly
www.terryholleyconsulting.com
to demonstrate having used the process
approach. An auditor could simply map out
the process using a process ow diagram,
ask questions following the process path
What does $4995 get your company?
and then record notes that support the
technique that was used. Five Days of On-Site
Its not my insistence against using a Consulting or Facilitation!
checklist, but rather the International Auto- Mix and match services and duration
motive Oversight Bureau regarding ISO/TS to best meet your organizations needs.
16949 third-party auditors. I agree that audi-
tors can use checklists to ensure complete
coverage of a process or standard. But I
Services
have seen many instances where checklists
Project Planning and Start-up Assistance
limit the effectiveness of the audit.
Project Management Methodology Development
Auditors might conduct the audit without
Project Management Fundamentals Facilitation
preparation and then ask only the questions
PM Certication Exam Prep Facilitation
on the checklist, or they might pencil whip
the checklist and not conduct an effective
Project Portfolio Management Training
audit. Internal audit programs should not Quality Management Facilitation
rely too much on checklists and should al-
low time for auditors to prepare before each Advantages
audit, using the checklist as a tool. All services tailored to your needs
Regarding the objective evidence of Raise competence & rapidly meet objectives
comprehensive coverage in the absence of
a checklist, there are several ways to pro-
vide evidence the requirements are evalu-
ated without using a checklist. One way Terry Holley, MBA, PMP
is to maintain thorough auditor notes that DFSS/Lean Certied
Terry has 20 years of corporate and
manufacturing experience rich in project
WRITE AWAY management, hands-on problem-solving,
To offer your own take, log on to www. and training. He is a university instructor
in business courses, including quality and
qualityprogress.com and post a com- project management.
ment on an articles page, or e-mail
editor@asq.org.

June 2008 QP 9
 EXPERTANSWE
Dening defect
Q: Is there a standard method for deter-
mining the number of opportunities that
exist when computing defects per million
opportunities (DPMO)?
A: To use the DPMO metric effectively, it is
important to answer the following questions:
What is a defect? I have always lived
by the principle that anything not done cor-
rectly the rst time is a defect. Of course,
this means understanding what it takes
to do it right the rst time and specifying
those conditions in advance. It also means
holding true to these denitions after they
have been established.
For example, if a unit met any of the
criteria for being declared defective but
was later found to be usable by a mate-
rial review board (MRB), the classication
as a defect should remain intact. Some
organizations are reluctant to embrace this
position because it adversely affects their
quality numbers or because they conclude
that since the unit is usable, it must not
be defective. A defect might not affect
usability.
How can a defect occur? To address
this question, many organizations have
compiled a list of error families and error
types within those families. Such a list
should be complete in identifying all pos-
sible error types (collectively exhaustive).
Also, each error type should be indepen-
dent of all other error types (mutually
exclusive). This allows you to recognize
the occurrence of multiple defects on any
given unit.
Avoid the temptation to exclude known
error types because they happen infre-
quently. In other words, if an error family
or error type is known to occur, include
it on your list. In addition, it is useful to
have an error family or error type deemed
other because you might lack the
foresight or wisdom to dene everything
in advance.
As you develop your list of defect types,
it is useful to dene them in pairs (too high
or too low, too long or too short), particu-
larly when physical, mechanical or electri-
cal characteristics are under consideration.
Use Table 1 as a starting point for such a
list.
Where can a defect occur? A defect can
occur during a process when the outcome is
not the outcome specied in advance.
Suppose an operator is soldering parts to
a circuit card assembly precisely at the pre-
scribed locations, according to the assembly
instructions and drawing. The instructions
specically indicate that no parts should
be soldered anywhere on the board other
than where specied (you are specifying
in advance the conditions under which a
defect can occur). There are 100 such parts
to be soldered, and the assembly operator
has soldered each part correctly.
During the soldering process, the opera-
tor slipped and applied solder on the card
in an area where no soldering was to be
placed. Fortunately, the additional solder
did not compromise the usability of the
completed circuit card assembly.
Did a defect occur? Yes, because solder
was applied to a location where it should
not have been applied as specied in
the assembly instructions and drawing.
Perhaps the card can be reworked and the
solder removed without destroying the
circuit card assembly. A defect did occur,
however, and it should remain in the com-

Defect families and types / TABLE 1 panys quality system.

So, to answer the question of whether
Physical Electrical Mechanical
there is a standard method for comput-
Dimensional Resistance Torque ing DPMO, the answer is yes and no. Yes,
Too long Too high Too high because the computational method of de-
Too short Too low Too low termining the number of DPMO is standard.
Color Capacitance Cross-threaded screw No, because the method for dening and
Wrong color Too high Other counting defects depends on each indi-
Weight Too low vidual organization and that organizations
Too heavy Voltage ability and willingness to create accurate
Too light Too high and meaningful defect data.
Orientation Too low
For an example of the application of DPMO and online-only
Wrong Other tables, visit www.qualityprogress.com.

Missing T.M. Kubiak, author and consultant

Other Weddington, NC

12 QP www.qualityprogress.com
RS
Overcoming resistance to change
Q: My senior managers have assigned me
to lead a cross-company team that will
implement a major quality initiative at my
company. I recognize that this initiative
will involve signicant change. How can I
overcome resistance to change?
A: Overcoming resistance to change can be
one of the most difcult assignments you
can take on. There are several approaches
that can be effective, depending on your
situation and environment.
In selecting your implementation team,
try to include people from across the or-
ganization. These people will be important
in communicating what potential change
issues exist in their areas and the progress
of your team in those areas.
Early in the process, identify all potential
barriers to change with your team and the
actions you can take to address them. A
force eld diagram can be a useful tool.
Identify the inuencers in your organiza-
tion. These peoples opinions are respected
by large groups of employees in the
organization. These inuencers are usually
managers, but they dont have to be. Meet
with the inuencers to understand how
they view the change. Work toward gaining
their support by showing how the change
will benet them and their group.
Remember that the most damaging
people to your effort are not the ones who
tell you they dont support the change, but
rather those who tell you they support the
change while telling their circle of inuence
they dont support the change.
AN EXPERTS OPINION
If you would like to reach QPs experts or pose a question for consideration,
visit www.qualityprogress.com and click on the Expert Answers department,
or e-mail editor@asq.org.
Clearly communicate the change and
how it will affect people in the company.
Be sure that senior leaders are part of the
communication process. Communicate
early and often. By doing so, youll keep the
rumor mill to a minimum and help reduce
fear of the unknown. Leverage your team
members and any existing communica-
tion vehicles (for example, meetings or the
company newsletter) to disseminate your
message.
Listen to concerns and show people
you are addressing their concerns as best
you can. Many times, change involves
losing something, even if its just the way
weve always done things around here.
Acknowledge the loss and show how the
changes will benet the people in your
company.
Develop support mechanisms for the
change and show people how to access
those mechanisms. This will help people
make adjustments and reduce anxiety.
Maintain exibility. Be willing to make
course adjustments based on your teams
observations and feedback.
Ken Cogan, senior manager of
performance management, Intelstat
Columbia, MD
Sample equation
Q: Everyone knows the old standby sam-
pling plan of square root of (N + 1) to de-
termine a sample size for discrete numbers
of materials. What is the exact source or
rationale behind this equation?
Carolyn Tomlinson
Greenville, NC
A: The rule dates back to the 1920s as an
easy-to-remember sampling scheme for
agricultural regulatory inspectors. It was
semiformalized in an unpublished report by
the Assn. of Ofcial Agricultural Chemists
(now AOAC) in 1927.
As common as the use was, most qual-
ity personnel believed that the scheme
was statistically questionable. It does meet
the denition of a sampling plan in that it
provides a sample size, an accept value (0)
and a reject value (1). Even though the plan
has been questioned, it was adopted by
many international and federal regulatory
agencies, including the United Nations, the
World Health Organization and the Food
and Drug Administration.
Hewa Sarandasa addressed this issue
in an article published in Pharmaceutical
Technology: The accuracy of the 95% con-
dence probability statement for mean was
compared for three distributions for sample
size obtained from the square root of N plus
one rule with the Edgeworth approximation
derived sample size. Results showed that
the sample size obtained from this rule is
not even enough to declare less than 20%
of defectives in a moderate size population
with a high degree of condence. Therefore,
the author concludes this rule should not
be used to select a sampling plan to infer a
population defective rate.1
REFERENCES
1. Hewa Sarandasa, The Square Root of N Plus One
Sampling Rule: How Much Condence Do We Have? Phar-
maceutical Technolgy, Vol. 27, No. 5, pp. 50-62.
BIBLIOGRAPHY
Keith Borland, The Fallacy of the Square Root Sampling
Rule, Journal of the American Pharmaceutical Association,
Vol. 39, No. 7, pp. 373-377.
Wayne A. Taylor, Acceptance Sampling Questions, www.
variation.com/FAQs.html (case sensitive).
I. Elaine Allen,
associate professor of statistics
Babson College, Wellesley, MA
June 2008 QP 13
 KEEPINGCURRE
PROCESS IMPROVEMENT
Groups Call for Voting Reform
A As political candidates duke it out for the chance to be on the ballot
in November, two professional societies are calling for an overhaul
of the voting process to win back voter condence.
Quality and statistics must play a key role in ensuring voting
procedures are carried out properly and vote totals are measured
accurately, ASQ and the American Statistical Assn. (ASA) recom-
mended in separate statements earlier this year.
Its very important for citizens to have as much trust in their
voting systems as they have in their medical care systems or air
travel, said Liz Keim, a past president of ASQ. So why not lever-
age some of the same basic tools that bring reliability to critical
functions to solve quality problems that continue to plague our
elections?
14 QP www.qualityprogress.com
Peter Lachenbruch, ASA president and board chairman, said
state and federal governments must adopt routine monitoring of all
electoral procedures to ensure continuous quality improvement.
Statistical methods can provide condence in well-conducted
contests, Lachenbruch said in a prepared statement. They can
also identify common sources of problems and help avert future
electoral failures.
ASQs recommendations include:
Establish standardized procedures, such as voting machine
start-up protocols, voter instruction procedures and steps for
secure handling of ballots and equipment.
Implement simple checklists based on the standardized proce-
dures.
Test the proposed solutions adequately.
Train the volunteer election workers on critical points in the
system where problems are likely to occur.
Standardize and document procedures at polling places.
Implement checklists for volunteers at polling places, similar to
preight checklists used by pilots.
ASA called on the federal government to:
Conduct broader research into the integrity of elections.
Provide tools to help election ofcials conduct high-integrity
elections.
Design acceptable ballots and procedures to test ballots on the
actual voting machines.
ASA and ASQ agree that audits must be part of the election
process overhaul.
ASA recommends audits of voter-veried hardcopy records to
provide highand clearly speciedlevels of condence in elec-
toral outcomes.
ASQ suggests two types of audits:
1. Post-election audits of results to detect anomalies that indicate
fraud or systemic errors in voting equipment.
2. Audits against an established standard. The Election Assistance
Commission is working to establish this standard.
ASQs recommendations are included in the latest ASQ Quarterly
Quality Report. You can access the entire report at www.asq.org/
quality-report/reports/200803.html.
NT
FOOD SAFETY

FOOD SAFETY MEASURES

RAMPED UP FOR OLYMPICS
Concerns over the safety of food at this summers Olympic Games has
prompted Chinas top quality ofcial to advise food suppliers to save video
records of their monitoring procedures in case tampering does occur.
The safety of Olympic food is our top priority, said Li Changjiang,
Whos Who in
NAME: Wilson Goncalves da Silva.

RESIDENCE: Londrina, Paran, Brazil.

Q
EDUCATION: He holds post graduate degrees in busi-
the director of Chinas general administration of quality supervision, ness administration from Alvarez Penteado, and in
inspection and quarantine. We must look carefully into every step of total quality management from Fundao Getulio Var-
the production process and be vigilant to possible sabotage. gas, both located in So Paulo, Brazil.
Changjiang asked suppliers to maintain video monitoring records of
CURRENT JOB: Packaging manager
their foods storage and production. Usually, suppliers overwrite moni-
at Sandoz do Brasil Ltd., a Novartis
toring tapes every 20 to 30 days.
company.

The safety of Olympic food is our top

priority. We must look carefully into
every step of the production process and
be vigilant to possible sabotage.
PREVIOUS JOB: Goncalves has worked in
the pharmaceutical industry since 1974.

INTRODUCTION TO QUALITY: Dur-

Chinese government ofcials are also planning to use global posi-

tioning satellites to monitor food production, processing factories and
ing his last year as a chemistry student, Goncalves
worked at Eli Lilly and Co. and came in contact with
different quality tools and quality manuals. He was
exposed to quality concepts such as current good
food hygiene to make sure safe food is delivered to the 10,500 athletes manufacturing practices, internal audits and standard
residing in the Olympic Village. operating practices.
Food products will be afxed with an Olympic food safety logistics
ASQ ACTIVITIES: An ASQ member since 1996, Gon-
code and transportation vehicles will be tracked using global position-
calves became country councilor for Brazil in 2000.
ing satellites.
He is a member of several divisions, including Lean
Meanwhile, a Chinese government news agency reported a World
Enterprise, Human Development and Leadership, and
Health Organization ofcial recently gave high marks to China and its
Food, Drug and Cosmetics. Goncalves is applying to
preparation for possible public health threats during the upcoming
be regional country councilor of the Food, Drug and
Olympics. The opening ceremony is Aug. 8.
Cosmetics Division.
Sources: The Associated Press, the International Herald Tribune and Asian News International.
OTHER ACTIVITIES: He is the director of the Interna-
tional Society of Pharmaceutical EngineeringBrazil.

PERSONAL: Married, three children.

FAVORITE WAYS TO RELAX: Reading, walking on the

beach and preparing a Brazilian barbecue.

QUALITY QUOTE: Companies are looking for more

problem solvers inside the workplace and encour-
aging quality managers to use their brains to solve
problems permanently.

June 2008 QP 15
 KEEPINGCURRENT
WORLD CONFERENCE 2008 ROUNDUP
Quality Professionals Coming Together
Nearly 2,000 quality professionals gathered in Houston last month
for ASQs annual World Conference on Quality and Improvement
and the Quality Institute for Healthcare. The conference followed
the theme Generation Quality, and a number of speakers and
sessions addressed that focus. Other themes coursed through the
conference schedule, including international quality and innovation.
Featured events
Rear Adm. T.K. Ken Mattingly, one of the heroes of the Apollo 13
space mission, retold the adventure of the space ight gone awry,
and how he and a team worked together to bring the crew safely
back to earth.
The biggest lessons Mattingly took away from the near-disaster
could be applied to any challenge: Have a clear goal.
Know what you have to accomplish and dont let anything get
in the way, Mattingly said.
Personal accountability is also a factor in any teams success, he
added. Its important for each team member to be able to say, It
wont fail because of me, he said.
MEMBERS OF THE Boeing Co.s Universal Splice Machine Project team
show off the gold award they received at the team competition in
Houston.
An international panel discussion, titled 21st Century Chal-
lenges Call the Quality Movement to Action, brought together
some of todays brightest thinkers in quality: James Faubion,
chair, department of anthropology, Rice University, Houston;
Marius J.S. Buiting, president of the European Society for Quality
16 QP www.qualityprogress.com
in Healthcare; Shoji Shiba, visiting professor at the Massachu-
setts Institute of Technology and professor emeritus, University
of Tsukuba, Japan; and Margaret Wheatley, president emerita of
the Berkana Institute.
They relayed their unique views on how crucial quality is when
addressing many of societys problems, and how quality can lead to
a richer life for everyone.
Remembering Juran
The death of Joseph M. Juran in late February was on the minds of
quality professionals as they gathered for the conference. Juran In-
stitute CEO Joseph DeFao offered a moving tribute and introduced
a video that led off conference activities.
One thing Dr. Juran would want mostdont forget the contri-
butions quality leaders have made. In the words of Juran, Keep the
revolution alive.
During the conference, a separate room with remembrances
was open to all attendees.
Team excellence awards
The Boeing Co.s Universal Splice Machine Project team took top
honors, winning the gold award at the 23rd International Team
Excellence competition.
A rigorous nalist competition included 26 teams from around
the world. Gold, silver and bronze awards were presented at the
conference.
The Silver Award went to Healthways Improving Member
Contact Rate team. Bayer MaterialSciences Optimizing the TL/LTL
Decision team took the Bronze Award.
ASQ ofcers
ASQ announced the names of its ofcers for the 2008-2009 term
at the annual business meeting, held in conjunction with the World
Conference. The newly elected ofcers who will assume their roles
July 1, are:
President Roberto M. Saco.
Chairman Michael D. Nichols.
President-elect Peter L. Andres.
Treasurer E. David Spong.
Newly elected board members are Darlene Deane, Alexis P. Gon-
calves, Marc Kelemen and Art Trepanier. Re-elected board members
are Belinda Chavez, Lou Anne Lathrop, Aimee Siegler and Steve
Wilson.
DATEINQUALITYHISTORY
QP looks back on an event or person that made a difference in
QP ONLINEONPAPER
the history of quality.

June 18, 1926

QUICK POLL RESULTS
Philip B. Phil Crosby was born in
Each month at www.qualityprogress.com,
www.qualityprogre readers can take a
Wheeling, WV.
short, informal survey,
rvey and we po
post the results. Here are the num-
Crosby was a businessman and author
bers from a recent Quick Poll:
who contributed to management theory
and quality management practices. He is
What ill effects has your company experienced as a result
widely recognized for promoting the concept of zero defects of the sluggish economy?
and for dening quality as conformance to requirements. Budget cuts: 28.1%
Crosby is credited with reinforcing the message, Do things Layoffs: 18.7%
right the rst time. He also became well known for acquaint- Hiring freeze: 18.7%
ing quality professionals with the following theories: The price None: 18.7%
of nonconformance is a measure of poor quality; prevention is All of the above: 15.6%
a means to eliminate quality problems; leadership is a require-
ment to make progress; teamwork is the principle for work; Visit www.qualityprogress.com to answer the most recent
and customer requirements dene the standard of quality Quick Poll question posted:
performance. Which method of measuring customer satisfaction is the
Crosbys rst book, Quality is Free, has been credited, in most effective?
part, with igniting the quality revolution in the United States
and Europe. He published a total of 13 books, including The TOPICS AND SUBTOPICS
Absolutes of Leadership in 1996 and Quality and Me, an autobi- At www.qualityprogress.com, past articles are organized and
ography published in 1999. divided into topics and subtopics. Customer loyalty and satisfac-
He died in August 2001 at the age of 75. tion, has been divided into three subtopics. The top articles for
each subtopic are as follows:

PROCESS MANAGEMENT Customer experience management: Eight Steps to Sustain

APQC RELEASES THREE NEW
PROCESS FRAMEWORKS
APQC, formerly known as the American Productivity and
Quality Center, has developed three industry-specic business
process frameworks for the consumer products, electric utili-
ties, and aerospace and defense industries.
The industry versions are based on APQCs Process Clas-
sication Framework (PCF), a taxonomy of cross-functional
business processes intended to allow objective comparison of
performance within and among organizations.
Supported by IBM, the three frameworks and newly updat-
ed cross-industry PCF are available at no charge for organiza-
tions to download at www.apqc.org/pcf. An additional seven
industry PCFs are scheduled for release later this year.
APQCs PCF now forms the foundation of the Open Stan-
dards Benchmarking Collaborative research.
Change, by John R. Schultz.
Customer requirements: AS&D Standards Revisions to Im-
prove Supplier Performance, by L.L. Buddy Cressionnie.
Customer satisfaction and value: Build Loyalty Through Experi-
ence Management, by Leonard L. Berry and Lewis P. Carbone.
To check out more articles and other articles under different
topics, click on Topics on at www.qualityprogress.com.
QUALITY NEWS TODAY
QP scours the news wires each day to bring you the latest local,
regional, national and international stories related to quality. Each
week, QP editors post about 10 articles a day on topics such as air
travel, healthcare and product recalls. Visit www.qualityprogress.
com and nd Quality News Today on the right-hand side of the
home page.

June 2008 QP 17
 KEEPINGCURRENT
ASQNEWS
CALL FOR 09 PAPERS Proposals for ses-
sions at next years ASQ World Conference
on Quality and Improvement in Minneapolis
will be accepted until Aug. 1. The confer-
ence will focus on ve areas: quality basics,
customer service, driving quality through-
out the organization, improved perfor-
mance and 21st century needs, competen-
cies and issues. Visit http://wcqi.asq.org for
more details on the submission process.

SR WEBSITE ASQ has launched a new

website, www.thesro.org, devoted to qual-
itys role in social responsibility (SR). The rst
phase of the website features user-generated
content about qualitys contribution to SR,
the ISO 26000 draft standard and related ASQ Ron Sachs, Consolidated News Photos

products, events and research. Malcolm Baldrige National Quality Award re- earlier this year to convince the president to
cipients at a White House reception following attend the ceremony to highlight the prestige
BUSH AND BALDRIGE (above right) Presi- the awards ceremony in Washington, D.C., of the award. Bush had not participated in
dent George W. Bush recognized this years in April. ASQ led a letter writing campaign four out of the last ve ceremonies.

WEBWATCH
This months Web Watch focuses on customer satisfaction. For
more quality related websites, visit www.asq.org.
www.csmassociation.org
The Customer Satisfaction Measurement Assn. is a membership or-
ganization that conducts benchmarking studies to identify practices
that improve customer satisfaction and overall operations. The site
lists 70 organizations as members, including Fortune 500 compa-
nies, educational institutions and government agencies.
www.intelliseek.com
Have a compliment or complaint about a product or service? Want
to read what other people think? This website allows consumers to
leave feedback on good and bad experiences with companies in a
variety of industries. Free of charge, users may enter comments or
read other users comments and rankings of individual companies.
FOUND AN INTERESTING QUALITY SITE? If you
come across a noncommercial site that could be useful to
other quality professionals, e-mail it to medmund@asq.org.
Free registration allows you to manage your letters and participate
in rankings and, according to the website, increases your chances of
receiving responses from companies.
www.statpac.com/customer-satisfaction.htm
This site provides simple guidelines for designing customer sat-
isfaction surveys and employee opinion surveys. Steps detailed
include dening the goals of the research, writing the individual
survey items and testing the reliability of a survey. The site also
provides free downloadable sample surveys.
www.theacsi.org
The American Customer Satisfaction Index (ACSI) tracks trends
in customer satisfaction and provides benchmarking insights into
the consumer economy for companies, industry trade associa-
tions and government agencies. ACSI rates industries such as
manufacturing, transportation, retail and nance on scales of
0-100 based on telephone interviews with consumers. The results
of these quarterly studies can be found on this website, organized
by industry and specic companies.

18 QP www.qualityprogress.com
STATISTICS
Is SportsCenter Next?
ESPN, the sports media giant, recently published a two-part series on build-
ing the best fantasy baseball team after talking to a group of statisticians
(and self-professed baseball fanatics) that rst had the work published in QP
last year.
ESPN.coms Jason Grey picked up on the study done by I. Elaine Allen,
an associate professor of statistics and entrepreneurship at Babson College
in Massachusetts (and a regular contributor to QPs Statistics Roundtable
column), Kirill Kustov, a Babson undergraduate statistics student and George
Recck, the director of the colleges math resource center.
Greys column and the QP article, Building a Better Fantasy Baseball
Team (pp. 24-29, April 2007), detail how the group applied multivariate and
univariate analysis to players statistics to reveal which players have the
potential to be the most valuable on a fantasy baseball team.
Visit www.espn.com and search under Grey Matter to view the columns.
You must pay a subscription to access this part of ESPN.coms fantasy
baseball section. To view the QP article, visit www.asq.org/quality-prog-
ress/2007/04/statistics/building-a-better-fantasy-baseball-team.html.
160,000
THE ESTIMATED NUMBER OF professionals who have received ASQ
certications since ASQ began offering them 40 years ago.
Mr. Pareto Head BY MIKE CROSSEN
SHORTRUNS
THE INTERNATIONAL ORGANIZATION FOR
STANDARDIZATION (ISO) has launched a proj-
ect committee to develop an international stan-
dard on energy management. The standard will
be based on continual improvement and the
plan-do-check-act approach used in ISO 9001
and ISO 14000 to provide compatibility and in-
tegration opportunities. For details, go to www.
iso.org/iso/pressrelease.htm?red=Ref1122
(case sensitive).
THE STANDARDS ENGINEERING SOCIETY
(SES) is sponsoring a paper competition as
part of the U.S. celebration of World Standards
Day, Oct. 23. The competitions theme is intel-
ligent and sustainable buildings. Submissions
must be received by SES by Aug. 29 to be
eligible for three cash prizes. For details, go
to www.ses-standards.org/displaycommon.
cfm?an=1&subarticlenbr=77.
THE AMERICAN SOCIETY FOR NONDE-
STRUCTIVE TESTING is developing a new
instruction series program for magnet particle,
liquid penetrant, ultrasonic, electromagnetic
and radiographic testing and is seeking quali-
ed authors. For more information, go to www.
asnt.org/latestnews/authors2007a.htm.
June 2008 QP 19
 The
TRUE TEST Of
Measuring In 50 Words
Or Less
advocacy, Despite its importance
in increasing prot-

purchasing ability, customer loyalty

measurement hasnt
kept pace with its tech-

and retention nology.

Using advocacy, pur-

can increase
chasing and retention
indexes to manage
loyalty is statistically

protability superior to using any

single question alone.
These indexes helped
predict the growth
potential of wireless
service providers and
PC manufacturers.
by Bob E. Hayes

20 QP www.qualityprogress.com
 CUSTOMER LOYALTY

Loyalty
THE CUSTOMER LOYALTY eld has experienced much tech-
nological innovation, such as automated reporting portals and integration
of attitudinal and behavioral data in customer relationship management
applications, over the past decade. The quality of measurement and under-
standing of the meaning of customer loyalty, however, have not kept pace.
A multidimensional measurement approach to conceptualizing customer
loyalty can offer new insights into growing business through new and exist-
ing customers.
There are several objective measures of customer loyalty:
Number of referralsword of mouth and word of mouse.
Decision to purchase again.
Decision to purchase different products.
Decision to increase purchase size.
Customer retention and defection rates.

June 2008 QP 21
 Customer loyalty can lead to protability gains.
Companies can grow through the referral process,
thanks to the acquisition of new customers. The idea
is that the customer acquisition process relies on exist-
ing customers to promote or recommend the company
to their friends, who, in turn, become customers.
Another way to strengthen the nancial growth of a
company is by having existing customers increase their
purchases or buy other products or services it offers.
Finally, nancial growth is dependent on a companys
ability not to lose existing customers at a faster rate
than it acquires new ones.
Customer defection rate is an important metric in the
wireless service industry, where processes such as num-
ber transfers and contract terminations make customer
defection measurement straightforward and routine.
Customer loyalty surveys
Customer surveys rather than objective loyalty mea-
sures are still the more commonly used way to assess
customer loyalty, for several reasons:
Customer surveys allow companies to quickly and
easily gauge levels of customer loyalty.
Results from customer surveys can be more easily
used to change organizational business processes.
Customer surveys commonly include questions
about customer loyalty, as well as the customer ex-
perience with products, services or support. Used
jointly, these questions can be analyzed to identify
reasons customers are loyal or disloyal.
Questions about expected levels of loyalty-related
behavior provide opportunities for companies to
look into the future regarding customer loyalty.
For each survey question or item, customers are
asked to rate their likelihood of engaging in a specic
behavior. Commonly used customer loyalty survey
questions mirror the seven objective loyalty measures:
1. Overall satisfaction.
2. Likelihood of a rst-time purchaser to repurchase.
3. Likelihood to recommend.
4. Likelihood to continue purchasing the same prod-
ucts or services.
5. Likelihood to purchase different products or services.
6. Likelihood to increase frequency of purchasing.
7. Likelihood to switch to a different provider.
Most of the questions allow respondents to indicate
their likelihood of behaving in different ways toward
the company (for example, 0 = not at all likely to 10
= extremely likely). The satisfaction question is some-
times used in customer loyalty measurement and is rat-
ed on a scale (for example, 0 = extremely dissatised
to 10 = extremely satised). For all questions (except
for likelihood to switch), higher ratings reect higher
levels of customer loyalty.
Measurement error
Objective measures of customer loyalty have minimal
measurement error associated with them because they
have unambiguous meanings.
The number of recommendations a customer makes
is clearly distinct from the number of repeat purchases
that customer makes. This is not to say these measures
of customer loyalty are unrelated, but instead that they

Wireless provider factor pattern are measurably different entitiessimilar to the fact

matrix / TABLE 1 that height and weight are different constructs but
are related to each other because taller people tend to
Factors1 weigh more than shorter people.
Measures of customer loyalty via survey questions,
Questions (items) 1 (Advocacy) 2 (Purchasing) 3 (Retention)
however, have measurement error associated with
Overall satisfaction 0.79 0.34 0.35 them. In psychological measurement terms, the loyalty
Choose again 0.71 0.41 0.43 questions are simply observable indicators of a single
Recommend 0.78 0.40 0.39 underlying construct. Specically, customers ratings
Purchase same 0.61 0.37 0.58 of each of the questions (loyalty items) is simply a

Purchase different 0.29 0.75 0.23

function of an underlying construct (loyalty).
Because of the way customers interpret survey ques-
Purchase increase 0.26 0.76 0.09
tions and the inherent error associated with measuring
Likelihood to switch to 0.32 0.12 0.68
another provider2 psychological constructs, ratings on the customer sur-
1. Based on a factor analysis with varimax rotation.
veys do not exactly match the true loyalty of the cus-
2. Reverse coded so higher scores mean lower likelihood of switching. tomers. That is, customers true levels of loyalty can

22 QP www.qualityprogress.com
 CUSTOMER LOYALTY

only be estimated through the questions on the survey.

As such, ratings need to be critically evaluated to ensure
PC manufacturer factor
the meaning behind the ratings is understood.1
pattern matrix / TABLE 2
Factors1
Factor analytic view Questions (items) 1 (Advocacy) 2 (Purchasing)
To determine the measurement properties of the loyal- Overall satisfaction 0.77 0.32
ty items, two separate studies were undertaken in June
Choose again 0.81 0.35
and July 2007, each within a specic industry: wireless
Recommend 0.90 0.34
service providers and PC manufacturers.
Purchase same 0.81 0.39
For each study, a sample of about 1,000 general
consumers ages 18 and older in the United States were Purchase different 0.49 0.54

asked about their attitudes toward their wireless ser- Purchase increase 0.35 0.92
vice providers or PC manufacturers. All respondents Purchase frequency 0.32 0.79
were interviewed to ensure they met correct proling 1. Based on a factor analysis with varimax rotation.
criteria and were rewarded with an incentive for lling
out the survey. Global Market Insite Inc. collected the involves some form of judgment when determining the
survey data. number of factors, as well as deciding which variables
The seven loyalty items listed earlier were included are related to the smaller set of factors.2
in the wireless service provider sample. For the PC The elements in the factor pattern matrix are called
manufacturer sample, the rst six questions were used factor loading and essentially reect the correlation
with an additional question (likelihood to increase fre- between each item and the three factors: advocacy,
quency of purchasing). purchasing and retention. Higher factor loadings indi-
In statistical terms, factor analysis is a data reduc- cate a stronger relationship between the item and the
tion technique that explains the statistical relationships underlying factor.
among a given set of variables using fewer unobserved The results of the factor analyses suggest the seven
variables (factors). In simpler terms, a factor analysis items measure fewer constructsthree for the wire-
tells us two things: less sample and two for the PC sample. Tables 1 and 2
1. The number of factors (constructs) being measured represent the factor pattern matrixes.
by the set of questions.
2. Which questions are related to which factors. Advocacy, purchasing and retention
Specically for our example, a factor analysis helps The items that load on the rst factor appear to have
determine whether the set of seven original loyalty ques- a strong emotional component, reecting the extent to
tions is actually measuring fewer constructs (factors). It which customers advocate the company. Consequent-
is important to note that an exploratory factor analysis ly, this factor was labeled advocacy loyalty. The items

Loyalty scores for wireless Loyalty scores for PC

providers / FIGURE 1 manufacturers / FIGURE 2
9.0 ALI 10.0 ALI
7.9 7.8 8.7
8.0 7.3 PLI 9.0 PLI
7.2 7.1 7.6 7.5 7.6
Customer loyalty

6.8 8.0 7.4 7.0 7.2 7.1

7.0 6.5 RLI
Customer loyalty

6.1 6.2 5.9 6.0 6.0 7.0 6.3 6.2 5.9

5.8 5.7 5.7
6.0 5.4 6.0 5.2
4.9 5.0
5.0
4.0
4.0 3.0
3.0 2.0
1.0
2.0
0.0
1.0
ple

es
)

ay
mp d
ll

rd
(HP

aq

ib
De

(Co ckar

hin
ka

tew

0.0
sh
Ap

ac

Alltel ATT/ Sprint/ T-Mobile Verizon

To

ac
a

Ga
t-P
t-P

eM

Cingular Nextel Wireless

et
et

wl
wl

He
He

June 2008 QP 23
 that load on the second factor reect specic purchas-
ing behaviors. This second factor is labeled purchas-
ing loyalty. For the wireless service provider sample,
the item that represents the third factor reects reten-
tion (the opposite of switching) and is, therefore, la-
beled retention loyalty.
Indexes for each factorially derived component of
customer loyalty can be calculated by averaging the
loyalty items that load highly on the same factor:
Advocacy loyalty index (ALI): Reects the de-
gree to which customers will be advocates of the
company (average across satisfaction, recommend,
choose again and purchase same).
Purchasing loyalty index (PLI): Reects the
degree to which customers will increase their pur-
chasing behavior (average across purchase differ-
ent, purchase increase and purchase frequency).
Retention loyalty index (RLI): Reects the de-
gree to which customers will remain with a given
company (single defection itemreverse coded).
Loyalty index reliability
Reliability deals with the extent to which measurement
is free from random error. For the wireless service pro-
vider sample, the reliability (Cronbachs alpha3, 4) of
the ALI was 0.92. The reliability estimate (Cronbachs
alpha) for the PLI was 0.82.
For the PC manufacturer sample, the reliability
of the ALI was 0.94. The reliability of the PLI was
0.87. These levels of reliability are considered very
good for attitude research (0 = no reliability, 1 =
perfect reliability)5 and suggest that each measure
has minimal measurement error.
Of particular interest are three specic loyalty items
that load on advocacy loyalty in each study:
1. Satisfaction.
2. Recommend.
3. Purchase same.
Net promoter score (NPS) developers say the rec-
ommend question is the best predictor of business
growth.6, 7 This conclusion has come under recent at-
tack from other researchers who have found that the
satisfaction and purchase same questions are just
as good as the recommend question in predicting
business growth.8-10
The current factor-analytic ndings cast additional
doubt on the conclusions by the NPS camp. The rec-
ommend question appears to measure the same un-
derlying construct as the other two loyalty questions.
There is no scientic evidence the recommend ques-
tion is, or should be, a better predictor of business
growth compared to other loyalty questions.
The NPS developers support the use of a single
question to understand customer loyalty. Current study
ndings dont support this single-item approach. There
is nothing unique and special about the recommend
question. Furthermore, single-item measures are less
reliable (contain more measurement error) than mul-
tiple-item measures.
Measuring loyalty with a single question is akin to
measuring math skills with a single-item math test. An
answer to the single item test would be a less reliable
reection of math skills than the combined answers to
a 50-item math test. Would you want your childs Scho-

Impact on objective Advocacy loyalty and new

loyalty measures / FIGURE 3 customer growth / FIGURE 4
8.0%
Percentage total new customer growth
Loyalty indexes (Q2 07)

ALI Verizon
7.0% Wireless
6.0% T-Mobile
PLI
(Q2-Q3, 07)

5.0% AT&T/
Cingular
4.0%
RLI Alltel
3.0%
Sprint/
0.0 0.2 0.4 0.6 0.8 1.0
2.0% Nextel
Correlation with objective metrics of loyalty
Percentage total new customer growth (Q2-Q3, 07) 1.0%
ARPU growth (Q2-Q3, 07) 6.0 6.5 7.0 7.5 8.0 8.5
Churn Q3, 07 Advocacy loyalty index (Q2, 07)

24 QP www.qualityprogress.com
 CUSTOMER LOYALTY

lastic Aptitude Test score to be determined by a single

question or the entire set of questions on the test?
Objective loyalty measures
Using the loyalty indexes in customer loyalty man-
for wireless providers / TABLE 3
agement is statistically better than using any single Percentage total
question because the indexes provide a more precise new customer
measure of loyalty than any of the items used alone. ARPU growth Churn growth Q2-Q3,
(Q2-Q3, 2007) (Q3, 2007) 2007
Alltel $1.86 1.9% 3.7%
Ranking companies
Figures 1 and 2 (p. 23) contain the average loyalty AT&T $0.19 1.7% 4.7%

indexes for the wireless service providers and PC Sprint/Nextel $-1.00 2.3% 2.3%
manufacturers, respectively. The ALI and PLI are T-Mobile $2.60 2.9% 5.9%
sensitive enough to detect differences across the Verizon $0.33 1.2% 6.8%
companies. It appears that the measurement preci- Wireless
sion of each of the loyalty scales can detect mean-
ingful differences across different companies, thus tionship between the PLI and ARPU growth.
enabling researchers and practitioners to reliably Finally, the RLI was the best predictor of actual churn
study different types of loyalty across different rates for wireless service providers. Companies with
groups of customers. higher RLI scores had lower churn rates compared to
companies with lower RLI scores. Figure 6 (p. 26) illus-
Predicting business growth trates the relationship between the RLI and churn rate.
To understand how well the ALI and PLI predict future The results show that loyalty indexes are predic-
growth, objective loyalty measures for the wireless tive of future business growth through new custom-
service providers were collected for the third quarter ers (new customer growth) and existing customers
of 2007 (www.ercewireless.com and quarterly reports (ARPU), suggesting that the ALI, PLI and RLI are use-
from providers respective websites). ful measurement instruments in managing customer
Each loyalty index was correlated with each of the loyalty and business growth.
following objective loyalty measures collected for that While the current results are based only on the
quarter (see Table 3 for values): wireless industry, the ndings showing the predictive
Average revenue per user (ARPU) growth, quarter power of the ALI and PLI are very compelling. Future
two to quarter three, 2007. research in other industries can help verify and extend
Churn for quarter three, 2007, reverse coded so the current ndings.
higher scores reected better retention.
Total percentage of new customer growth, quarter
two to quarter three, 2007, estimated from churn
rate and net new customers Purchasing loyalty and
The correlations for each loyalty index with each ARPU growth / FIGURE 5
objective loyalty measure are in Figure 3. Each loyalty
$3.00
index (quarter two, 2007) was differentially related to T-Mobile
$2.50
ARPU growth (Q2-Q3, 07)

the objective loyalty measures (quarter three, 2007).

$2.00 Alltel
The ALI had its greatest impact on new customer
$1.50
growth. Companies with higher ALI scores experienced
$1.00
greater new customer growth compared to customers
$0.50 Verizon
with lower ALI scores. Figure 4 illustrates the relation- AT&T/ Wireless
$0 Cingular
ship between the ALI and new customer growth.
$(0.50)
The PLI, however, was highly predictive of ARPU Sprint/
$(1.00) Nextel
growth. Companies with higher PLI scores also expe-
$(1.50)
rienced greater ARPU growth compared to companies 4.5 5.0 5.5 6.0 6.5
with lower PLI scores. Figure 5 illustrates the rela- Purchasing loyalty index (Q2, 07)

June 2008 QP 25
 CUSTOMER LOYALTY

Retention loyalty and PC manufacturer from existing customers. Ad-

ditionally, T-Mobile custom-
churn rate / FIGURE 6 loyalty grid / FIGURE 7 ers indicate they are likely
3.0% Faster to increase their purchase
T-Mobile Apple
2.8% behavior at a rate comparable
2.6% to the customers of Alltel and

Purchasing growth
2.4%
Churn (Q3, 07)

Sprint/ Hewlett- Verizon.

2.2% Nextel Hewlett- Packard
Packard (HP) New-customer growth po-
2.0% (Compaq)
Alltel tential for AT&T is on par
1.8% eMachines
AT&T/ Toshiba Dell with the industry average.
1.6% Cingular
1.4% Gateway Sprint/Nextel will experi-
Verizon ence slower growth in both
1.2% Wireless
1.0% customer acquisition and
5.5 6.0 6.5 7.0 Slower
Slower Faster increased purchases from
Retention loyalty index (Q2 07)
New customer growth existing customers.
Advocacy loyalty, purchas-
Loyalty grid ing loyalty and retention loy-
The ALI assesses new customer growth, while the PLI alty each provide unique information regarding the
assesses purchasing growth. The Hayes loyalty grid quality of customer relationships. Loyalty indexes
charts the ALI and PLI and helps companies understand predict actual future growth (business outcomes) and
where they rank in the competitive landscape with re- can help companies effectively measure and manage
spect to predicted business growth. Figures 7 and 8 are the different types of customer loyalty. They allow
two examples of the Hayes loyalty grid. companies to better manage their customer relation-
As seen in Figure 7, there is considerable variability ships to maximize growth through new and existing
across PC manufacturers with respect to their growth customers.
potential. Clearly, Apple has high levels of both advoca-
cy loyalty and purchasing loyalty. Compared to other PC REFERENCES AND NOTE
1. American Educational Research Assn., American Psychological Assn. and
manufacturers, Apple should expect to see faster growth National Council on Measurement in Education, Standards for Educational
and Psychological Testing, American Psychological Assn., 1985.
with respect to acquiring new customers and increasing 2. Interested readers can learn more about this topic in B.E. Hayes, Measur-
the purchasing behavior of existing customers. ing Customer Satisfaction, second edition, ASQ Quality Press, 1997.
3. L.J. Cronbach, Coefcient Alpha and the Internal Structure of Tests,
Among the remaining PC manufacturers, HP (Com- Psychometrika, Vol. 16, No. 3, 1951, pp. 297-334.
4. L.J. Cronbach and P.E. Meehl, Construct Validity in Psychological Tests,
paq) and Dells growth Psychological Bulletin, Vol. 52, 1955, pp. 281-302.
potential are on par with 5. J.M. Nunnally, Psychometric Theory, second edition, McGraw-Hill, 1978.

Wireless provider
6. F.F. Reichheld, The One Number You Need to Grow, Harvard Business
the industry average, while Review, Vol. 81, 2003, pp. 46-54.

loyalty grid / FIGURE 8 Gateway, Toshiba and eMa- 7. F.F. Reichheld, The Ultimate Question: Driving Good Prots and True
Growth, Harvard Business School Press, 2006.
chines are expected to ex- 8. Claes Fornell, Sunil Mithas, F.V. Morgensen and M.S. Krishan, Customer
Satisfaction and Stock Prices: High Returns, Low Risk, Journal of Marketing,
Faster perience slower growth in Vol. 70, 2006, pp. 1-14.
both customer acquisition 9. T.L. Keiningham, Bruce Cooil, T.W. Andreassen, and Lerzan Aksoy, A Longi-
tudinal Examination of Net Promoter and Firm Revenue Growth, Journal of
and increased purchases Marketing, Vol. 71, 2007, pp. 39-51.
Purchasing growth

Alltel 10. N.A. Morgan and L.L. Rego, The Value of Different Customer Satisfaction
T-Mobile Verizon from existing customers. and Loyalty Metrics in Predicting Business Performance, Marketing Science,
Wireless
As you can see in Fig- Vol. 25, No. 5, 2006, pp. 426-439.

ure 8, Alltel and Verizon

AT&T/ appear in the upper right
Cingular
BOB E. HAYES is the president and founder of Busi-
quadrant, suggesting they ness Over Broadway, a customer satisfaction and
Sprint/ are poised to experience loyalty research consultancy in Seattle. Hayes holds
Nextel a doctorate in industrial-organizational psychology
faster growth with respect from Bowling Green State University in Ohio and has
Slower
conducted survey research for companies such as
Slower Faster to customer acquisition
Oracle, Siebel Systems, Agilent Technologies and
New customer growth and increased purchases Cisco Systems.

26 QP www.qualityprogress.com
28 QP www.qualityprogress.com
Now
Hear
This
In rst year of eligibility, nonprots
by Brett Krzykowski,
assistant editor

listening skills lead to Baldrige honors

THE CHALLENGES PRESENTED when gathering
customer feedback are as varied as the methods used for the actual
gathering. Every organization has its own set of obstacles to over-
come, and not every one will nd a way to do so. The ones that suc-
ceed, however, are rewarded for their perseverance.
The U.S. Army Armament, Research, Development and Engi-
neering Center (ARDEC) and the city of Coral Springs, FL, are well
aware of the payoff. The former tangled with the difculty of com-
municating with a customer base in a war zone half a world away,
while the latter ew in the face of governments focus on the here
and now to see the big picture when serving its customers. As a
result, the two became the rst nonprot organizations to receive
Malcolm Baldrige National Quality Awards.

June 2008 QP 29
 Customer Servicemen
Army weapons R&D center soldiers on with help from VOC

CUSTOMER FEEDBACK is a key Its critical to what we do, Denery said.

In 50 Words component of improvement. The challenge We understand that, to a large extent, the
Or Less has always been nding the most effective safety of the United States is dependent on
Using quality tools, ARDEC way of procuring it. As technology has grown how strong our military is. We want our sol-
fosters a culture of im-
increasingly sophisticated, organizations have diers to come home safely. Were keenly aware
provement while serving
tried comment cards, telephone surveys and that the lives of the warghters depend on the
a customer base that
includes the U.S. Army. any number of internet-fueled enticements to products we put in their hands. We dont want
Voice of the customer is get answers. it to be a fair ght on the battleeld.
critical to ARDECs suc- Those responses have proven difcult to
cess and ability to meet obtain, though, even while targeting a popu- Enterprising voyage
customer requirements.
lation that has web access at its ngertips. ARDECs focus on VOC manifests itself
For its efforts, ARDEC
was one of two nonprot Now imagine how daunting the task is to an through an approach dubbed enterprise ex-
organizations that received organization with customers whose ngers cellence (EE), which was constructed using
a 2007 Baldrige award. are more likely to be curled around the trig- a Baldrige framework and, in addition to
ger of a rie 6,000 miles away. VOC, employs lean Six Sigma (LSS), capabil-
Thats the reality faced by the U.S. Army ity maturity model integration (CMMI) and
Armament Research, Development and En- standards set by the International Organiza-
gineering Center (ARDEC), which considers tion for Standardization (ISO).
its emphasis on voice of the customer (VOC) But while ARDEC owes its recent honor
one of the reasons it was chosen to receive a to the implementation of EE that took place
Malcolm Baldrige National Quality Award in in 2005, Denery said, its Baldrige roots reach
2007the rst Department of Defense (DoD) back more than a decade prior and are inter-
organization to do so. twined with the frustration that accompanied
One of two nonprot enti- the organizations stalled improvement efforts.
LSS cost benet/ ties to receive the awardthe The DoD and the Army had studied the
avoidance / TABLE 1 rst time nonprots received benets of employing the Baldrige frame-
the honorARDEC is the R&D work as early as 1988 and, in 1994, decided
ARDEC cost Baldrige winner
Fiscal year benet/ cost benet/ center for armaments used by to implement a program to transform a
avoidance ($M) avoidance ($M)
the U.S. Army, Special Opera- wasteful environment often mocked for its
2001 $37.76 $0.47 tions Command and other or- extravagant spending on pedestrian items.
2002 $1,511.43 $2.67 ganizations. Tasked with out- We didnt want to have the $500 hammers
2003 $670.12 $3.22 tting U.S. soldiers stationed anymore, Denery quipped.
2004 $105.17 $2.43 around the world, including But after winning Baldrige-based com-
2005 $541.73 $3.77 the Middle East, its focus on petitions in New Jersey, which houses 84%
2006 $58.04 $3.35 VOC is less a luxury than a ne- of the ARDEC workforce, the organization
cessity, said Donelle Denery, examined its results and estimated that its
2007 YTD $9.30
chief of the strategic manage- point totals consistently fell in the middle
Total $2,933.55 $15.9
ment and process ofce. of the 1,000-point scale used in the Baldrige

30 QP www.qualityprogress.com
 BALDRIGE AWARD

competition. For an entity not willing to

accept middling performances of any
kind, hitting that ceiling and turning back
wasnt an option. So, after commission-
ing an LSS study and enlisting the help
of consulting rm Booz Allen Hamilton,
ARDEC was informed the ceiling would
remain as long as it continued to cling to
its product-based structure.
A competency-based system was pro-
posed, which, as Denery recalled, no-
body in the government had ever tried
before. It was like tiptoeing out on a
limb and not knowing whether the limb
was going to break. But, from the lean
Six Sigma study we had done, we knew
it was going to work. The amount of risk
is proportional to the reward. Thats
what put us over the top.
Even before that bellwether moment,
LSS had become ingrained in ARDECs
ARDEC DESIGNERS perform t checks of gunner protection kit weapon interfaces.
culture, to the point where more than
one-third of the workforce is trained in
the discipline today and every member of senior leader- central, which deploys eld and science advisors to the
ship is either Black Belt (BB) certied or BB trained. war zone. In addition, ARDEC has 700 integrated prod-
In fact, ARDEC sets the DoD benchmark for LSS uct teams, all of which include a customer representa-
thanks to a culture that achieved a cost benet/avoid- tive. But the prime example of the organizations efforts
ance of more than $2.9 billion between 2001 and the to heed the concerns and complaints of its customers
time of the organizations Baldrige application last
year (see Table 1). That number is closer to $3.2 billion
now, Denery said. The next best performance within Customer support triangle / FIGURE 1
the DoD is 40% of that total by an organization with a
e terface
workforce three times as large. s int rface t in of
es ec
in

of

ARDEC goes so far as to call LSS the improvement

j

c
Pro

Establish
Bus

ce

engine of EE, which is appropriate considering the requirements

MDTs APOs IPTs
program that serves as the cornerstone of the organiza- (Proposals)
tion rose from a BB project initiated by ARDEC Direc-
Systems
tor Joseph Lannon. The approach, which was imple- Constant engineers
mented in 2005, has among its components a quality interface
e s
re

management system that applies CMMI and ISO 9000 ut nt

Sa m )
qu rve

ec me
tis e
ire

x
standards to the organizations systematic processes E ire s)
fy nts
(S

qu view
u

and a balanced scorecard to drive its strategic intent. r e e

y

But the facet of EE that drives down product cost, ac- (R

Customers
cording to Denery, is VOC.

Finding their voice

When ARDEC underwent a thorough reorganization in MDT = market develpment teams
APO = ARDEC project ofcers
2004, one of the new ofces it created was warghter IPT = integrated product teams

June 2008 QP 31
 are its conferences with soldiers recently returned from
the war zone.
During what amounts to a 400-person focus group, sol-
diers talk about their experiences in the eld, which prod-
ucts worked well, which products didnt, and innovations
they could use in the future. The discussion also sheds
light on product uses that ARDEC, with all of its simula-
tions and other testing, never would have imagined.
When your life is on the line, soldiers become very
creative with how they use products, Denery said. We
may have designed a product for one purpose, and it
almost becomes like the pharmaceutical industry. You
develop a drug for Parkinsons disease, and you nd
out that it helps something else. That happens with our
products. We design them for one thing, but soldiers
get creative and nd out it works for something else.
We need that real-time feedback from them about
how a product is working, what we can do to improve
it, what they need next, and its a wonderful interface.
Those are great conversations for us.
All of those efforts fall within ARDECs customer
support triangle (Figure 1, p. 31), which ensures that
VOC is a constant inuence on the day-to-day process-
es within the organization. That approach is one of the
ways ARDEC achieves the latter part of its most criti-
cal task: getting superior products in the hands of U.S.
soldiers and getting them there fast.
During one of its listening sessions, ARDEC was
informed that soldiers perched in the top turret of a
Humvee might as well paint targets on themselves con-
sidering the unfettered sight lines enjoyed by enemy
Customer satisfaction
ratings / FIGURE 2
FY= Financial Year Target: FY05-3.70, FY07-3.75, FY09-3.80
4.0 Good
Average ratings
3.8
3.6
3.4
3.2
3.0
FY00 FY01 FY02 FY03 FY04 FY05 FY06 FY07
YTD
ARDEC average
Government best-in-class benchmark
Industry best-in-class benchmark
Strategic management system target
32 QP www.qualityprogress.com
snipers. That led to the creation of a gunner protec-
tion kita shielded turret that provides a signicant
barricade while still providing the weapons operator a
good view of the battleeld.
While ARDEC isnt capable of large-scale produc-
tion, Denery pointed out, it has metalworking facili-
ties that can create mock-ups and turn out items on a
limited basis. We started with that limited production
and then contracted with industry to produce them to
get them into the hands of the troops very quickly, she
said. We designed about 20 of the kits, got them out in
the eld, got feedback on what worked and what didnt
work, and now virtually every Humvee in the Middle
East has a gunner protection kit on it.
Read and react
Easing the process of gathering feedback from 6,000
miles away is ARDECs quick reaction task force
(QRTF), which created a formal process using a web-
based tracking tool that gathers questions and needs
from soldiers in the eld and puts them on the fast
track to a solution.
In the fourth quarter of scal year 2006 through the
second quarter of 2007, the QRTF closed out 80% of
inquiries in less than 72 hours. That contributed to
2007 customer satisfaction ratings well above gov-
ernment and industry best-in-class benchmarks (see
Figure 2).
Those numbers wouldnt be possible without total
buy-in from everyone in the workforce. But if the data
isnt enough proof of ARDECs focus on improvement,
theres the near-constant presence on the 10 Greatest
Inventions of the Year list handed out by the Army,
the award for Army Research and Development Lab
of the Year won two of the last four years and, now,
the Baldrige trophy the organization took home from
Washington, D.C., in April.
Earning one of the top accolades the quality world
has to offer doesnt mean ARDEC is going to stray from
its path, though. Instead, it plans on keeping improve-
ment opportunities rmly in its sights.
We have a very solid approach to what we do. We
have a deep deployment within our organization, we
have integrated and aligned strategies and operations
within ARDEC, and weve had multiple cycles of learn-
ing, Denery said. Its a very trite phrase, but quality
truly is a journey, not a destination. No matter how
good you are, theres always room for improvement.

Far-Sighted
Long-range focus allows city government
to celebrate Baldrige recognition
AMERICANS HAVE grown accustomed to
politics-as-usual, particularly during the nonstop cov-
erage of the current race for the White House. The
political world is one that seems to cater to whatever
might get a candidate or elected ofcial immediate re-
sults or a bump in the polls.
Thats what made the accomplishments of the city of
Coral Springs, FL, so astounding to those who saw rep-
resentatives from the community receive its 2007 Mal-
colm Baldrige National Quality Award. In the rst year
nonprot organizations could apply for the honor, Coral
Springs became the rst local government in the nation
to take home the award by proving the emphasis on the
present at the expense of the future doesnt permeate
every inch of the political scene.
The city made its rst move down the path to Baldri-
ge recognition in 1993 with the implementation of a total
quality management program. Prior to that, however, its
system fell in line with just about every other govern-
ment in the country as a reactive entity.
Eventually, it became clear that the usual way of
doing things wasnt getting the results city leaders
wanted. That gave rise to the rigorous planning pro-
cess Coral Springs ofcials and employees steadfastly
adhere to today, even when staring opposition in the
faceliterally.
Politicians often are paying the most attention to
the squeaky wheel in front of them. They make deci-
sions based on politics, said Susan Grant, the citys
director of HR. But the data analysis piece has really
allowed us to move away from that reactive mode. Yes,
there might be a single citizen in front of you at a com-
mission meeting complaining about this or that. But
our survey results might show that 99% of the residents
feel differently.
It really allows you to be thoughtful, to be proac-
tiveusing survey data, using other process data, us-
ing our environmental scan. All of that put together
allows us to plan in a meaningful way.
BALDRIGE AWARD
More accurately, it allows Coral
Springs to plan in a strategic way. The
In 50 Words
City Commission instituted the pro-
Or Less
During its improvement
cess of strategic planning in 1997, and efforts, Coral Springs, FL,
in doing so created an ongoing system de-emphasized immediate
that started and ended where any good results.
company nds its alpha and omega The resulting planning pro-
its customers, which the city denes as cess became part of the
citys corporate culture.
residents and businesses.
The shift in philosophy led
to robust nancial health,
Welcome feedback lower crime rates and a
The two-year planning process includes Baldrige award.
the creation of strategic and business
plans, which set priorities and objec-
tives, and the budget, which makes the
plans a reality by dening programs and services that
will serve the communitys citizens. But the bookends
holding up the strategy development and deployment
model are the citys customers, a focus evident in Cor-
al Springs multipronged approach to communicating
with its customer base.
No medium has gone unexplored by Coral Springs,
which disseminates information via its call center
OUTREACH PROGRAMS like this one at a local school are one way Coral
Springs communicates with its customer base.
June 2008 QP 33
 and website, e-mail, Coral Springs magazine, CityTV,
CityRadio and the consumer-friendly City Hall in the
Mall. On top of that are the citizen advisory commit-
tees and boardsone for each strategic priority
that meet monthly with a city staff member (usually a
department head) and the annual Slice of the Springs
meetings held in each of the citys six geographic seg-
ments to give residents and businesses one more op-
portunity to provide feedback.
That customer-centric approach has yielded impres-
sive results, including a 10-year run of residents qual-
ity ratings topping 90% and a satisfaction rating from
the business sector that hasnt dipped below 92% in the
last four years.
Laying the myriad choices at the feet of the citizenry
was actually the easy part of a dual-input setup that
also includes data analysis of a multitude of measure-
ments. The city didnt always emphasize the latter por-
tion, though, and it might not have if not for a sparsely
worded note from the Florida Sterling Council, which
awards a Baldrige-based honor in the state.
[The measures] have evolved over time, Grant
noted. On our rst feedback report from the Sterling
award, we had absolutely no strengths in any of the
results categories. Its pretty tough to see Strengths:
none noted. But thats where you start. You have to
nd out what youre not doing right and what you need
to improve on.
What emerged from those humble beginnings is a
measurement system that links to every activity out-
lined in the strategic and business plans. Process own-
ers are tasked with tracking the process measures,
which are aligned with in-process measures that deter-
mine whether the process will meet requirements.
A prime example of the system in action is the track-
ing of emergency medical service response times. Re-
sults are analyzed daily to determine whether variation
has occurred and, in turn, whether corrective actions
need to be taken. Total response time is then broken
down to determine the cycle times for four segments
intake, dispatch, turnout and traveleach of which
have their own performance goal.
Finally, that data is reported to the State of Florida
Department of Health in just one of a series of pro-
grams Coral Springs is involved in to see how it stacks
up against numbers gathered from across the state and
country.
Compare and contrast
Not too long ago, the city cast a wide net with its
benchmarking activities. Prior to the most recent
Baldrige cycle, however, the decision was made to be

Crime rate incidents per Residents quality

100,000 (composite index ratings for police
indicator) / FIGURE 1 department / FIGURE 2
15%
Good

93% 95% 95% 94% 94% 93% 92% 92% 91%

100%
15% 90%
80% Good
70%
15% 60%
50%
40%
5,093

5,168

4,886

15% 30%
4,391

3,988

20%
3,507

3,390

3,194

3,085

10%
3,001

2,731

15% 0%
1999 2000 2001 2002 2003 2004 2005 2006 2007
Police ACSI City 9 City 37
0%
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Coral Springs compares favorably to city 9 and city 37 (the
Coral Springs Broward County State National only other ICMA benchmark cities with a similar question
in their resident surveys), and the American Customer
The City of Coral Springs crime rate is lowest in Florida and the fourth Satisfaction Index for police service in suburban areas.
lowest in the nation for cities with populations of 100,000-499,999.
The city had the 10th lowest overall crime rate in the United States in
all categories.

34 QP www.qualityprogress.com
 BALDRIGE AWARD

more selective, despite the dearth of data among its

governmental brethren.
Finding comparison data in the public sector is not
easy, Grant said. Its probably not easy in any sector,
but we think its more difcult for us because govern-
ment doesnt have a long history of performance mea-
surement. In the past, we just kind of pulled compari-
sons from wherever we could get them. One of the bigger
improvements we made in this cycle of the application
was being more systematic about our comparisons.
That meant scanning the roster of the International
City/County Managers Assn. (ICMA) Center for Perfor-
mance Management and culling a list of 80 to 100 cit-
ies to the 20 highest-performing entities that are demo-
graphically similar to Coral Springs. And when certain
numbers arent available, the city doesnt hesitate to
contact those comparable cities to procure the data.
There are times, however, when the ICMA mea-
sures fell short due to differences in the participants
climates, economic situations or state laws. Coral
Springs addressed those shortcomings in 2005, when it
teamed with 10 local Florida governments to form the
Florida Benchmarking Consortium. The group, which
has since expanded to 37 members, makes it possible
for participants to compare performance data when
its necessary to include regionally specic factors.
But the city doesnt stop there. The police depart-
ment, for example, provides performance data to other
departments in Broward County, the Florida Department
of Law Enforcement and the Uniform Crime Reporting
System of the U.S. Department of Justice and FBI.
Those efforts led, in part, to a crime rate that has been
cut nearly in half over the last decade to 2,731 incidents
per 100,000 people (see Figure 1)the lowest crime rate
in the state and the fourth lowest in the nation among cit-
ies with populations between 100,000 and 499,999. As a
result, quality ratings for police have been above the 90%
mark every year since 1999 (see Figure 2).
On paper, those are just a lot of numbers, lines and
bars. In reality, it was the proof that sold the quality
movement to any skeptic that came along.
TECHNOLOGICAL INVESTMENTS make it easier for the Coral Springs Police Depart-
Getting over those humps was difcult at rst, ment to serve the community and contribute to a consistently high quality rating
Grant recalled. We had to produce results that backed for the department.

up what we were doing. If those didnt comeand they

didnt come right awayit was going to be a hard sell.
You have people asking why they should do business
this way. But when our results showed this was what
made our customers happy, it wasnt a tough sell.
WINNING APPROACH
Like how this years nonprot Baldrige award recipients attacked their
improvement projects? Did they miss some opportunities? Make your
opinion heard in the articles comment section at www.qualityprogress.
com or e-mail editor@asq.org.

June 2008 QP 35
Incredible
Injuries, costs persuade a
Boeing team to develop a
safety plan

In 50 Words
Or Less
To eradicate injury rates
and workers compensa-
tion costs, a C-17 team
embarked on a world-
class safety journey.
Brainstorming, bench-
marking and shbone
by Nicole Adrian,
diagrams helped identify contributing editor
root causes and the nal
solution.
The solution, the Inte-
grated Safety Manage-
ment System, yielded
substantial cost savings
and a great reduction in
injury rates.
 TEAMS

Journey
IN RESPONSE TO employee safety concerns and
rising workers compensation costs, a team at Boeings
C-17 site developed a solution to prevent injury and save
money.
The journey of the C-17 World-Class Safety Team from
Boeingan aerospace and defense corporationended
with the implementation of the Integrated Safety Man-
agement System, which has 12 essential elements, in-
cluding open communication processes and challenging
goals and objectives.
Throughout the journey, the team repeatedly used
many quality tools and methods, including benchmark-
ing, brainstorming and value stream mapping. Compara-
tive metrics, force eld analysis and a shbone diagram
helped create the nal solution.

June 2008 QP 37
 The C-17 program at Boeings Long Beach, CA, facili- A 12-question employee safety survey was conduct-
ty employs approximately 6,500 people to manufacture ed, and the results indicated an average score of 2.2.
15 C-17 aircrafts each year. The team chose the world- Results for world-class companies are between 2.5 and
class safety project for two reasons: it recognized that 3.0. In 2004, a consultant rated the C-17 program as re-
keeping the workforce safe shows how a company active, based on its position to the far left of the safety
values employees, and how signicant safety is in run- culture maturity curve (see Figure 1).
ning a successful business. The safety team used numerous quality tools to se-
The safety team tied for second place with two oth- lect the project and determine why it was important.
er teamsincluding another from Boeingto earn a These tools included:
silver award at the 2007 International Team Excellence Injury frequency and cost data.
Competition. The competition is sponsored by ASQs Consultant, benchmarking and culture survey data.
Team and Workplace Excellence Forum, and the pre- Fishbone diagram and brainstorming.
sentations are judged annually at ASQs World Confer- Vision and principles.
ence on Quality and Improvement. Value stream mapping.
Before launching the project, the team reviewed Project strength matrix.
the data on injury frequency and associated costs to Gap and force eld analysis.
help understand the C-17 sites current state. Team Safety culture maturity model.
members benchmarked best-in-class practices exer- Additionally, the team used a Six Sigma strength
cised by premier companies and used gap analysis to analysis tool to assess the likelihood of the projects
help understand the programs strengths and weak- success. The tool analysis elements included impact
nesses. Additionally, strategies and tactics were iden- on employees, customers, cost, competitiveness and
tied and deployed, and a consultant from an outside resources.
company helped deploy the safety program. The safety journey project was ultimately selected
because the C-17 sites injury rates were unacceptable.
Alcoa, the company with the best rates, was 15 times
Intangible and tangible better. C-17s lost workday case rate comparisons for
results / FIGURE 1 2004 were 1.97, while Boeing came in at 1.65, DuPont
at 0.39, Raytheon at 0.21 and Alcoa at 0.12 (see Figure
Na
tu 2). By reducing the rate by 10% per year, it would take
ral Safety culture maturity curve
ins the company at least six years to obtain a lost workday
tin
cts case rate of 1.0.
Sup
erv Rising costs were another important concern to
isio
n address. C-17 workers compensation costs increased
Self 74% from 2002-2003, which added costs to each C-17
2004 state
Teams aircraft produced.
2006 state
From project selection to nal implementation,
Reactive Dependent Independent Final goal numerous stakeholders from different areas were en-
Lack of Management Individual Open gaged in the project. These stakeholders included em-
management commitment recognition communications
involvement Value all people Personal value Help others ployees from:
Delegated to Training Personal conform Leadership.
safety Dened task knowledge, Others keeper
manager Rules/ commitment, Networking Production operations management.
Compliance procedures and standards contributor United Aerospace Workers (UAW).
is the goal Leader Internalization Care for others Safety.
Safety by expectations Care for self Team culture
natural Process Practice, habits Organizational U.S. Air Force.
instinct emphasis, Increased pride Stakeholder involvement included brainstorming,
Common and goals communications
sense Condition of benchmarking, training and focus meetings. This selec-
Blame employment tion was based on direct involvement in manufacturing
culture
processes and program leadership responsibilities. Key

38 QP www.qualityprogress.com
 TEAMS

needs from stakeholders ranged from involvement in

project design, deployment and oversight, to produc-
Comparative metrics
tion quality, cost and scheduling. Stakeholder needs
analysis of data to select
were quantied using a scoring tool that measured the nal root causes / FIGURE 2
relationship needs and impacts against project and 2004 lost workday case rate comparison
business criteria. 10
9 good

Lost workday case rate

8
Hopes for the future
7
The safety team established an aggressive goal of
6
achieving 50% improvement across four different inju-
5
ry and cost metrics: lost workday cases, lost workdays, 4
workers compensation allocations and recordable 3
Best Benchmarked
case rate. Team members wanted the project to have 2
in class industry
a direct effect on four goals: eliminating employee in- 1
juries, shaping the safety culture, reducing cost and 0

a
on

tin

nt

ic

an

TX
ie
co

,W
sustaining future business.

tr

ra
Po

m
he

ar

og

o,
ec
Al

t,
og
um

on
M

Du

ni
yt

et
ol

El

pr

to
Injury, cost and culture analysis tools were used to Ra

nt
hn
ed

er
Gr
l
ra

An
Re
17

Ev
he

up
ne
Te

C-
assess the project degree of impact on goals. Based on

n
ck

hr
Ge

Sa
d
Lo

rt
ite

benchmark research, injury reduction was anticipated
to be 25% in the rst year of the project. Standard in-
jury rate metrics were used as the monthly measure-
ment tool. The safety culture improvement was also
expected to be 25% in the rst year.
Additionally, cost reductions were expected to be
in the millions in each of the rst two years, primarily
due to injury prevention. The injury, culture and cost
improvements reduce risk and minimize losses.
The C-17 team estimated a labor loss improvement
of 1.5% for the production workforce in the rst year,
which was conrmed by reduced lost workdays that
provided competitive advantage in acquiring new busi-
ness.
Stakeholders were expected to be impacted as well.
These impacts across all stakeholder groups typically
involved:
Increased time for involvement in safety activities.
Reduced injuries and associated costs.
Increased productivity due to a reduction in labor
loss.
Increased cost due to initial safety improvements.
Final root causes and a solution
After using quality tools and methods to determine
possible root causes, the team used the same tools to
pinpoint the nal root causes. These tools included:
Brainstorming: Provided the team a means to so-
licit different ideas and opinions regarding safety im-
provements.
No
Un
Boeing large
Aerospace companies
airframe sites
Organization
Benchmarking: Provided comparative metrics with
other world-class companies.
Gap analysis: Identied the current state and gaps
to attain future state.
Force eld analysis: Categorized the driving forc-
es for and against change, and established priority of
factors on each side of the balance sheet.
Fishbone diagram: Identied potential causes in
an orderly manner and summarized major causes un-
der six main categories.
These tools and data, as well as stakeholder involve-
ment, found seven nal root causes for why the C-17
program had so many lost workdays and high work-
ers compensation costs. These root causeswhich
the team validated using the same methods, tools and
datawere:
No personal accountability for safety performance.
Lack of communication.
Inadequate procedures, training and rules.
Lack of management commitment.
Not a business priority.
No safety integration.
Complacent safety culture.
To select the nal solution, the team used three
methods: comparative metrics, force eld analysis and

June 2008 QP 39
 a shbone diagram. First, using comparative metrics, The system was developed after establishing 12 essen-
the C-17 program was evaluated against premier com- tial elements:
panies with low occupational injury and illness rates, Visible, demonstrated commitment.
aerospace industry competitors and other Boeing large Clear, meaningful policies and principles.
airframe sites. Challenging goals and objectives.
The force eld analysis data indicated the forces High standards of performance.
working against improving the C-17 safety program Line management accountability.
were three times higher than the forces for those im- Integrated organizational structure.
provements. Forces against, including lack of man- Supportive safety staff.
agement involvement, no personal accountability for Progressive motivation.
safety and lack of corrective action, outweighed the Continuous training and development.
forces for, including team based culture, safety being a Thorough investigations.
priority and all employees being safety leaders. Effective observations.
Finally, a shbone diagram evaluated various ele- Open communication processes.
ments in categories of measurement, people, product, The need to implement this system was validated
environment, method and material to identify causes through a consultant assessment. The Integrated Safe-
for the lack of a robust safety program. This analysis ty Management System was deployed through the C-17
showed what caused the lack of a robust safety pro- executive safety council, which includes the leadership
gram, including weak safety goals, unsafe acts and team, UAW leadership, employee involvement/safety
poor safety communications (see Figure 3). awareness teams, Air Force customers, the safety staff
After assessing the current state, determining root and executive-led subcommittees.
causes and involving stakeholders, the team developed
and deployed a new C-17 Integrated Safety Manage- Addressing employee concerns
ment System, which team members believed was the Employees had some concerns about the proposed
foundation for achieving a world-class safety culture. plan. Some saw the Integrated Safety Management
System as another avor of the
month program and believed
Cause and effect analysis methods and tools it was an unnecessary culture
used to identify possible root causes / FIGURE 3 change.
The safety team addressed
Causes Effect the resistance through numer-

Measurement People Product

ous means of communication,
Lack of management and training was provided to
Oil metrics Conguration of aircraft
commitment production line management on
Safety audits and involvement Complexity of product
the elements of the new system,
Lack of safety training Moving surfaces
Weak safety goals and then was owed down to all
Aging workforce Repetitive work
Data systems employees.
nd al

Unsafe acts
n
co ysic

Production tooling
itio

Additionally, four subcom-

No personal
Ph

Hand tools Lack of mitteesincident investigation,

accountability
Assembly orders
Production schedule priority
Out of position tasks
Complacent safety culture Lack of procedures
and policies
No discipline to process Poor safety
communications
Unsafe conditions
Incident investigation
(root-cause analysis)
Business priority
No safety rules
Environment Method
robust
safety rules and procedures, safety
Hazardous materials program observations, and activities and
involvementwere developed to
Parts in large containers
integrate the elements of the plan
Large and bulky into the nal solution.
subassemblies
Three of the four subcommit-
Hazardous waste tees have developed procedures
and databases to govern their
Material
processes and evaluate their re-

40 QP www.qualityprogress.com
 TEAMS

sults. These provide evidence that the changes made are

being sustained. The fourth subcommittee uses various
safety activities and communication methods to involve
employees and improve the safety culture.
Numerous systems were created and installed to
measure and sustain the results of the safety system.
They include the safety observation database, which
validates by comparing monthly reports from the da-
tabase, and the vision support plan and safety deck,
which measure numerous aspects of safety from the
business-wide level to the site level and are reviewed
monthly.

Seeing results
The Integrated Safety Management System resulted in
numerous intangible results, including the shift from a
reactive state to an independent one in which the ma-
jority of individuals take responsibility for their own
MEMBERS OF THE C-17 World-Class Safety Team were (front row, from left) Michelle
safety (see Figure 1, p. 38).
Alcaraz, Teresa Jugan, Jana Rountree, Anna Garcia, (back row, from left) Kenneth
The C-17 program has also seen evidence of move- Elmore, Ken Drew, Mike Aber and Cliff Tanigawa.
ment toward the nal goalan interdependent state
in which employees act as each others keepers, take
pride in the organization and communicate openly. rmed by feedback during the regular meetings.
Other intangible results included improvements in The safety team credits its success to its use of the
safety, morale and pride (the site earned multiple team training guides, which added skills necessary for
awards for its world-class safety journey), leadership working together effectively and identifying indicators
and commitment, and stakeholder collaboration. These of team dysfunction. Team members also give credit to
intangible benets are quantied in the feedback from their self-management style. Each member worked in-
the production operations employee satisfaction sur- dependently and with other co-workers to collaborate
vey, which showed a 25% increase in feeling safe from on the project. Effective communication was another
accidents and health hazards in my work area. key to team success. Routine project interactions al-
Additionally, numerous tangible benets were lowed team members to review progress by identifying
achieved: quality, cost, schedule and risk issues.
The cost savings in workers compensation alloca- The tremendous progress on injury reduction and
tions was in the millions. culture change gave the team something to be proud
The business unit goal for recordable injury rate re- of. The C-17 program is now considered an example of
duction was 9.5% and the C-17 programs rate was excellence for safety, and the teams efforts have been
actually reduced by 32%. used as a benchmark internally and externally.
The lost workday rate goal reduction was 30% and
the C-17 programs rate was actually reduced by
46%.
Safety observation is being conducted by leadership
to eliminate at-risk work practices.
These results were shared with the various stake-
holders involved in the project. The team determined
the results were successfully communicated and the
PERSONAL ACCOUNTS ONLINE
To read rst-person accounts of the World-Class Safety Teams project
current stakeholders were satised by comparing and experience at the International Team Excellence Competition,
previous safety culture data to current. This was con- visit www.qualityprogress.com.

June 2008 QP 41
 Two
Than
In 50 Words
Or Less
Organizations need
nancial management
controls to comply with
the Sarbanes-Oxley
Act (SOX).
ISO 9001 can be
leveraged for SOX
compliance in docu-
mentation, internal
audit processes, and
internal corrective
and preventive action
processes.
This can save money by Vivek Vic Nanda
by eliminating waste
caused by duplication.

42 QP www.qualityprogress.com
 SARBANES-OXLEY ACT

Are Better
One
ISO 9001 and SOX commonalities
help enhance internal controls
MANAGEMENT NEED and my personal curiosity recently led me to
take a closer look at the Sarbanes-Oxley Act (SOX). My work pointed out similarities
and differences between SOX and ISO 9001 and provided a list of ISO 9001 clauses
that, if implemented, directly support SOX compliance.
The purpose of this exercise was to understand whether and how an organizations
ISO 9001 based quality management system (QMS) could be leveraged for SOX com-
pliance, making SOX compliance more efcient. By nding synergies between SOX
and ISO 9001 and eliminating duplication of implementation efforts, I found organiza-
tions can reduce the cost and resource needs associated with separate and discon-
nected efforts.

June 2008 QP 43
 Why SOX?
Congress enacted SOX in 2002 in response to major
corporate and accounting scandals, such as those
that occurred at Enron and Tyco. These scandals
resulted in a decline of public trust in accounting
and reporting practices.
SOX has the following stated objectives:
Restore investor condence in U.S. nancial mar-
kets damaged by scandals.
Improve corporate accountability.
Improve transparency and quality of nancial re-
porting.
Provide reasonable assurance that preparation of
nancial statements is in accordance with generally
accepted accounting principles.
SOX also led to the establishment of a new quasi-
public agency, the Public Company Accounting Over-
sight Board (PCAOB), which is charged with oversee-
ing, regulating, inspecting and disciplining accounting
rms in their roles as auditors of public companies.1
The SOX legislation itself is wide-ranging and estab-
lishes new or enhanced standards for all U.S. public
company boards, management and public accounting
rms.
The act contains 11 sections. The two most interest-
ing deal with corporate responsibility for nancial re-
ports and management assessment of internal controls.
1. Section 302Corporate responsibility for
nancial reports: This section requires certifying of-
cers (generally the CEO and CFO) to establish and
maintain an internal control framework over nancial
reporting. The word establish refers to documenta-
tion, deployment, ongoing monitoring and assessment,
and reporting.
Disclosure is required when changes made to inter-
nal controls have had or will have a material effect on
nancial reporting. This disclosure includes reasons
for the change in controls when the change was made
to correct material weakness (a signicant deciency
that prevents reasonable assurance that misstatements
will be prevented or detected in a timely manner).
Section 302 also calls for quarterly and annual cer-
tication.
2. Section 404Management assessment of in-
ternal controls: This section mainly deals with the
assessment of the effectiveness of the internal control
framework. It requires annual assessment by manage-
ment and independent auditors, and reporting of results.
44 QP www.qualityprogress.com
Other important points about SOX that might not
be obvious are:
Company management, not the external auditors,
has nal responsibility for nancial statements and
internal controls. Therefore, external audits are nei-
ther sufcient nor a safeguard.
Internal audits can help management verify that
controls are in place.
Audits are not the answer either, because they are
periodic and based on a sampling approach.
Preventive and detective controls therefore need to
be embedded in the system.2
Achieving compliance
The Securities and Exchange Commission (SEC) iden-
ties the Committee of the Sponsoring Organizations
of the Treadway Commission (COSO) framework as
a method for achieving compliance with SOX require-
ments.
COSO, established in 1985, is a private organization
that promotes corporate governance and improves the
quality of nancial reporting. Therefore, it predates the
SOX legislation. COSO authored a document called the
Internal ControlIntegrated Framework (IC/IF) to
assess effectiveness of internal controls in a company.3
This has become the de facto internal control frame-
work adopted by most companies implementing SOX.
The COSO IC/IF denes ve areas that, when imple-
mented, can help support the requirements set forth
in SOX:
1. Risk assessment.
2. Control environment.
3. Control activities.
4. Monitoring.
5. Information and communication.
Use of the COSO IC/IF is not mandated but is
strongly endorsed by SOX.
In todays organizations, IT is an important piece of
the control framework because automated processes
and complex transaction processing applications rely
on the underlying IT infrastructure. Therein lies the
main deciency of the COSO IC/IF: Although it includes
IT control considerations, companies have found this
guidance to be insufcient.
To ll this need, the IT Governance Institute au-
thored an IT control framework, Control Objectives
for Information Technology (COBIT).4
COBIT is the most widely used IT control frame-
 SARBANES-OXLEY ACT

work in companies implementing SOX requirements.

There are other IT internal control frameworks,
Relationship summary / FIGURE 1

such as ISO 17799, which is a code of practice for in-

formation security management, and the Information
Technology Infrastructure Library.
Key areas of IT responsibility in SOX implementa-
tion include:
Understanding the companys internal control pro-
gram and its nancial reporting process.
Mapping the IT processes that support the internal
control program and nancial reporting process.
Identifying risks related to these processes and IT
systems.
Designing and implementing controls to mitigate
the identied risks.
Documenting and testing IT controls.
Updating IT controls as needed, when internal
controls change or the nancial reporting process
changes.
Monitoring IT controls for continued effectiveness.
A summarized relationship between SOX, COSO IC/
IF and COBIT is shown in Figure 1.
SOX and ISO 9001
Figure 2 shows the close similarities between not only
IT control Internal control SOX compliant
framework framework nancial
(for example, Requires (for example, Requires reporting
COBIT) COSO) system
(for IT
aspects
only)
some key requirements in SOX and ISO 9001, but also
the governance and auditing processes supporting
both.
Just as the SEC establishes the rules and provides
the high level governance for SOX, the International
Organization for Standardization (ISO) performs a
similar role for ISO 9001. The ANSI-ASQ National Ac-
creditation Board (ANAB) plays a role similar to the
PCAOB by serving as the U.S. accreditation body for
management systems.
ANAB accredits registrars for ISO 9001 QMSs
and ISO 14001 environmental management systems,
as well as a number of industry-specic require-
ments. Finally, just as nancial auditors provide
independent verication of the internal controls with-
in a company in accordance with SOX requirements,

SOX and ISO 9001 / FIGURE 2

SEC ISO

SOX ISO 9001

Rules, overall (law) Rules, overall (quality management standard)
oversight oversight

Oversight Financial Oversight Quality auditors

PCAOB ANAB
auditors (registrars)
Guideline/ Guideline/
reference reference

COSO ISO 90003

Verify Verify
framework ISO 19011
compliance compliance
COBIT
PCAOB Auditing
Standard No. 2

Company Company

ISO 90003 is ISOs guideline for the application of ISO 9001 to computer software.
ISO 19011 is ISOs guideline for quality and environmental management system auditing.

June 2008 QP 45
 quality auditors provide independent verication of a
companys QMS to ISO 9001 requirements.
While drawing such comparisons between the gov-
ernance and auditing processes of SOX and ISO 9001 is
an interesting exercise and certainly enhances our un-
derstanding of them, the similarities dont end there. In
fact, the insight becomes much more valuable as you
begin to peel the onion and take an even closer look.
Table 1 lists some similarities and differences between
SOX and ISO 9001.
Lets look at some specic areas of ISO 9001 that
organizations can easily leverage to directly contrib-
ute to SOX compliance. The resulting consolidation
of effort can translate into signicant cost savings and
elimination of waste in duplicated efforts.
First, SOX calls for quality control of nancial re-
ports, and processes and systems that combine to pro-
vide inputs to those reports. ISO 9001 processes that
can be reused for SOX compliance include but are not
limited to:
Documentation management and control, including
versioning, review and approval.
Internal audit process.
Internal corrective and preventive action processes.
The ve elements of COSO IC/IF can be mapped to
ISO 9001 clauses and subclauses, for example:
1. Control environment4.1, 5.3, 5.4.1 and 6.2.1.
2. Risk management8.4.
3. Information and communication5.5.1, 5.5.3, 7.2.3
and 7.4.2.
4. Monitoring8.2.1, 8.2.3, 8.2.4 and 8.4.
5. Control activities5.6, 8.5.1, 8.5.2 and 8.5.3.
This is not a complete mapping and is provided for
illustrative purposes only. There are additional clauses
in ISO 9001 to which COSO framework requirements
can be mapped.
This means that in the process of using COSO IC/IF
for achieving compliance to SOX, an organization can
leverage whatever has already been implemented for
the corresponding ISO 9001 clauses.
ISO 9001 also provides an excellent mechanism to
manage risk in an organizationand risk, or rather
the avoidance of risk, is the primary focus of SOX.
Therefore, the discipline and transparency of opera-
MORE ON SOX
Several articles on the Sarbanes-Oxley Act have appeared in QP since 2003.
Access them at www.qualityprogress.com.
46 QP www.qualityprogress.com
tions that an ISO 9001 compliant QMS provides is a
powerful tool for SOX compliance.
Further, by virtue of the internal auditing (subclause
8.2.2), measurement (subclauses 8.2.3 and 8.2.4) and
customer satisfaction requirements (subclause 8.2.1)
in ISO 9001, the standard provides a good measure
of the status and effectiveness of an organizations
processes. This directly contributes to validation of
internal processes and controls, which is the area of
primary interest for SOX, as well as a public companys
nancial auditors.
Finally, ISO 9001 places a great deal of emphasis on
continual improvement. This again directly contributes
to the primary focus of SOXthe strength of internal
controls within the organization, the continual valida-
tion of those controls for robustness and strengthening
of those controls as needed.
Industry experience
Organizations are increasingly realizing that the exist-
ing QMS (typically an ISO 9001 based QMS) in compa-
nies can be enhanced for nancial management report-
ing and assurance.
There are, however, limited case studies or lessons
learned reported from the integration of the two sys-
tems. Primarily, this is because nance personnel are
not generally involved with QMSs, so they might not
see a link. Also, quality assurance personnel might not
effectively communicate or recognize the value of link-
ing the two systems.5, 6
At the same time, companies that have been able
to leverage their QMSs for SOX compliance report ve
key areas in which quality personnel have been able to
help nance personnel with SOX:
1. Supporting nancial operations and controls by
identifying nonvalue added costs, achieving early
risk identication and corrective and preventive ac-
tions, and understanding operational processes.
2. Building process awareness and mentoring on qual-
ity techniques, such as process mapping for nance
processes.
3. Managing risk.
4. Establishing an internal audit program.
5. Developing business process metrics.
These organizations also report the following key
lessons learned from the SOX compliance efforts:
Section 404 has helped increase awareness
of or lack of internal controls.
 SARBANES-OXLEY ACT

One size does not t all. Senior management of
smaller companies should tailor the controlsand
their assessment and testingto the size of the or-
ganization.
Signicant cost can be expected in the rst year of
SOX implementation.
The commonalities between SOX and ISO 9001 are
ments, Public Company Accounting Oversight Board, 2004.
2. Scott Green, Managers Guide to the Sarbanes-Oxley Act: Improving Internal
Controls to Prevent Fraud, John Wiley, 2004.
3. Committee of the Sponsoring Organizations of the Treadway Commission,
Internal ControlsIntegrated Framework, 1992.
4. Control Objectives for Information Technology, IT Governance Institute,
2004.
5. Sandford Liebesman, Quality Practitioners and Effective Corporate Gover-
nance, Quality Progress, March 2004.
6. Liebesman, Use Management Tools to Mitigate Risk from SOX: Can a QMS/
EMS Help You Deal With the SOX Act? Informed Outlook, 2004.

not only in the areas of governance and auditing pro-

cesses already described in this article, but also in the
specied requirements.
Given that ISO 9001 is by far the worlds de facto
QMS and that most public companies are already reg-
istered to ISO 9001, these organizations can rapidly
achieve SOX compliance or enhance the robustness
of their internal controls for SOX by leveraging their
existing QMSs.
REFERENCES
1. PCAOB Auditing Standard No. 2: An Audit of Internal Control over Financial
Reporting Performed in Conjunction With an Audit of the Financial State-
BIBLIOGRAPHY
Liebesman, Sandford, John Walz and Paul Palmes, Sarbanes Oxley Law:
Using QMS/EMS to Reduce Business Risk, QuEST Forum Best Practices
Conference, September 2004, Dallas.
Palmes, Paul, and Sandford Liebesman, Qualitys Path to the Boardroom,
Quality Progress, October 2003.
Stimson, William, Sarbanes-Oxley and ISO 9000, Quality Progress, March
2005.
VIVEK VIC NANDA is a quality manager for Motorola
Inc., Horsham, PA. He earned a masters degree in
computer science from McGill University, Montreal,
Quebec. Nanda, a senior member of ASQ, is an ASQ
certied quality manager, quality auditor and software
quality engineer. He wrote ISO 9001:2000 Achieving
Compliance and Continuous Improvement in Software
Development Companies (ASQ Quality Press, 2003).

Comparison of SOX and ISO 9001 / TABLE 1

SOX ISO 9001

Specify requirements to
meet but not how to meet
them (nonprescriptive)
Severity of audit
nonconformances
Yes. Companies use guidance documents such as
COSO and COBIT to implement SOX requirements.
Major (material weakness); minor (signicant
deciency, internal control deciency); or
observation.
Yes. Companies use documents and books that
offer ISO 9001 implementation help.1, 2
Major, minor or observation.

Consequence of major Management is not generally permitted to Precludes the company from being registered to
Similarities

nonconformances conclude that the companys internal control over ISO 9001 until nonconformance is resolved.
nancial reporting is effective (as of the end of
most recent scal year) if there was a material
weakness reported. Permissible exclusion:
Despite a material weakness, management can
report internal controls to be effective only if
internal controls were changed to eliminate the
condition, the control change was sufciently
in advance of the as of attestation date in the
management report to the nancial auditors,
and the revised controls have been tested for
effectiveness over a period of time.
Compliance Mandatory for public companies. Mandatory only if a company chooses to register
to the ISO 9001 standard.
Differences

Legislation versus standard Legislation. Standard.

Required documentation SEC and PCAOB have not specied what or ISO 9001 clearly identies required procedures.
how much documentation is necessary for SOX
compliance.

1. Nanda, Vivek, ISO 9001:2000 Achieving Compliance and Continuous Improvement in Software Development Companies, ASQ Quality Press, 2003.
2. ISO 90003:2004, Software EngineeringGuidelines for the Application of ISO 9001:2000 to Computer Software, ISO, 2004.

June 2008 QP 47
Scorecard helps Sears simplify
QMS compliance, supports
continuous improvement efforts
In 50 Words THERES ONE TOOL you wont nd for sale at Sears.
Or Less One of the retail giants divisions has started using a qual-
After achieving ISO 9001
certication, Sears Home
ity management tool extensively to maintain and improve its
Services needed a way quality management system (QMS).
to monitor its quality
management system in Sears rst began its ISO 9001 efforts nearly a decade ago.
its regions.
Managers and executives
Since then, more than 450 units within Sears Home Services
developed a scorecard to have registered to ISO 9001.
measure the health and
effectiveness of process- Once the QMS was in place, the challenge became how to
es within the division.
The scorecard improved
continually improve it. How can a corporation as colossal and
internal communication complex as Sears motivate its employees to continually do a
and motivated employ-
ees to continuously better job in so many different locations around the country?
improve.
 PROCESS IMPROVEMENT

Whos
KEEPING
Score?
by Rebecca
M. Neenan

June 2008 QP 49
 The company keeps score, and it keeps score often.
Specically, Sears Home Services has created a unique
tool called the QMS compliance scorecard.
In basic terms, the scorecard takes a snapshot of
how the divisions regions are keeping up with compli-
ance, said Cris Shale, ISO program manager for Sears
Home Services.
The scorecard works as a basic grading system
for the different regions and their managers, enabling
them to see areas of success and places to improve.
The scorecard also provides an easy and efcient
mechanism to quickly communicate with executive
management on the health of the QMS, not the com-
panys nancial performance.
Scorecard origins
In January 2005, the executive team requested a way to
communicate the ongoing status of the QMS.
With more than 450 Home Services locations under
the QMS scope, it became difcult for the executive
team to monitor the health of each individual location
solely based on audit reports, Shale said.
Because ISO 9001 terms
are not always common
knowledge, the company
needed a method for
The scorecard has motivated
monitoring and report-
ing management reviews,
employees to remain compliant
corrective and preventive
action, and its internal au-
to processes and supported
dit program to all levels of
management.
continuous improvement.
After investing money in
ISO 9001 efforts, standards
and management systems were implemented, and Sears
had a way to ensure things were done the same way in
all of its operations. But there wasnt a way to measure
how effective and healthy the processes were.
The scorecard became that clear-cut solution.
The level of accountability and focus of ISO 9001 var-
ied by region. Sears didnt want one regional manager
engaged in the QMSdriving the intent of ISO 9001 and
continuous improvementmore than another regional
manager, said Joan Trach, vice president of operations
and administration for Sears Home Services.
The idea of developing a scorecard came from the
managers and executives dedicated to continuous im-
provement, who realized a visual way of presenting this
50 QP www.qualityprogress.com
evaluation was needed, explained Dennis McQueeney,
East region operations manager.
Corporate managers responsible for ISO 9001 and
continuous improvement ll the scorecard out while
associates in the eld see the results and understand
what it takes to get an A score. Based on this meth-
od, employees cannot cheat on their scorecard to get a
passing grade. Their role is to understand the method
and requirements to maintain their QMS.
During the rst two years of using the scorecard,
employees were feeling out the tool to see how it
worked, McQueeney said.
I think our employees are growing into it even as
we speak, McQueeney said. But were getting to the
point where were becoming mature and were pretty
condent that things wont change now.
The scorecard has evolved just like any tool or
technology evolves: Quantitative measuresactual
metricsare being used more to take subjectivity out
of the picture.
The scorecard began tracking progress by the cal-
endar year, but today it is being transitioned to corre-
spond to Sears scal year calendar. This aligns with
the companys prot and loss statements. The delivery
business was also added to the scorecard in the third
quarter of 2007.
Out of this, the scorecard has become a perfor-
mance management tool of the highest level and a way
to see how well regions remain compliant.
How it works
Sears uses the scorecard approach to monitor the
Home Services facilities currently under the ISO 9001
scope in six regions. This includes the following opera-
tions:
Carry-in and in-home services.
 PROCESS IMPROVEMENT

Parts distribution centers.

Region truck assortment organizations.
Sears scorecard results / TABLE 1
National parts sourcing center. A scores increased 771%.
Sears technical assistance center and li-
F scores decreased 57%.
brary imaging system.
Inventory rebalancing center. QMS scores for corporate businesses increased 31%.

Corporate headquarters. QMS scores for carry-in region 1 increased 1.1%.

The individual scorecards for the opera-
QMS scores for carry-in region 2 increased 21%.
tions are completed each quarter during the
scal year and measure several key con- QMS scores for in-home region 1 increased 10%.
cepts, including: QMS scores for in-home region 2 increased 23%.
Recent third-party audits (whether the
QMS scores for in-home region 3 increased 26%.
district has passed or failed).
Management reviews. QMS scores for in-home region 4 increased 33%.
Timeliness of completion of manage-
ment reviews.
Corrective and preventive actions. aged to perform a self-assessment using the same
A units ability to adhere to several procedures. scorecard method.
Preventive maintenance of equipment. Following the Sears Home Services evaluation
Environmental and safety requirements. teams assessment, the units have the opportunity to
Most areas of the scorecard allow grades of either review the results and report any errors the team might
an A or F, while other areas allow for a C. Some of the have made. The nal version is then issued.
sections are color-coded using the basic three levels Shale and Don Munn, the acting manager of compli-
of green (passing), yellow (needing improvement) and ance and ISO 9001 for Sears Home Services, and others
red (failing). on the ISO 9001 compliance team are in charge of as-
For example, in terms of the corrective action re- sessing and computing the scores. They have become
port (CAR) process, there are three grades for putting procient in lling out the scorecards by trying their
a CAR into process in a certain amount of time. If a best to strip all subjectivity out of the process.
CAR has been put into the system within 90 days, that Assessments of the QMS categories are done by re-
area then receives an A. In 120 days, the area gets a C. viewing records in the respective databases. If theres a
Any more than 120 days without a CAR being entered question on the data, an e-mail can be sent to the own-
results in the grade of F. er of the data (usually the ISO 9001 management repre-
One of the areas graded is equipment. For this, the sentative) for clarication before a grade is assigned.
scorecard looks at the percentage of tools requiring
maintenance. An operation can have up to 10 pieces of How it helps
equipment past due for maintenance and still receive a C, The scorecard has made communication among corpo-
but they must have no equipment past due to earn an A. rate executive management, continuous improvement
The Sears tool grades all areas measured and trans- managers and eld managers responsible for the QMS
lates them into the three colors: red for a problem in much easier. It has motivated employees to remain com-
the QMS, yellow to signal room for improvement and pliant to processes and support continuous improve-
green to designate the area as healthy. ment efforts.
Although it started as a basic concept, Sears has made It really applies to the ISO 9001 side of things. It
some tweaks and additions along the way. For example, makes sure that were cognizant of who has been keep-
QMS and compliance categories have been added, and ing up with the various parts of the QMS, said Randy
the grading has been adjusted to raise the bar. Tuggle, former region manager of process and compli-
Sears added another step to the scorecard pro- ance. Because youre graded once a quarter, you then
cess. The blank scorecard template is made available know whether you need to work on a particular thing
to the eld, and the units being graded are encour- so people are always improving.

June 2008 QP 51
 PROCESS IMPROVEMENT
For example, if executive management sees yellow
and red on the sheet, it can immediately see which unit
contributed to that score and go to that level to deter-
mine the reasons the scores were low.
If the scorecard shows red in an individual unit, the
unit might fail an upcoming surveillance audit.
The score then reduces the units overall annual
performance. It is in the employees best interest not
only to maintain the standards of the QMS, but also to
know how the unit is being measured.
The scorecard has become a
performance management
tool of the highest level.
The scorecard also helps determine where more time
might be needed for mentoring, coaching or training.
Because all areas of the Sears Home Services divi-
sion are kept in check and measured through the score-
card, Sears is ready when registrars visit to audit. Sears
doesnt want to wait until the week before an audit and
scramble to make things look good. The scorecard pre-
pares Sears for the future, Tuggle said.
When Tuggle joined his region in 2005, his team had
six red areas on the scorecard. Now they have more
greens. Tuggle applauds the scorecard for keeping his
region in check and letting them make the QMS better.
Trach said her team didnt have any green areas
when she joined Sears in 2002. Now her team has
moved up to 85% green and 15% yellow.
The management team has spoken highly of the
scorecard because executives can quickly see how
the QMS is being maintained. Although detailed in-
formation is graded, the scorecard presents it in a
simple, obvious package.
While Sears also uses surveys, the Six Sigma Black
Belt methodology, internal audits and kaizen principles,
the scorecard is the one tool that ensures the company is
meeting the ISO 9001 requirements. The scorecard also
SEE A SEARS SCORECARD
Visit www.qualityprogress.com to see the scorecard that Sears Home Services
division uses to rate how its regions maintain and improve the QMS.
52 QP www.qualityprogress.com
reminds employees of Sears promise to provide quality
products and services and satisfy customers.
Friendly competition
Among all of the in-home, carry-in and delivery regions,
the one earning the highest grade wins a trophy, and the
top performing district in each region also receives one.
It has proven to be effective in driving competition
in the eld and created good-natured jesting back and
forth, Munn said.
Tuggle explained that ISO
9001 is a way of life. The score-
card helps put that notion into
place and keeps ISO 9001 new
and fresh. The scorecard has
gotten attention and has been a
great tool to allow Sears to mea-
sure and motivate continual im-
provement.
The trophy that goes along
with the scorecards has started to become coveted, so
theres some competition with it, and I think thats hu-
man nature and can only bring better performances,
Tuggle added.
Scorecard success
Since Sears implemented the scorecard two years ago,
all regions have maintained or improved their scores,
even in the face of an ever-changing market, consolida-
tions and associate turnover, Shale said.
The winning units are not always perfect, but they
have a majority of green scores compared to yellow or
red. The competition for these winning scores is erce
and has even led to scorecard discussions with manag-
ers ghting for every grade.
Between the rst quarter of 2006 and the fourth
quarter of 2007, many of the scorecard grades swung
dramatically. See Table 1 (p. 51) for some of the re-
sults.
The numbers really tell the story of how the QMS
scorecard has driven all of our regions to maintain and
improve our dedication and commitment to quality ser-
vice and business, Shale said.
Rebecca M. Neenan is a freelance writer in Nashville, TN.
 T
Special Advertising Section

O
ASQs
software directory P
& BUYERS GUIDE
T
S
earching for the best quality software for your needs
can be a daunting taskwhere do you even begin? Its

E
crucial to find the solution that best fits your business
needs but that also integrates seamlessly with your current
systems. These are, without a doubt, difficult choices. With
that in mind, ASQ presents its Annual Software Directory
and Buyers Guide, a one-of-a-kind resource to help make
your researchand your decisioneasier. With tons of
informative listings on the top vendors, you might find
yourself referring to it year-round.
N
Vendors Top 10 tips for making
the right software choice:
1. Ask other ASQ members for their feedback and referrals.

T
2. Get all stakeholders involved and make a comprehensive
checklist of your requirements. Differentiate what you must
have and what you would like in a perfect world.
3. List all the programs with which the software must interface
to ensure they work together.
4. Get a free trial of the software so you can try it firsthand.
5. Ensure that the software company has good training and
customer service.
I
6. Fully understand the cost and determine any hidden or extra

P
costs like consulting fees.
GET 7. Get a list of references.
8. Find out how long they have been in business and what new
products have been developed.

S
9. Try calling the support line with a question or two.
10. Keep and refer to ASQs Software Directory and Buyers Guide
and tell them you heard about them from ASQ.

June 2008 QP 53

InfinityQS
Real-Time SPC Software for Your Shop Floor, Enterprise, and Supply Chain
InfinityQS is the No. 1 ranked provider
of real-time Statistical Process Control
(SPC) software and technical services to
manufacturers worldwide. (ARC Advisory
Group) With a 17-year history of helping
businesses automate and implement SPC
solutions, InfinityQS has established a
strong presence throughout the U.S. and
internationally. To support its thousands
of customers, InfinityQS employs
industrial statisticians, ASQ Certified
Quality Engineers, ASQ Certified Quality
Technicians, and IT professionals with a
broad range of skills.
Thousands of manufacturers worldwide
rely on InfinityQS software to help reduce
scrap and improve their bottom line.
Customers include Rockwell Automation,
Caterpillar, Honeywell, Ethicon Endo-
Surgery, Ball Corp., Duracell, Quaker
InfinityQS International, Inc.
14900 Conference Center Drive
Chantilly, VA 20151
Data-Driven Decision Making
Software Tools for Six Sigma and Manufacturing Intelligence
COMPREHENSIVENESS
STATISTICA offers an extremely
comprehensive assortment of analytical
tools, graphics, and reporting capabilities.
USABILITY
The simple, intuitive, and highly
customizable interfaces found in STATISTICA
ensure that users will adopt it quickly.
INTEGRATION
STATISTICAs simple integration with
virtually all database platforms ensures that
data-driven decision making will harness all
of the data thats available.
OPEN ARCHITECTURE
Quality control activities require
collaboration among many individuals, and
STATISTICA offers an open architecture and
Contact StatSoft at 918-749-1119, or visit www.statsoft.com to learn more about our
high performance analytical solutions.
54 QP www.qualityprogress.com
Special Advertising Section
Oats, Tyson Foods, Raytheon, and Osram
Sylvania.
InfinityQS solutions are 100-percent
scalable and easily customizable to
enable manufacturers to reduce scrap and
rework, improve quality, and significantly
decrease costs. With automated gathering,
analysis, and reporting of data across the
enterprise, on the extranet and through
the Web, operations of any size can
benefit from InfinityQS.
Formerly known as SPC Quality Suite,
ProFicient is InfinityQS recent upgrade
to its industry-leading SPC software. It
automates data collection and analysis
on the plant floor, helping quality
professionals make informed decisions to
improve their manufacturing processes in
real-time, before defects occur. It offers
1-800-772-7978
sales@infinityqs.com
www.infinityqs.com
tools to ensure analysis results are easily
accessible and transferable.
FEATURE RICH APPLICATIONS
The STATISTICA Enterprise software system
is the data analysis platform used by Six
Sigma and manufacturing intelligence teams
around the world. STATISTICA provides a
comprehensive suite of data analysis, data
mining, and visualization tools all within a
single software platform, including a rich
palette of features:
Largest selection of analytical techniques
and graphics
Simple connections to data repositories
Analysis templates
Report templates
Interactive filtering
Security management
a clean and simple interface, automatic
recording and alerting of events, more
than 300 different types of control charts,
unlimited data sorting on any chart or
graph, and other unique features.
eSPC extends the capabilities of SPC
software to the entire enterprise and
supply chain. It is a Software as a Service
(SaaS) solution for real-time quality
management, requiring minimal IT
infrastructure and support. Manufacturers
can conduct plant-to-plant or supplier-
to-supplier comparisons to scientifically
improve quality.
HIGH PERFORMANCE
ANALYTICAL SOFTWARE
SOLUTIONS
The STATISTICA product line is utilized
across many industries by leading
global corporations in areas such
as manufacturing, semiconductors,
pharmaceutical, chemical, petrochemical,
food processing, automotive, heavy
equipment, insurance, telecom, R&D, and
others. STATISTICA is used in mission-
critical manufacturing applications and in
regulated FDA-controlled industries.
STATISTICA has achieved an unprecedented
record of recognition from both users and
expert reviewers. STATISTICA has received
the highest rating in EVERY comparative
review in which it has been featured since
its first release in 1993.
Throughout the Ages weve sought wisdom
in the Mountains.You still can
Come to Knoxville, Tennessee
where useful techniques, and the way of thinking that makes them work, are taught.
Learn from Donald J. Wheeler.
You and your company will be glad you did!

Understanding Statistical Process Control

SPC for Manufacturing & Process Industries
September 8 - 11, 2008

Practical Data Analysis

Doing Six Sigma Better
October 13 - 16, 2008

Successful Experimentation
Making Industrial Experiments Work
October 27 - 30, 2008

Understanding Variation
In-House Only

Evaluating the Measurement Process

In-House Only

1.0

EMP III
0.9

0.8

0.7

0.6
EVALUATING THE
0.5
MEASUREMENT PROCESS
&
Probability of Detecting Three Standard Error Shift

0.4

0.3
USING IMPERFECT DATA
0.2
I II III IV

0.1
DONALD J. WHEELER
I
0.0
1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0
Intraclass Correlation Coefficient

Statistical Process Control, Inc. & SPC Press www.spcpress.com

5908 Toole Drive, Suite C Knoxville, Tennessee 37919 USA 1-800-545-8602

Enterprise Quality Management
and Regulatory Compliance Solution
AssurXs CATSWeb is a flexible,
all-in-one platform that automates
quality management and regulatory
compliance-related processes so issues
can be globally managedincluding
detection, corrective action, and trend
analysis. It helps collect, organize,
analyze, and share information to
better manage and improve quality and
compliance performance everywhere in
your enterprise.
By providing clear views and shared
information across functional areas,
AssurX creates the power to easily detect,
correct, and analyze issues globally so
companies can continuously improve
operations, compliance, and their
bottom line.
AssurX, Inc.
18525 Sutter Blvd., Suite 150
Morgan Hill, CA 95037
888-9-ASSURX
SigmaXL + Excel = Easy Statistical
and Graphical Analysis!
SigmaXL is a leading provider of user-friendly
Excel add-in tools for statistical and graphical
analysis.
Our flagship product, SigmaXL, was designed
from the ground up to be cost-effective,
powerful, and easy to use. It enables users to
measure, analyze, improve, and control their
service, transactional, and manufacturing
processes. As an add-in to the already familiar
Microsoft Excel, SigmaXL is ideal for Six Sigma
training and is used by leading Six Sigma
consultants. It is rapidly becoming the tool of
choice for Six Sigma professionals.
Unique Features
SigmaXL includes lean and DMAIC templates
such as Team/Project Charter, SIPOC Diagram,
Data Measurement Plan, Cause and Effect
Matrix, FMEA, Quality Function Deployment,
Pugh Concept Selection Matrix, Control Plan,
Takt-Time Calculator, and Process Load Balance.
SigmaXL
1-888-SigmaXL (1-888-744-6295)
or 1-416-236-5877.
56 QP www.qualityprogress.com
Special Advertising Section
With CATSWeb, you can manage:
CAPA
Product Defects
Nonconformances
Supplier Quality
Change Control
Training
Complaints
Returns
Audits
Management Dashboards, KPIs,
and Metrics
The CATSWeb solution comes with a
built-in charting and report feature that
allows you to view, modify, and add
Web: www.assurx.com
E-mail: sales@assurx.com
Measurement Systems Analysis includes
a random worksheet generator with user
specified number of parts, operators, and
replicates. The analysis adds confidence
intervals to the MSA metrics.
The two-sample comparison test automatically
tests for normality, equal variance, means, and
medians, and provides a single report with
rules-based yellow highlight to aid the user in
interpretation of the output. Low p-values are
highlighted in red indicating that the results
are significant.
Design of Experiments provide a unique view
power analysis as you design feature allowing
you to readily see the trade-off between the
cost of experimental runs/replicates and the
corresponding statistical powerbefore you
finalize the design.
Multiple Linear Regression accepts both
continuous and categorical predictors. An
Web: www.sigmaxl.com
E-mail: sales@sigmaxl.com
reports without having to install and
learn third party reporting tools, enabling
you to monitor KPIs and drill down to
the underlying datain real time, all
the time.
Reduce problems by tracking trends
over time by site, department, product,
activity, or any other parameter. Manage
resources by identifying and working on
the largest issues first. Prevent recurring
problems by determining root causes and
implementing lasting effective solutions.
Professional Services and Support
AssurX maintains a staff of highly
experienced business process engineers
who are available to assist you in
identifying operational areas and
processes that can be improved, resulting
in higher product quality, improved
operational efficiencies, and cost savings.
interactive Predicted Response Calculator gives
95 percent confidence intervals and 95 percent
prediction intervals.
The Control Chart Selection Tool simplifies the
selection of SPC charts. The control chart tools
allow you to add data to an existing chart,
scroll through the chart data, split limits by
historical groups, and easily read a summary of
test violations.
Cost Effective
Priced at $199, SigmaXL is a fraction of the
cost of any major statistical product, yet it has
all the statistical and graphical functionality
most professionals need.
A free 30-day trial version is available for
download from our Web site. Quantity,
educational, and trainer discounts are
available.

Creating a Culture of Quality
For years, the manufacturing industry
has been in the midst of a struggle
between two opposing forcessupply
chain management and supply quality
management. So often organizations find
that the need to ensure that product
reaches store shelves quickly is at odds
with the painstaking attention to quality
in the manufacturing process. The quality
of manufacturing must be linked directly
to the efficiency in the supply chainso
that product arrives to the shelves
quickly and with absolutely no defections
or, worse, harmful components.
The biggest challenge facing the
manufacturing of goods is a genuine
disconnect between C-level management
and those responsible for ensuring
strict quality procedures. The only
Sparta Systems, Inc.
2137 Highway 35
Holmdel, NJ 07733
732-203-0400
Are You Having Trouble Scoring?
WE PROVIDE A PERSONAL APPROACH &
REAL-WORLD, PRACTICAL EXPERIENCE.
CONSULTING & TRAINING IN:
ISO 9001, TS 16949, ISO 14001,
ISO 27001, ITIL, Lean, Six Sigma,
AS9100, ISO 17025, ISO 18001
CALL TO FIND OUT HOW
OUR TEAM CAN HELP YOURS.
Special Advertising Section
way to confront these challenges is customer that counts. A culture of quality
for organizations to create a culture is not simply a way to reduce compliance
of quality that begins at the top and costsits essential for the health of
permeates through the supply chain. the industry.
Continuous improvement needs to start
at the top. Only when this philosophy
has taken hold will we see a real
difference in the industry, both for
the organizations looking to improve
their bottom line and for consumers
seeking quality.
In the manufacturing industry, ensuring
that the raw materials and finished
products meet the requirements and
specifications of government regulatory
bodies can be enormously difficult. But
the real reason why quality management
is so important has nothing to do with
battling government agenciesits the
info@sparta-systems.com
www.sparta-systems.com
GS-23F-8060H
Visit us on-line for Web Specials
www.EuroQuest.net
June 2008 QP 57
 Special Advertising Section 2 0 0 8
Software

ILITY
DUCIB
ENT

ENT
Showcase

G
APPIN
LOPM

EPRO

O YM

RO L

CE
N/M
DEVE

DEPL
ND R

RAN
& Directory

ONT
T

TATIO
EM EN

ITY A

ASSU
ARE

AND
T

SS C
N
N

DS

S
ACTIO
RIME

IQUE
OFTW
TION
S

ETHO
949
NAG

M EN
E

TION
TABIL

N
RO CE
ING
N
UDIE

RAN

LITY
RVIC
RD

IO
ISITIO
G

EXPE

ECHN
6

T
E AC

SOLV
AWA

T MA

DO CU
T
RKIN

ED S

TIVE

AL M
TS 1

FUNC
ASSU

QUA

ARA
Y ST

ENT

AL P
M EN
EPEA
R SE
N

RCE
ING

N
RS

N
RATIO

ACQU

Y
ECTIV

LATIO

E
AGEM

IGMA
N OF

4000

CHI T
OMIZ

ENTA

BILIT
ECTIO

000/

PREP
M EN
BILIT
HMA

ING
WAR
ITING

RIGE

ISTIC

ISTIC
PUTE

MME

BLEM
SURE
OME

LIER
000
GE R
SULT

CESS

LITY

LITY

R
TRAIN
DESIG

DOCU
CORR
CALIB

RELIA

TAGU
ISO 9

ISO 1
DATA
CAPA

PREV

SIMU

SUPP
SIX S

OTHE
BALD

BENC

CUST

CUST

QS-9

SOFT
MAN

STAT

STAT
COM

INSP
E-CO

TEST
QUA

QUA
MEA
GAU
CON
AUD

PRO

PRO
Company
AssurX x x x x x x x x x x x x x
18525 Sutter Blvd., Suite 150
Morgan Hill, CA 95037
P:408-778-1376
www.assurx.com
AssurX provides an all-in-one platform that
automates quality management and regulatory
compliance related processes so issues are globally
managedfrom detection to corrective action.
CEBOS Ltd. x x x x x x x x x x x x x x x x x x
5936 Ford Road, Suite 203
Brighton, MI 48114
P:810-534-2222
www.cebos.com
CEBOS develops MQ1, a suite of enterprisewide
quality management, compliance, and business
process management software. MQ1 was created
to streamline workflow, increase productivity, and
consolidate multiple applications into one easy-to-
use, functionality rich system.
CyberMetrics Corporation x x x x x x x
P:800-777-7020
www.cybermetrics.com
DataNet Quality Systems x x x x
24567 Northwestern Highway
Southfield, MI 48075
P:248-357-2200
www.winspc.com
DataNet Quality Systems delivers real-time SPC
software and services that empower manufacturers
to improve their products, processes, and
profitability.
EtQ, Inc. x x x x x x x x x x x x x
399 Conklin St., Suite 208
Farmingdale, NY 11735
P:516-293-0949
www.etq.com
EtQs Web-based enterprise quality management
system automates quality and compliance business
processes for document control, risk assessment,
corrective action, nonconforming materials, audits,
reporting, and more!
GrayTech Systems, Inc. x x x x x x x x x x x x x
P:770-499-1510
www.graytechsystems.com
Harrington Group x x x x x x x x x x x x x x x x x x x x x x x x
11501 Lake Underhill Road
Orlando, FL 32825
P:407-382-7005
www.harrington-group.com
Harrington Group is an advanced technology
company that produces enterprisewide and desktop
software solutions to help organizations meet their
quality management and regulatory compliance
requirements.
IBS America, Inc. x x x x x x x x x x x x x x x x x x x x x x x x x
P:781-862-9002
www.ibs-us.com

58 QP www.qualityprogress.com
 Special Advertising Section 2 0 0 8
Software

ILITY
DUCIB
ENT

ENT
Showcase

G
APPIN
LOPM

EPRO

O YM

RO L

CE
N/M
DEVE

DEPL
ND R

RAN
& Directory

ONT
T

TATIO
EM EN

ITY A

ASSU
ARE

AND
T

SS C
N
N

DS

S
ACTIO
RIME

IQUE
OFTW
TION
S

ETHO
949
NAG

M EN
E

TION
TABIL

N
RO CE
ING
N
UDIE

RAN

LITY
RVIC
RD

IO
ISITIO
G

EXPE

ECHN
6

T
E AC

SOLV
AWA

T MA

DO CU
T
RKIN

ED S

TIVE

AL M
TS 1

FUNC
ASSU

QUA

ARA
Y ST

ENT

AL P
M EN
EPEA
R SE
N

RCE
ING

N
RS

N
RATIO

ACQU

Y
ECTIV

LATIO

E
AGEM

IGMA
N OF

4000

CHI T
OMIZ

ENTA

BILIT
ECTIO

000/

PREP
M EN
BILIT
HMA

ING
WAR
ITING

RIGE

ISTIC

ISTIC
PUTE

MME

BLEM
SURE
OME

LIER
000
GE R
SULT

CESS

LITY

LITY

R
TRAIN
DESIG

DOCU
CORR
CALIB

RELIA

TAGU
ISO 9

ISO 1
DATA
CAPA

PREV

SIMU

SUPP
SIX S

OTHE
BALD

BENC

CUST

CUST

QS-9

SOFT
MAN

STAT

STAT
COM

INSP
E-CO

TEST
QUA

QUA
MEA
GAU
CON
AUD

PRO

PRO
Company
iGrafx x x x x x x x x x x x x x x x x x x x x
7585 SW Mohawk St.
Tualatin, OR 97062
P:503-404-6050
www.igrafx.com
iGrafx solutions help organizations reduce
costs, waste, and risk by allowing IT, business
analysts, and process initiative practitioners to
understand and improve processes across functional
boundaries.
InfinityQS International x x x x x x x x x x x x x x x x x x x x x x x
14900 Conference Center Drive, Suite 525
Chantilly, VA 20151
P:703-961-0200
www.infinityqs.com
InfinityQS is the No.1 independently ranked
real-time SPC software provider to manufacturers
worldwide. InfinityQS solutions dramatically reduce
costs and improve quality by identifying potential
issues before defects are produced.
Intelex Technologies, Inc. x x x x x x x x x
366 Adelaide St. W, Suite 100
Toronto, ON M5V1R9
P:416-599-6009
www.intelex.com
Intelex is a leading provider of quality
management systems. Its software is designed to
drive performance by enabling organizations to
effectively track, analyze, and report on product
defects, supplier management, corporate objectives,
and more.
Lifestar x x x x x x x x x x x x x x x x
2253 S. Oneida St., Suite 3D
Denver, CO 80224
P:888-468-1537
www.qimacros.com
QI Macros SPC Software for Excel includes control
charts, histograms, QFD, DOE, PPAP, and FMEA.
$139/copy. Download a 30-day evaluation copy at
www.qimacros.com/freestuff.html.
Minitab Inc. x x x x x x x x x x x x x x x x x
Quality Plaza, 1829 Pine Hall Road
State College, PA 16801
P:800-448-3555
www.minitab.com
Minitab Statistical Software, Quality Companion by
Minitab, and Quality Trainer by Minitab provide
a complete solution for Six Sigma and quality
improvement projects. Free trial versions are
available at www.minitab.com/downloads.
Pilgrim Software x x x x x x x x x
P:813-915-1663
www.pilgrimsoftware.com
PQ Systems, Inc. x x x x x x x x x x
P:800-777-3020
www.pqsystems.com
Quality America, Inc. x x x x x x x x x x x x x x
1984 N. Kolb Road
Tucson, AZ 85715
P:800-722-6154
www.qualityamerica.com
Founded in 1982, Quality America offers statistical
process control and document control software; ASQ
certification materials; SPC training; and Six Sigma
software, training, and consulting.

June 2008 QP 59
 Special Advertising Section 2 0 0 8
Software

ILITY
DUCIB
ENT

ENT
Showcase

G
APPIN
LOPM

EPRO

O YM

RO L

CE
N/M
DEVE

DEPL
ND R

RAN
& Directory

ONT
T

TATIO
EM EN

ITY A

ASSU
ARE

AND
T

SS C
N
N

DS

S
ACTIO
RIME

IQUE
OFTW
TION
S

ETHO
949
NAG

M EN
E

TION
TABIL

N
RO CE
ING
N
UDIE

RAN

LITY
RVIC
RD

IO
ISITIO
G

EXPE

ECHN
6

T
E AC

SOLV
AWA

T MA

DO CU
T
RKIN

ED S

TIVE

AL M
TS 1

FUNC
ASSU

QUA

ARA
Y ST

ENT

AL P
M EN
EPEA
R SE
N

RCE
ING

N
RS

N
RATIO

ACQU

Y
ECTIV

LATIO

E
AGEM

IGMA
N OF

4000

CHI T
OMIZ

ENTA

BILIT
ECTIO

000/

PREP
M EN
BILIT
HMA

ING
WAR
ITING

RIGE

ISTIC

ISTIC
PUTE

MME

BLEM
SURE
OME

LIER
000
GE R
SULT

CESS

LITY

LITY

R
TRAIN
DESIG

DOCU
CORR
CALIB

RELIA

TAGU
ISO 9

ISO 1
DATA
CAPA

PREV

SIMU

SUPP
SIX S

OTHE
BALD

BENC

CUST

CUST

QS-9

SOFT
MAN

STAT

STAT
COM

INSP
E-CO

TEST
QUA

QUA
MEA
GAU
CON
AUD

PRO

PRO
Company
SigmaXL x x x x x x x x x x x
151 Bloor St. W, Suite 890
Toronto, ON M5S 1S4
P:888-SigmaXL
www.sigmaxl.com
Sigma XL + Excel = Easy Statistical and Graphical
Analysis. Sigma XL is cost-effective, powerful, user-
friendly, and ideal for Six Sigma training! SigmaXL
offers a free 30-day trial.

Sparta Systems x x x x x x x x x x
P:732-203-0400
www.sparta-systems.com
Statistical Process Controls, SPC Press x x x x x x x
5908 Toole Drive, Suite C
Knoxville, TN 37919
P:865-584-5005
ww.spcpress.com
Statistical Process Controls, Inc. was started by
Donald J. Wheeler in 1982 and provides world-class
training in data analysis. SPC Press, a part of that
company, provides books and training materials.
StatPoint Technologies Inc. x x x x x x x x x x x x x
2325 Dulles Corner Blvd., Suite 500
Herndon VA 20171-4676
P:540-364-0420 x350
www.statgraphics.com
Unmatched statistical power meets ease of use.

StatSoft Inc. x x x x x x x x x x x x x x x x
P: 918-749-1119
www.statsoft.com
Title 21 Software x x x x x x x x x x x x x x x x x x x x x x x x
101 Convention Center, Suite 700
Las Vegas, NV 89109
P:925-989-6161
www.title21software.com
Title21 Softwares document control, quality data
management, and training (21 CFR Part 11) are
extremely user-friendly, integrated software.
Highly-configurable and easily adoptable, you will
realize the fastest ROI with Title21.
TriloByte Statistical Software x x x x x x x x
Hradistska 300
Pardubice, Czech Republic, Europep 53352
P:420 46 6615 725 STATISTICAL
SOFTWARE

www.trilobyte.cz
Founded in 1990, TriloByte is a European research
and development company focused on statisitical
analysis. Its main product, QCEXPERT, is the
statistical software package for advanced statistical
analysis in industrial QC and research.

StatSoft Web site

SOFTWARE REVIEW

The StatSoft Web site (www.statsoft.com) provides not only in-depth information about the
broad line of STATISTICA solutions, but also an award-winning Electronic Statistics Textbook.
Information can be found to assist users in understanding a broad selection of cutting-edge
methods to meet their data analysis needs. They can also visit any of the 24 StatSoft offices
around the world.
StatSoft, Inc.
Phone: 918-749-1119

60 QP www.qualityprogress.com
3.4 PER MILLION
Data Dependability
Improve data accuracy for managing, improving processes
AS QUALITY or Six Sigma professionals,
we have been taught to address the issue
of data accuracy and integrity from the
statistical viewpoint.
Both the Certied Six Sigma Black Belt
(CSSBB) exam and the Certied Quality
Engineer (CQE) exam test these concepts,
although they are described somewhat
differently in their respective bodies of
knowledge (BoKs).
The data accuracy and integrity de-
scription in the CQE BoK (section VI.A.4)
states:
Describe the characteristics or prop-
erties of data (for example, source/re-
source issues, exibility and versatility)
and various types of data errors or poor
quality such as low accuracy, inconsis-
tency, interpretation of data values,
and redundancy. Identify factors that
can inuence data accuracy, and apply
techniques for error detection and cor-
rection. (Apply)
This column will address most of the el-
ements of this subtopic by discussing why
data accuracy is important, the numerous
causes of poor data accuracy, various tech-
niques for improvement and the placement
of effective data collection points.
Importance of data accuracy
Poor data accuracy undermines two key
precepts in a quality-focused or Six Sigma
organization: fact based and data-driven
decision making.
When faced with data that has been
rendered useless because of data accuracy
issues, decision makers must revert to
relying on intuition.
LOOKING FOR MORE 3.4?
If you liked this article, dig back to past columns at www.qualityprogress.com.
BY T. M. KUBIAK
This might provide the naysayer with
reason and justication to ask why this Six
Sigma approach wont work here. Remem-
ber, Six Sigma drives cultural changes in or-
ganizations, and accurate data are necessary
to drive and sustain real improvement.
In addition, some organizations might
be required to submit highly accurate data
because of regulatory or reporting require-
When faced with data that has
been rendered useless, decision
makers must revert to relying
on intuition.
ments, or because of contractual require-
ments. This is particularly true in indus-
tries such as aerospace in which data, in
addition to product, are a deliverable.
In fact, most organizations have
processes that deliver only data to other
organizations.
Consider the example of payroll data
regularly reported to the Internal Revenue
Service. The consequences associated
with poor data accuracy can be quite
troublesome and have a highly negative
impact on the individual depending on the
data.
Minimizing poor data accuracy
There are many causes of poor data ac-
curacy. Table 1 (p. 62) includes a few of
the most signicant causes. Consider how
these issues relate to your own experience
and create your own list so you can mini-
mize the impact of poor data accuracy in
your organization.
Table 2 (p. 63) lists techniques used
to minimize the occurrence of poor data
accuracy or its impact.
Remember, processes are measurable
and generate data. Treat poor data ac-
curacy as you would any other quality de-
fect: Use the plethora of tools available to
you as a quality or Six Sigma professional
to ferret out root cause. Chart data defects
as you would product defects. Make them
available for all to see.
Common data collection points
Five common data collection points are
summarized in Table 3 (p. 64) and are as
follows:
End of process: Traditionally, orga-
nizations have collected data to measure
process performance at the very end when
everything is said and done and its usually
too late to take corrective action.
In process: Over time, more progres-
sive organizations have realized that data
should be collected upstream and during
the process. Data collection at critical
subprocesses allows corrective action to
be taken earlier when change is still an
June 2008 QP 61
 3.4 PER MILLION

Common causes of poor data accuracy / TABLE 1

Issue Comment
Multiple points of entry When data entry is centralized to a single individual, entry errors are usually minimized over time because of
the individuals experience and the ability to clarify questions with the data originator. When data originators
become responsible for their own data entry, data accuracy usually diminishes, particularly when frequency
of data entry decreases or there is high turnover among the data originators.
Limited use of data Such validation would include the presence or absence of data, data ranges, inconsistent data and other
validation logical checks. Surprisingly, computer technology has not completely eliminated this cause.
Batching input versus This scenario usually results in instantaneous differences, but the differences are usually reconciled with
real-time input/ time. Consider the example of an individual inquiring about a checking account balance through an ATM, a
synchronization of bank teller or online. Because of different cutoff and system update times, the individual will see up to three
system updates different account balances.
Multiple means of data When multiple data correction access points are present, the opportunity to have synchronization issues
correction access to increases. For example, suppose there are three upstream systems prior to the current system. Label them
upstream systems 4, 3, 2 and 1 for the current system. If corrections are made to 4, then 3, 2 and 1 are updated. However, if
corrections are made to 2, only 2 and 1 are updated.
Unclear directions This issue becomes increasingly important when multiple points of data entry exist.
Lack of training This issue becomes increasingly important when multiple points of data entry exist.
Ambiguous terminology Ask an operations manager for monthly data and you get monthly data according to the calendar month.
Ask for the same data from an accountant and you get monthly data by scal month. Different months
lengths yield different data.
Manual versus automated Automated data entry is usually preferred over manual data entry, particularly if you accept the premise that
means of entry people are fallible.
Rounding Improper rounding. For example, when rounding up is appropriate instead of rounding down. Also, the
cumulative effects of rounding can generate errors. For example, computers store and calculate using
numbers to 16, 32 or even 64 digits. An individual using numbers rounded to four digits in the same
calculations can arrive at a result that is substantially different than the results arrived at by a computer.
Order of calculations 36 9
Consider (1/9) . If you rewrite this equation as 36 1 and multiply through, you obtain 324. If you divide out
1/9 and round to four decimals you obtain 0.1111. Dividing 36/0.1111 you obtain 324.0324. Different order,
different answers.
Calculation errors This is simply an arithmetic error or an error in the specication of the formula to be used. For example, I
worked at two organizations that computed associate turnover incorrectly.
Inadequate measurement This is an issue commonly known among Six Sigma professionals who have been taught to examine the
systems adequacy of the measurement system rst. Examples include Kappa studies and gage repeatability and
reproducibility studies.
Units of measure not This type of error occurs when the wrong use of measure is assumed or used, such as foot instead of inch,
dened yard instead foot, or box instead of each. The last example can signicantly impact inventory levels.
Failing to use the proper On Sept. 23, 1999, after a 286-day journey from Earth to Mars, NASA lost the $125 million Mars Climate Orbiter
system of measurement after the spacecraft entered orbit 100 kilometers closer than planned. NASA used the metric system, while its
partner, Lockheed Martin, the organization that designed the navigation system, used English units.
Field inconsistencies/ Moving an n character eld to an n-1 character eld or smaller can result in eld truncation, which would
truncation among impact all calculations that use it. For example, 100,789 is moved into a three-character numeric eld. The
systems or data elds result would be 789, not 100,789.
Similarities of characters Several letters and numbers can easily be confused, particularly when handwriting is poor or when reading
other than the top sheet of a no carbon required form. Common examples include:
0 O (oh) 4 H 8 B
1 l (el) 5 S 9 g (gee) or q
2 Z and occasionally Q 6 G
3 B 7 Z
Examples of possible confusion between letters:
c-e; C-O; E-F; m-n; M-N; O-Q; P-R; r-v; u-v; U-V; v-w; V-W; v-y
Copying errors Duplicates of duplicates can result in image degradation and impact the readability of the copy. Also, copy
truncation occurs when the image being copied is close to the edge of the paper. Try this with your credit card
statement sometime. Ensure the purchase amounts are adjacent to the edge of the copier glass.

62 QP www.qualityprogress.com
option. This move facilitates more effec-
tive process management and allows the
Useful data collection techniques / TABLE 2

organizations to better gauge the overall Technique Comment

health of their processes.
Points of convergence: This occurs
when multiple streams of processes
come together. Collecting data at this
point makes sense because it helps or-
ganizations understand how each input
stream affects the downstream subpro-
cesses.
Points of divergence: This is nothing
more than a ip of the points of conver-
gence. Keep in mind that a decision point
is a specialized point of divergence with
two distinct outputs.
Decision points involving yes or
no decisions are useful data collection
points, particularly when the yes path is
the desired path and the no path is the
undesired path.
Understanding the percentage of
products or services taking the no path
becomes an important piece of data nec-
essary for driving process improvement.
Across functional boundaries: These
are nothing more than process hand-off
points. However, responsibilities and
accountability usually change across func-
tional boundaries.
As such, collecting data at these points
helps organizations understand how
each functional department supports or
detracts from process efciency and effec-
tiveness. It also helps minimize organiza-
tional nger pointing with data and fact.
Accuracy is essential
This column covers the numerous causes
of poor data accuracy, a variety of
techniques available to help minimize or
Walk the process rst Always walk the process rst so everything that follows in
this table has meaning and applicability.
Make the data collection Minimize the need to perform calculations and maximize the
process simple use of direct data collection. Direct data collection occurs
when you take a reading from an instrument and log or enter
that piece of data. In this case, a simple data transference has
occurred.
Dene collection points Table 3 identies typical and useful data collection points.
Use checksheets and These are simple but very effective tools and are often
checklists overlooked. Perhaps they are too simple.
Bridge computer gaps Some organizations still have processes requiring individuals
to extract data from one computer report and enter it into a
database residing on a different computer. In some cases, it is
the same computer.
Minimize other What is other? When collecting nominal or categorical
data, it is often the last, and sometimes the largest, bar on
a Pareto chart. A substantial amount of data classied as
other usually indicates that data originators either dont
understand clearly how to classify the data, or categories are
not mutually exclusive. The opportunity exists to classify data
into multiple categories. As a result, data originators might
disagree with one another and the measurement system
becomes inadequate.
Limit options The more categories of data there are, the greater the
possibility of a data collection error. This technique should be
balanced against the previous minimize other technique.
Establish the rules This technique is best illustrated by an example. Consider an
organization that receives payments stating that any receipt
received after 3 p.m. will be credited on the next business day.
Address timing and This technique is important when a data collection point is
sequencing inserted to collect data from multiple input streams. In this
case, be particularly aware of the possibility of generating
a bottleneck by waiting for a specic data input stream.
Otherwise, it might be necessary to collect the data at
another point downstream. If this is not possible, rebalancing
the upstream processes might be required.
Time stamp data This technique facilitates traceability of activities and events
through time based data and allows for more thorough root
cause analysis.
Red tag a unit of Track a specied unit of product or service through the entire
product or service process to understand how it is used to generate data or how
data are collected from it.
Chart data accuracy Remember data collection is part of the process and
therefore subject to charting techniques, such as control
charts and run charts.

eliminate these causes and the ve com-

mon data collection points in processes in
which the capture of data can have a sig-
nicant impact on understanding process
performance.
The delivery of data might be a regula-
tory or contractual requirement. Conse-
quently, poor data accuracy can adversely
impact an organization. Data defects like
product defects can and should be charted
and improved using the variety of tools
at the disposal of the quality or Six Sigma
professional.
While the current CSSBB BoK does not
identify data accuracy in the same way as
the CQE BoK, data accuracy is essential
for managing and improving processes

June 2008 QP 63
 3.4 PER MILLION

Common data collection points (DCPs) / TABLE 3 with sustained results. We look forward
to seeing the description of data accuracy
Point Example updated in the next revision of the CSSBB
End of process BoK to reect the key elements of this
A B C DCP
important topic.
In process
A DCP B DCP C BIBLIOGRAPHY
Benbow, Donald W., and T.M. Kubiak, The Certied Six Sigma
Black Belt Handbook, ASQ Quality Press, 2005.
Points of convergence Certied Six Sigma Black Belt Body of Knowledge, ASQ, 2001.
A Certied Six Sigma Black Belt Body of Knowledge, ASQ, 2007.
Certied Quality Engineering Body of Knowledge, ASQ, 2001.
Certied Quality Engineering Body of Knowledge, ASQ, 2006.
DCP C CNN.com, Metric Mishap Caused Loss of NASA Orbiter,
Sept. 30, 1999. www.cnn.com/tech/space/9909/30/mars.
metric.02.
B

Points of divergence
B

A DCP
T.M. KUBIAK is a consultant based
in Charlotte, NC. He is a co-author
C
of The Certied Six Sigma Black Belt
Handbook. Kubiak, a senior mem-
Across functional ber of ASQ, serves on many ASQ
boards and is a past chair of ASQs
boundaries Dept. A Dept. B Publication Management Board.

ASQ CERTIFICATION + Upper Iowa University MBA =

Access to the World of Opportunity

Combine your ASQ Certification with an accredited MBA
in Quality Management from Upper Iowa University and
maximize your opportunities for career advancement.
Whether you have a technical degree or one in business, the
Upper Iowa University MBA in Quality Management is designed
to integrate contemporary quality management concepts with
the knowledge needed to manage people and resources.
Graduates of this program are expected to be able to assume
leadership roles advancing quality in their organizations.
Online courses that fit your schedule.
No residency requirement
Start in any of 6 terms each year
Highly qualified and supportive
faculty

Developing Leaders Since 1857

Upper Iowa University is accredited by the Higher Learning Commission and is a member of the North Central Association.
The MBA Program and business degrees offered are further accredited by the International Assembly for Collegiate Business Education.

UPPER IOWA UNIVERSITY Contact American Society for Quality:

1.800.603.3756 online@uiu.edu www.uiuonline.info www.asq.org or telephone: 1.800.248.1946
On Campus Online Correspondence U.S. & International Centers for certification information.

64 QP www.qualityprogress.com
Professional Networking
Through Each One Reach One
Jim Smith
Joined ASQ in 1979
Fellow Member
Active participant in the Each One Reach One referral program
Dianne Smith
Over the years, I have recruited more than 400 people, from several areas Joined ASQ in 2006
around the United States, and continue to stay in touch with many of them. Recruited by Jim Smith
At the end of the day, quality needs to be personal.

A very important aspect of growing as a professional is the chance to build relationships with your
peers. ASQ provides its members with a great opportunity with the Each One Reach One member
referral program. With this program, you have the chance to create your own network of professionals Just another
with other members in your local Sections, Forums and Divisions, and communities of practice, while
gaining rewards for recruiting new members. With more than 6,000 members referred through Each reason ASQ
One Reach One every year, you have the opportunity to make a difference in someones career. is your
For every Individual member you recruit you will earn 5 ASQ Bucks, and for every Site (formerly professional
Sustaining) member you earn 120 ASQ Bucks (1 ASQ Buck = $1). You can use your ASQ Bucks association
toward:
of choice.
Membership renewal* Conference of your choice
Quality Press purchases Training and certification

Remember your name and member number must be entered in the Member Referred By line on the
application of the new member you recruited in order to qualify for rewards. Get started today by using
the application on the back of this ad.
*NEWuse any amount toward your membership renewal.

LEARN MORE AT WWW.ASQ.ORG/EORO.

 OFFICE USE ONLY
MBHKE78
Membership Application
PRIORITY CODE ________________________

Order Number _________________________

Member Number _______________________

Preferred Mailing Address: Home Business Industry: Healthcare Service

Government Education Customize Your Membership Experience
Mr. Ms. Mrs. Dr. Male Female Manufacturing Member Type:
1 Full (formerly Regular) $125 Associate $72
Date of Birth: _________ /_________ / _________ Forum- or Division-only $31 $
Month Day Year The one geographical Section included with Full (formerly Regular) membership
2 will be determined by your primary address.*
Sections (geographic or the e-Section) may be added to any member type for an
First Name Middle Initial Last Name additional $20.00 each. Visit www.asq.org for a listing of available Sections.

Sections , , $
Company Name Job Title * Contact ASQ to change your assigned Section.

Forum or Division Selection

3 As part of your Full (formerly Regular) membership you receive membership
Business Address Ste.
in one topic- or industry-specific Forum or Division. For Forum- or Division-only
membership and Full (formerly Regular) membership use the list below to indicate
the Forum or Division number and name.
City, State/Province Zip+4/Postal Code Country
included
$
(#) Name

Home Address Apt./Ste. Additional Forums and Divisions may be added to all levels of membership.
Please indicate in the list below the additional Forums or Divisions you would
like and total the number you have selected.
City, State/Province Zip+4/Postal Code Country Audit (19) Human Development and
Automotive (3) Leadership (13)
Aviation, Space and Defense (2) Inspection (9)
Area Code/Business Telephone Area Code/Home Telephone Biomedical (10) Lean Enterprise (23)
Chemical and Process Industries (4) Measurement Quality (17)
Customer-Supplier (15) Product Safety and Liability
Preferred E-mail Address Fax
Design and Construction (20) Prevention (25)
Education (21) Quality Management (1)
Which one of the following best describes your title? Electronics and Communications (5) Reliability (8)
Energy and Environmental (11) Service Quality (16)
Accountant Clinician Facilitator Manager Retired Student
Food, Drug, and Cosmetic (7) Six Sigma (26)
Advisor Coordinator Foreman Mechanic Scientist Superintendent
Government (22) Software (14)
Administrator Consultant General Nurse Six Sigma Teacher Healthcare (18) Statistics (12)
Analyst Contractor Manager Owner Black Belt Technician Team & Workplace Excellence (27)
Associate Controller/ Inspector President Six Sigma Unemployed Additional Forum and Division selections:
Auditor Comptroller Instructor Principal Green Belt Full (formerly Regular) or Associate member _____ x $10 = $
total

CEO Director Librarian Professor Specialist Forum- or Division-only member _____ x $31 = $
total
Chemist Engineer Machinist Programmer Staff Publications
Statistician
4
Quality Progress Domestic Canadian International Price/cost
Full (formerly Regular) Member included included** included**
ASQ does not sell e-mail addresses to third parties. Associate Member $55.00 $90.00 $90.00 $
Forum- or Division only Member $75.00 $110.00 $110.00 $
Mailing Lists Electronic Only $55.00 $55.00 $55.00 $
Occasionally ASQ shares its mailing list with carefully selected quality-related Air Service Fee n/a $30.00 $35.00 $
organizations to provide you with information on products and services. Please
check this circle if you do not wish to receive these mailings. Six Sigma Forum Magazine $45.00 $65.00# $70.00 $
Journal of Quality Technology $30.00 $51.00# $49.00 $
Software Quality Professional $45.00 $65.00# $70.00 $
Member Referred By:
Quality Management Journal $50.00 $80.00# $74.00 $
Quality Engineering $34.75 $51.25# $51.25 $

Member Name Member Number Technometrics $30.00 $30.00# $30.00 $

The Journal for Quality and
#
Participation $45.00 $65.00 $70.00 $

Payment Information **If you do not choose the airmail option, delivery can take approximately 8-12 weeks.
#
GST included $
Total of all magazines and journals
Check or money order (U.S. dollars drawn on a U.S. bank) Make check payable to ASQ.
MasterCard Visa American Express (Check one) TOTAL OF ALL ITEMS (1-4): $
Please submit your application with remittance to:
Cardholders Name (please print) ASQ You may also join online at
P.O. Box 3066 www.asq.org or by calling ASQ
Milwaukee, WI 53201-3066 Customer Care at 800-248-1946
Card Number Exp. Date or 414-272-8575.
or fax to 414-272-1734.
New memberships are effective upon receipt of payment. New members receive one year of
Cardholders Signature
membership from the date they join. Members are billed prior to the anniversary date of their
membership for next years dues. Memberships, even those paid by employers, are nontransferable.
All prices are subject to change.
Cardholders Address In becoming an ASQ member, you have the duty to follow the ASQ Code of Ethics and Society
governing documents.
STANDARDS OUTLOOK
Off to a Good Start
Initial contact can be formal or informal
but must include seven action items
CLAUSE 6.2.6 of the U.S. supplement
for the International Organization for
Standardizations (ISO) auditing standard
provides guidelines for establishing initial
contact with the auditee. The auditee usu-
ally knows there is going to be an audit
but doesnt yet know the specics.
ANSI/ISO/ASQ QE19011S-2004,
Guidelines for Quality and/or Environ-
mental Management Systems Auditing,
says the initial contact with the auditee for
the purpose of arranging the actual audit
can be formal or informal.
The dictionary denition for informal
The level of formality will
depend on the type of audit.
misses the mark for the auditing eld. The
dictionary links informal with the word
casual, an activity carried out without
formal prescribed procedure or ceremony,
and lists a synonym as unofcial.
Initial contact with the auditee is not
unofcial because the audit program
manager or lead auditor must follow
certain guidelines. Further, I would never
describe any part of an audit as casual.
The person making the initial contact
should align the formality with the situa-
tion and comply with audit program pro-
cedures. An example of being very formal
could be a letter informing the auditee of
its obligations, asking the auditee organi-
zation to provide specic information and
BY J.P. RUSSELL
stipulating the required communication
methods. The letter could be signed by
the audit team leader and audit program
manager, with an optional cover letter
signed by the person or the representa-
tive of the organization that requested the
audit (audit client).
Formal or informal
A fax or an e-mail is less formal than a
letter sent by express mail, although all
provide a record of communication. E-
mail responses are normally more highly
structured than rst contacts made by
telephone. I would, however, recommend
following up a telephone conversation
with a note or an e-mail reviewing what
was discussed and agreed on.
Finally, dont overlook a face-to-face
meeting that could be considered informal
or formal. For internal audits, this could
be chatting in a hallway or dropping by
the managers ofce. For external audits,
it might be a formal meeting in the general
managers ofce or in a conference room.
The level of formality in the initial
contact will depend on the type of audit.
External audits are more formal than
internal ones. Third-party regulatory and
certication audits are more formal than
internal department audits.
Who and what
The auditing standard supplement then
says contact should either be made by
those responsible for managing the audit
program (audit program manager) or the
audit team leader (lead auditor). The stan-
dard lists action items to be completed as
part of the initial contact. The seven ac-
tion items make up the purpose or reason
for the initial contact with the auditee:
1. Conrm authority.
2. Provide information about timing and
the audit team.
3. Establish communication channels.
4. Request access to documents and re-
cords.1
5. Determine site safety rules.
6. Make arrangements for the audit.
7. Agree on attendance of observers and
need for guides.
1. Conrm authority
The audit team and its organization must
rst verify that it has authority to conduct
the audit. The lead auditor or audit pro-
gram manager does not ask the auditee
for the authority, but instead states that an
audit is planned and asks if there are any
problems or issues regarding that plan.
Normally, authority for the audit has
already been communicated by top man-
agement, and the auditee will say he or
she was expecting it or knew something
was planned. Every once in a while, how-
ever, the auditee representative will say,
No one told me there was going to be an
audit. Or they might ask if the audit team
has approval from an executive or clear-
ance to view proprietary operations.
June 2008 QP 67
 STANDARDS OUTLOOK
For internal audits, authority to con-
duct the audit has been previously agreed
to by management, but there could be
department issues regarding approval or
the timing of the audit.
If you are conducting an audit to verify
compliance to regulations, the auditing
organization has legal authority to con-
duct the audit. For second-party audits,
in which the customer has contractual
authority to conduct the audit, it is still im-
portant to ensure the supplier is OK with
the plan to conduct an audit.
Though not specically mentioned in
ISO 19011S, conrming authority should
include the type of auditquality manage-
ment system, integrated management
system or processto be conducted.
Conrmation of authority is a simple,
straightforward action item, but if skipped
or assumed, it can cause a considerable
amount of angst, confusion and embar-
rassment later on.
MORE ON AUDITING
To read more articles on standards and auditing, go to the QP website
at www.qualityprogress.com.
68 QP www.qualityprogress.com
2. Provide information
This step answers the when and who
questions. Composition of the audit team
needs to be shared with the auditee rep-
resentative and should include how many
auditors, auditors in training, interpret-
ers, technical experts or specialists, and
observers will participate.
If certain auditors are going to focus on
specied areas, as in a process audit, they
can also be identied. For instance, one
auditor might be assigned to order entry
and another to the rebuild shop. (They
should also be identied in the audit plan.)
Next, the auditor needs agreement on
when to conduct the audit, although a
date or time period might have already
been established in a published schedule
or agreement between the client, audit
program manager and auditee manage-
ment.
The lead auditor and audit program
manager will say the audit is scheduled for
a particular date or time period (sometime
this week, perhaps). Normally, the auditee
acknowledges the date. Things change,
however, and circumstances can dictate a
change in the date or time period.
It is wise to reschedule if events
beyond the auditees control occur and
could put achievement of the audit
objectives at risk. Perhaps the audi-
tee organization received a statewide
award and is going to be visited by its
state congressional representative that
day. Perhaps the day scheduled for
the audit turns out to be the same day
the Environmental Protection Agency
or National Institute for Occupational
Safety and Health decided to audit the
organization.
Changing the timing, however, might
change the audit team composition, de-
pending on individual availability.
3. Communication channels
Normally, the auditee organization has
identied a person to represent it by being
a conduit for all communications. The
communication channel could be a person
or the means for transmitting information.
Perhaps all requests for information will be
channeled through an audit coordinator.
The type of things that might need to
be communicated are: requests for infor-
mation, changes in schedules or logistics,
audit team changes, or changes that have
been agreed onor will be agreed on
when the audit plan is issued.
Establishing communication channels
might also include naming the preferred
communication tool. The lead auditor or
audit program manager might want to be
contacted by e-mail, and the auditee repre-
sentative might want communications to
go to his or her mobile phone. In some
cases, a fax might be preferred because it
creates a record.
The higher the noncompliance or
nonconformity risks, the more formal
the communication channels should be.
Audits that represent a high risk to the au-
ditee or audit program should be recorded
in some manner for retrieval purposes. In
some cases, it might also be prudent to
request e-mail or voicemail verication
that the message was received.
For internal audits, the auditee
representative might be the department
manager, management system representa-
tive or someone assigned the role of audit
coordinator. ISO 19011s U.S. supplement
requires the auditee representative for
second-party audits to have the necessary
knowledge and authority to respond to the
needs of the audit that are communicated
by the audit team.
4. Document and record access
Since the audit plan has not been issued
yet, it is important to ensure that access to
documents and records is not a problem.
If it is a quality audit, the auditor will want
It is important to ensure that access
to documents and records is not
a problem.
access to documents related to the qual-
ity system. If it is a process audit of the
scheduling section, for example, the audi-
tor will want access to those documents
and records.
This is not a major issue for internal
audits, but there could be condentiality
issues for an external audit. With electron-
ic documentation systems, there could
be internet or intranet security concerns.
Electronic access might also require
some minimal training on how to navigate
through the software program and how to
access the help desk.
Access to documentation could be one
of the communication channels you need
to establish. Identify the contact person
for accessing needed documents and re-
cords. Determine the necessary protocols
in advance. If not now, agree on when and
who should be involved in establishing the
protocols.
5. Determine site safety rules
Almost all organizations have some type
of personal safety rules. If it is a manufac-
turing or heavy equipment site (such as a
construction site or an airport), there will
be many rules. If it is an insurance ofce,
the only rules might relate to emergency
exits or bomb threats.
I recommend the audit team leader
or audit program manager be proactive
by ensuring and verifying that he or she
is given all the rules. It can be upsetting
to nd out during or after the audit that
those who cross a yellow line on the shop
oor are supposed to wear ear protection
or steel-toed shoes. This oversight puts
the auditor in a position of breaking a per-
sonal protection rule as he or she veries
conformity to other rules.
Canned safety presentations some-
times must be viewed before access
to a property is allowed. This must be
conveyed to the audit team prior to the
audit. I have arrived to conduct an audit,
only to learn I must watch a 30-minute
safety presentation before I start. This
is an immediate disruption and puts the
audit team 30 minutes behind schedule
from the get-go.
If the safety presentation is avail-
able online as a streaming video or slide
presentation, you can include it in your
preparation steps before arriving at the
audit location. Later, when the audit plan
is distributed, the audit team leader might
again ask about personal safety require-
ments.
6. Arrange the audit
I interpret making arrangements for
the audit as conveying the meeting and
workspace requirements with the auditee
representative. Conversely, the auditee
representative might share circumstances
that are particular to his or her situation.
For external audits, the audit team will
need a base camp to get organized, review
results and perhaps conduct meetings.
This could be a conference room, training
room or empty ofce. Either during the
initial contact or when the audit plan is
distributed, the audit program manager or
lead auditor needs to tell the auditee rep-
resentative specically what is needed in
the room, such as electricity, internet ac-
cess, telephone, ip chart, tables or desks,
and chairs. If the audit is an external one,
the auditor might ask about lodging in the
area for the audit team.
7. Observers and guides
Guide and observer responsibilities are
discussed in ISO 19011S, clause 6.5.3. At
the initial contact step, information about
the possibility of an observer attending
the audit must be shared.
Perhaps someone could be observing
for the purpose of certifying the auditor
organization, or someone from the audi-
tor organizations management could be
observing for audit program or auditor
performance assessment.
For most external audits, guides should
be requested. For internal audits, guides
might not be needed unless the operations
are complex and the auditors are unfa-
miliar with departments or the functional
areas being audited.
Ensuring all seven requirements are
addressed will get you off to a good start.
Planning, as in the plan-do-check-act
cycle, is essential for ensuring that the do-
ing part of the cycle is effective.
After the initial contact has been estab-
lished, your next step might be to conduct
a review of the documents and records
requested to ensure conformity to the
audit criteria.
NOTE
According to ISO 9000:2005, a record is a special kind of
document stating results achieved or providing evidence
of activities performed. Procedure documents and record
documents are quite different. For this article, the author
chose to name documents and records separately to avoid
confusion and repeated explanations that documents
include records.
J.P. RUSSELL is president of J.P. Russell
& Associates, Gulf Breeze, FL, and
managing director for the Quality
WBT Center for Education at www.
QualityWBT.com. He is a fellow of
ASQ, ASQ certied quality auditor,
voting member of the American
National Standards Institute/ASQ Z1
committee and member of the U.S. technical advisory group
for International Organization for Standardization technical
committee 176. Russell is the author of several ASQ Quality
Press books, including Internal Auditing Basics (second edi-
tion) and Process Auditing Techniques, and editor of the ASQ
Audit Handbook (third edition).
June 2008 QP 69
 CAREER CORNER BY JOE CONKLIN

A Ready Rsum
Use performance review notes to keep your rsum up to date
DID YOU EVER suddenly hear about You can leave your rsum untouched task of keeping your rsum fresh.
that job youve been dreaming of? Ever for months and years and then face a Using performance review notes is
had to look for a new job when you least time-consuming rewrite when time is of the best way to keep your rsum ready.
expected it? At times like this, a ready the essence. Or, you can avoid some pain By performance review notes, I mean
rsum is your best friend. and invest a few hours each year to the the documentation of your accomplish-
ments, plans, training and improvement
WARREN T. KOSTUS suggestions you make prior to your actual
14 Points Lane review.
Algood, CA 91900-1006 Ideally, your boss will accept your
Tel: 555-937-1958 comments and incorporate them into the
Fax: 555-931-1984 nal write-up. If not, your notes are still a
wtkostus@pdqxyz.abc convenient way to organize your thoughts.
OBJECTIVE If you update your rsum every year,
Senior QA engineer with 20 years experience in quality and reliability requirements for state-of-the-art the parts you are most likely to revise are
computer hardware manufacturer. Seeking division manager position in product testing or warranty your objective, contact information and
function. Proven track record of integrating cost of quality tools, teamwork concepts and Six Sigma current job details. Any notes you prepare
principles in product testing operations. before a performance review are fodder
PROFESSIONAL EXPERIENCE for updating the current job section of
Senior QA engineer Report to plant quality manager of $110 million major computer hardware manu- your rsum.
facturer. Supervise four quality technicians. Manage warranty database and reporting function with

$750,000 budget. Choosing accomplishments

Consider the rsum (at left) of Warren
Hardenuf Hardware Inc., Algood, CA (2000-present) Kostus, a ctional quality assurance engi-
Key accomplishments: neer in the warranty data operation of a
Led team that reduced cycle time for warranty forecasts by 25% in two years. computer hardware manufacturer.
Implemented employee suggestions that reduced overhead budget by 15% in 18 months. Kostus current employer conducts
Eliminated one month backload in loading warranty survey data by computerizing data semiannual performance reviewsone
loading function. at midyear and the other at year-end. The
Designed feedback system to incorporate product engineers suggestions for midyear review provides an opportunity
streamlining warranty evaluation procedures. to address emerging issues and problems
QA engineer Report to reliability manager. Train design engineers in procedures for reliability test affecting the departments ability to meet
design and performing failure mode effects analysis. its goals for the year.
For his last midyear review, Kostus
4Ever Testing Inc., Algood, CA 1997-2000 reported the following accomplishments
Key accomplishments: to his supervisor:
Recommended and installed new reliability modeling software for design engineers. Made department record-retention
More accurate models helped engineers achieve 35% improvement in mean time period compliant with new standard.
between failure for new product. Trained staff in state-of-the-art war-
Trained 150 shop oor personnel in control charts as part of companys statistical ranty cost forecasting techniques.
process control program from 1996 to 1998. Reduced variation in critical widget Reorganized monthly forecast le so
dimensions by 12%. most important forecasts appear rst.
Set new record for timely submission

70 QP www.qualityprogress.com
 of departments contribution to annual
division review.
Streamlined and simplied department
audit procedures.
Helped interdepartmental team nal-
ize recommendations for use of the
internet to enhance major customer
database updates.
The following accomplishments were
noted for his last year-end review:
Automated key quality edits of major
department publications.
Implemented improved orientation
program for departments new hires.
Helped record number of staff mem-
bers obtain permission to share inter-
esting research ndings at professional
society conference.
Automated annual regeneration of criti-
cal product mailing list.
Improved release of key warranty
reports so they were current within 60
days instead of 90.
Kostus is now ready to update his rsu-
m. His objective and contact information
are the same as last year, so he concen-
trates on the description of his current
job. The department lost one technician to
retirement. Investment in modern report-
ing technology required an increase in
his budget, and the resulting productivity
improvements allow the remaining techni-
cians to perform the same work as before.
Not having room to list all 11 ac-
complishments for the last year on his
rsum, Kostus wants the best balance of
old and new. After some thought, he decides
to remove two of the old ones and replace
them with two new ones.
The last accomplishment he listed on
his performance review notesspeeding
up release of warranty reportshas the
advantage of being an easily grasped mea-
sure of success, so he chooses it as one of
the two. And, to better highlight his grow-
ing ability to manage and develop people,
he selects the accomplishment about the
new hire orientation program.
After 10 minutes of cutting, pasting and
editing, the part of Kostus rsum pertain-
ing to his current job is updated (see Key
Accomplishments Section of Undated R-
sum). If his next big career opportunity
knocks, he is ready to open the door.
JOSEPH B. CONKLIN is a statistician
with the U.S. Department of Energy,
helping to assess the quality of its
natural gas survey data. Conklin
earned a masters degree in statistics
from Virginia Polytechnic Institute. An
ASQ member, he holds the follow-
ing certications: quality engineer,
reliability engineer, quality auditor, quality manager and
software quality engineer.

Quality Education
UPDATED RSUM SECTION
for
Key accomplishments:
Quality Professionals
Led team that reduced cycle time for warranty
FLEXIBLE
forecasts by 25% in two years. Complete ALL of your
Master of coursework online at your
Implemented employee suggestions that re- Science location.
duced overhead budget by 15% in 18 months. in AFFORDABLE
Quality All distance students pay the
Improved release of key warranty reports so Assurance same low tuition rate of $227
they are current within 60 instead of 90 days. per credit hour.
Graduate RIGOROUS
Implemented improved orientation program for Green Belt Learn from PhD-level
departments new hires. Certicate instructors who know the
theory AND its application in
the quality !eld.

www.msqa.edu/online
PEAK PERFORMANCE 678-915-SPSU (7778)
For a more in-depth discussion and example of performance
review memos, read Joe Conklins earlier column, Making 1-800-635-3204
Performance Reviews Work for You, at www.qualityprogress.
com.
1100 SOUTH MARIETTA PARKWAY
MARIETTA, GEORGIA 30060-2855

June 2008 QP 71
 STATISTICS ROUNDTABLE
The Reality of
Residual Analysis
Its easy to overlook this important technique
when evaluating an analysis model
IN THE WORLD of statistics textbooks,
independent random samples of size 30
from a normal distribution are a dime a
dozenthe norm rather than the excep-
tion. You simply perform the correct
analysis (perhaps the one you just studied
in the most recent chapter), everything
goes like clockwork, and you move on to
the next problem.
Unfortunately, the real world stubborn-
ly refuses to conform to this alternative
reality of statistics textbooks. In the real
world, you rarely nd normal distribu-
tions, much less independence. In addi-
tion, the variable of interest is typically
impacted by variables not in your data set,
as unfair as this might sound.
Therefore, you dont have the luxury of
simply doing the obvious correct analysis
and moving on. Rather, to avoid the pot-
holes in the road to statistical insight, you
should always take a critical look at your
Residual plots, patterns and trends
Plot Pattern
Run chart (sequence) Shifts and trends
Normal probability Nonlinear
Predicted values V-shaped
Curved
Histogram Non-normal
Outliers might show up in any of these plots.
72 QP www.qualityprogress.com
BY ROGER W. HOERL
own analysis to ensure you havent missed
something important.
Statistics, not futures
One way to do thiscertainly not the
only wayis through residual analysis.
No, were not talking about the futures
market. These are statistical residuals, not
nancial residuals.
Residuals from any model are helpful
in evaluating the adequacy of the model
itself relative to the data and any assump-
tions you might make in the analysis. A
residual is simply the difference between
the observed value of y (the response
variable of interest) and the value of y
predicted by the model:
Residual = y observed y predicted
There is one residual for each ob-
servation. Statistical software typically
standardizes residuals to put them on a
common scale. How this gets done goes
/ TABLE 1
Source of problem (?)
Missing variables
Special cause
Residuals not normal
Consider transformation
Consider transformation
Consider higher order terms
Consider transformation
Residuals not normal
Consider transformation
beyond the scope of this article. Obvi-
ously, if you had a model that t the data
perfectly, the residuals would all be zero.
As noted earlier, analysis of the residu-
als is an effective method for assessing the
t of the model to the data and determin-
ing whether the model is useful. The rec-
ommended approach is to study a variety
of residual plots and look for patterns and
trends.
Generally speaking, the absence of
any patterns and trends (for example, a
random scatter of points) indicates the
model is adequate. In other words, the
residuals will be boringwith no note-
worthy patterns or trendsbecause an
adequate model accounts for the system-
atic (predictable) behavior in y. Therefore,
what remains in the residuals will be just
random variation or noise.
If the model is not adequate, however,
such as a model that does not account
for curvature in y, then the residuals will
contain both random variation and curva-
ture. So, counter to your intuition, you are
hoping to see nothing in residual plots and
are concerned about seeing something
interesting.
This interesting pattern, however,
might alert you to something important
that you overlooked. In general, you
should not consider the modeling process
as complete until the residuals have been
thoroughly evaluated.
When estimating a statistical model,
the analyst often assumes the residuals
Minitab residual plots for goodwill / FIGURE 1
scale (normal probability chart).
Residuals versus observation sequence
Residual model diagnostics
(run chart or individuals control chart).
Normal plot of residuals I chart of residuals
1
These plots are not independent of
5 5
each other and should be interpreted as a
1
3.0SL = 1.885 set. It is not uncommon for a problem or
Residual

Residual
7 7 =
0 0
7
X = 0.001043 issue to show up in more than one plot. It
7 3
-3.0SL = -1.883 is also important to recognize there is a lot
1
of judgment used in the interpretation of
-5 -5 1
-2 -1 0 1 2 0 10 20 30 40 50 residual plots. Your ability to properly in-
Normal score Observation number
terpret these plots grows with experience.
Histogram of residuals Residuals vs. ts
50
5 Residuals vs. predicted values
40
If the model gives an adequate t to the
Frequency

30
Residual

data and the typical assumption of inde-

20
10
0 -5
-4 -3 -2 -1 0 1 2
Residual
are sampled from a common normal dis-
tribution with constant variance and no
correlation between different residuals.
These assumptions are not made in every
analysis, but they are frequently made.
There are numerous residuals tests
Minitab residual plots for log
0
3 4 5 0 500
Fit
and plots that you can make to evaluate
these assumptions. Here are four plots to
consider:
Residuals versus predicted values.
Histogram of the residuals.
Residuals versus normal probability
pendent normally distributed residuals is
satised, the plot of the residuals versus
predicted values should be boring and
1000
show no pattern or trend.
When you observe a nonrandom pat-
tern, you should consider changing the
form of the model. One pattern commonly
observed is an increasing variation in the
residuals as the predicted values increase
(V-shaped pattern or megaphone shape).
This indicates the assumption of equal
variance of the residuals is not satised by
the data. The V-shaped pattern suggests
the variation increases with the average.

(goodwill) / FIGURE 2 In this case, a transformation, possibly a

log transformation (for example, analyze
Residual model diagnostics
log y in place of y) or weighted regression
Normal plot of residuals I chart of residuals
should be used. As a rule of thumb, the log
4 4 1
3 3 3.0SL = 2.777
transformation is often helpful if the ratio
2 2 of maximum y to minimum y is greater
Residual
Residual

1 1
= than about 10.
0 0 X = 0.003094
-1 -1 A curved relationship between the
-2 -2
-3.0SL = -2.771
residuals and predicted values suggests
-3 -3
-2 -1 0 1 2 0 10 20 30 40 50 that curvilinear terms (for example, x2)
Normal score Observation number should be included in the model. With
Histogram of residuals Residuals vs. ts several predictor (x) variables, it often
15 4 helps to plot the residuals versus each
3
x variable to pinpoint the source of the
10 2
Frequency

Residual

1 curvature. A transformation of y, such as

0 the log transformation mentioned earlier,
5
-1
might help remove the curvature and the
-2
0 -3 need for additional terms to be added to
-3 -2 -1 0 1 2 3 4 0 1 2 3
Residual Fit
the model.

June 2008 QP 73
 STATISTICS ROUNDTABLE

Histogram of residuals
A histogram provides a view of the overall
distribution of the residuals. Do they ap-
pear bell-shaped as you often expect? Are
there any outliers? Outlier residuals result
when the model does not adequately t
one or more observations.
When outliers exist, they typically show
up in several, if not all, of the residual plots.
Outliers suggest the observation is due to
a special cause (perhaps a measurement
error or keying mistake) or the model is
inadequate (missing variables and missing
model terms).
Residuals vs. normal
probability scale
The normal probability plot of the residu-
als is constructed to check on the normal
distribution assumption of the residuals.
This technique is better to use than the
histogram because the linearity pattern
you are looking for is easier for people to
perceive than a bell-shaped pattern. The
histogram, however, can help you pick up
other abnormalities, and it is still a useful
plot.
In the normal probability plot, the

FRESH
residuals are ordered by size and plotted
versus the normal probability scale, which
is, fortunately, calculated by the computer.
If the plot approximates a straight line,
the assumption that the residuals follow a
SPC charts normal distribution is reasonably satised.
Keep in mind that no real data will ever

Instantly follow a normal (or any other) distribu-

tion exactly. So you are not looking for
from any data a perfect line in this plot, just a general

source linear trend.

The presence of a curvilinear relation-
ship in the normal probability plot of the
residuals suggests the normal distribution
assumption is not satised. The curvature
Point,
P i click,
li k chart
might be because the relationship between
y and one or more of the x variables is not
Download a 30-day trial at www.pqsystems.com modeled properly.
CHARTrunner is SPC software that fetches data from Excel, Access, For example, you might need to add
Oracle, SQL Server, and other databases to generate up-to-the-minute an x2 term to the model. In this situation,
charts. It is the only process charting software that always provides as noted earlier, it might also be appropri-
fresh charts with no ate to consider transformations for y and
importing, exporting, one or more of the predictor variables
copying, or pasting.
Youve really got to
see it to believe it.
STATISTICS ROUNDTABLE
CORRECTION
The April Statistics Roundtable, It
Depends, contained the wrong
chart in Figure 2, EWMA chart
using subgroup six of n = 2. An
updated version of the column
is available online at www.asq.
org/quality-progress/2008/04/
North and South America Australia and Asia Europe and Africa statistics/statistics-roundtable-it-
Call 800-777-3020 800-777-3020 03-9770-1960 01704-871465 depends.html. We regret the error.

74 QP www.qualityprogress.com
(xs). These plots are diagnostic. Again, cal assumptions are almost always for the BIBLIOGRAPHY
Box, G.E.P, W.G. Hunter and J.S. Hunter, Statistics for Experi-
considerable judgment is required to use residuals, not the original y variable. The menters, Wiley Interscience, John Wiley & Sons, Inc., 1978.

them properly. y variable will generally contain both

ROGER W. HOERL is manager of Gen-
random and systematic variation. eral Electrics global research applied
Residuals vs. observation statistics lab. He has a doctorate in
applied statistics from the University
sequence REFERENCE of Delaware. He is an ASQ fellow, an
ASQ Brumbaugh Award recipient and
This plot of residuals versus time sequence 1. Roger W. Hoerl and Ronald D. Snee, Statistical Thinking:
Improving Business Performance, Duxbury Press, 2002. a GE Global Research College fellow.
should show no trends if the model is
adequate and no special causes if you are
using the individuals control chart. A trend
would suggest there are one or more vari-
ables not currently in the model that have
changed during the time spanned by these
data, and it would improve the predictabil-
ity of the model if this variable could be
added to the model.
A summary of some patterns and trends
found in residual plots and some likely
problem sources are summarized in Table
1 (p. 72).

An example
Figures 1 and 2 (p. 73) are sets of residual
plots using Minitab software.1 They come
from a regression analysis of a nancial
accounting variable: goodwill.
In Figure 1, you see a number of
problems, including outlier residuals in
the individuals (I) chart, curvature in the
normal plot, a very peaked histogram and
less variation at low values of y than high
values of y (residuals versus predicted
values or ts).
Based on this residual analysis, the ana-
lysts tried a log transformation of y and one
x variable, and repeated the analysis, pro-
ducing the residual plots in Figure 2. You
can see these look much more random and
boring, although there is still one outlier.
Rarely do we nd perfect residuals in
real applications.
ProFicient Plants are:
Proactive. They prevent defects before they result in
scrap, rework, or even recalls.
Process-Oriented. They understand process behavior across all
parts for true process control.
Productive. They have insight into planning and
maintenance to increase productivity and
reduce costs.
Only ProFicient SPC software from InfinityQS can provide the real-
time statistical tools you need to ensure optimal product quality and
reduce the costs of scrap and rework. ProFicient is uniquely able to
quickly and easily analyze multiple parts across various processes
on a single chart for a 360-degree view of process behavior.
Call 1-877-772-7978 or visit www.in!nityqs.com to learn more about making
your plant ProFicient.

Final warning Pro!cient manufacturing needs ProFicient software.

Some Six Sigma practitioners with limited
backgrounds in theoretical statistics have
recommended doing normality or other
tests of the y variable before modeling. www.in!nityqs.com 1.800.772.7978
This is not good practice, as the theoreti-

June 2008 QP 75
 QUALITY IN THE FIRST PERSON
All Roads Lead to Quality
Navy experience, ASQ certication and
mentors carve out one mans career path
AT 17, most high school students are in his children. He made it quite clear
trying to decide what they want to do that I was on my own if I wanted to go
with their lives. I thought I knew, but as to college or technical school. Consider-
I have seen with many co-workers over ing the high cost of four years of higher
the years, original aspirations often lead education, I did not relish the idea of
down another path. At graduation, I paying off student loans for many years
never would have thought that in 10 years following graduation.
I would work in quality assurance.
My father was a hard-working man Dipping into quality
with strong values who felt it was During my senior year, I looked at the mili-
important to instill a good work ethic tary as an option. A recruiter encouraged
me to test for an advanced program in the
nuclear power eld. I decided to take my
chances, and two weeks after graduating
from high school I was in boot camp.
During this time, I learned that the
Navy wanted me to go into the me-
chanical eld rather than electronics,
where my main interest was. I was a bit
disheartened, but I believe everything
happens for a reason. Little did I know
that it would help lay the groundwork for
where I am today.
Following 18 months of training in me-
chanics and nuclear power, I was assigned
to the USS Lafayette. I was eventually
assigned the additional collateral duty
of calibration petty ofcer, where I was
taught to do some gauge calibrations and
coordinate calibration of other gauges and
instruments. I was responsible for making
sure everything was calibrated according
to schedule and maintaining the records.
As this process is often viewed as a
SHARE YOUR JOURNEY
If you have an interesting quality story to tell, consider submitting it for this
column. E-mail your article to editor@asq.org and share your journey.
76 QP www.qualityprogress.com
BY HENRY W. ROBINSON
quality assurance function, this was my
rst exposure to the eld of quality
although I hadnt considered any career
in the quality eld at that time. During
my four years onboard the Lafayette, I
also assumed responsibility for maintain-
ing the reactor plant manuals, standard
operating procedures and all reactor
plant records.
Through schooling and maintenance
work, I became familiar with the Navys
quality assurance program, which was
designed to maintain the safety of the
nuclear submarine eet. While under-
going a two-year refueling operation, I
participated in various inspections and
testing to ensure uid systems met all
necessary requirements.
These job functions further laid the
groundwork for the future, although
again, I never thought of the quality eld
as being a viable career path.
After four years onboard the Lafayette,
I became a Navy recruiter. I was proud of
my accomplishments and what the Navy
had done for me, and I was eager to share
my experiences and help others achieve
their goals through the organization.
While this assignment did not have much
to do with quality, it allowed me to guide
young men and women into opportunities
they might not have had elsewhere. I was
successful in this position, and I achieved
numerous citations and awards, including
the Naval Achievement Medal.
After completing my tour as a recruit-
er, I decided to pursue a civilian career.
I completed my last year in the Navy at
I believe that individually we are all capable
of many things, but collectively we are
capable of greater things.
the Philadelphia Naval Base onboard the
USS Independence. There, I worked in en-
gineering as the training coordinator and
division ofcers assistant.
The job reinforced my commitment
to the importance of training, which
has been a steadfast cornerstone of my
career and is something I place great
importance on today.
Moving on
While waiting to hear back about two
nuclear power plant operator positions,
I applied for a job as a quality control
technician performing product testing at
Electro-Science Laboratories (ESL).
I looked at this as a temporary job until
something better came along, but within
a few months, I started to realize that I
really enjoyed doing this kind of work and
started thinking about the quality eld as a
career avenue.
My quality manager at ESL, a men-
tor to me, encouraged me to join ASQ
and learn more about the eld. I gladly
accepted additional responsibilities and
pursued training courses that would
allow me to further pursue a career in
quality. During this time, I also completed
an associates degree in liberal arts.
After working at ESL for about 18
months, I was promoted to a quality
control supervisor position. I then decided
to look into ASQs certication programs.
It seemed that the quality eld was not
structured in quite the same way as many
of the mainstream engineering disciplines.
While there were a multitude of small
training courses, there were not many
colleges or technical schools specializing
in quality. I did not believe my two-year
liberal arts degree was going to do a lot
for me, as it became apparent that recog-
nition in this eld was largely based on
experience and certication.
The next logical step for my career
was to become a quality engineer.
Typically, trying to promote yourself as
a quality engineer without a bachelors
degree or experience in the position can
be quite difcult.
ASQ certication was a means to that
end. Those who have pursued this know
how difcult it can be and what it takes
to achieve it. It truly recognizes knowl-
edge and experience.
After working at ESL for three years, I
achieved ASQs quality engineer certi-
cation, something I am still proud of 16
years later.
Certication lands a job
The certication allowed me to apply for
a quality engineer job with ITT-Goulds
Pumps. The manager, an ASQ certied
quality engineer, was well aware of the
certication process. Recognizing this
certication and my past experiences
with the Navy and ESL, he hired me. He
became another career mentor, as he
encouraged me to further my education
and training.
As a quality engineer, I completed my
bachelors degree in quality engineering
and became a certied quality auditor. My
manager, looking to the future, made sure
I was exposed to all different aspects of
the quality eld through varying job as-
signments and training opportunities.
After working as a quality engineer for
seven years, my boss decided to retire,
and I was offered the position of quality
systems manager, which I have held for
eight years. Most recently, I was pro-
moted to quality assurance manager with
some additional responsibilities as my
previous position.
Role reversal
In my role as quality assurance manager
and previously as quality systems man-
ager, the mentoring I previously received
has inspired me to encourage others in my
department to seek training opportunities
that will better them personally, provide
greater versatility of resources within the
department and strengthen our abilities to
better serve the needs of the business.
I believe that individually we are all ca-
pable of many things, but collectively we
are capable of greater things. Continuing
to develop ourselves through training
whether it be formal or on the jobis key
to our success in meeting the needs of the
business and our customers in this ever-
changing world.
HENRY W. ROBINSON is a quality
assurance manager for ITT-Goulds
Pumps in Ashland, PA. He earned his
bachelors degree in quality engineer-
ing. Robinson is a senior member of
ASQ and a certied quality auditor
and engineer.
June 2008 QP 77
 QPTOOLBOX
Food grade lubricant and transient signals. It supports record-
Dow Corning has unveiled a lubricant ing up to 20 analog chan-
for use in food and beverage processing nels to an internal hard
plants. drive at sample rates
The Molykote P-1900 Food Grade Paste of up to 500 kHz and
can be used to lubricate sliding surfaces a bandwidth of 100
and friction contacts exposed to heavy kHz per channel.
loads, especially at low and medium The Dash 20HF
speeds. The lubricant has a low coefcient recorder can
of friction, good water resistance and high provide real-time
load-carrying capacity, and it can be used in display and analysis of
temperatures ranging from -30 to 300 C. data. High frequency data can be
The paste meets FDA regulations and saved directly to a hard drive for review
is safe in food and beverage processing and analysis later.
plants where incidental food contact might The recorder is for use in aerospace,
occur. automotive, electric utility, pulp and paper,
E-mail: industrial@dowcorning.com; visit metal manufacturing and medical settings.
www.molykote.com. Call: 877-867-9783; visit www.
astro-med.com/tmindex.html.
High frequency recorder line
Astro-Med, a recording systems manufac- Indoor/outdoor keypads
turer, has released a high-frequency data Illinois Tool Works Switches has unveiled a
recorder called the Dash 20HF. line of access control keypads.
The portable unit was designed for the The units, called ActiveMetal keypads,
long-term capture of high-frequency data are activated by touching the raised
sculpted keys and
are constructed
in solid metal. The
keypads are available
in both 12-key and
16-key congurations.
Intended for indoor
or outdoor use, the
keypads are resistant
to weather elements,
as well as water and
dust.
The ActiveMetal
keypads, rated to op-
erate in temperatures
from -40 to +85 C,
are unaffected by
vibration and humidity
78 QP www.qualityprogress.com
and are rated at
+9 to +24 VDC
at 50mA (max).
The keypads
can be used
for point-of-sale
kiosks, security
and alarm systems,
keyless entry points,
gas pumps, ATMs, clean rooms, and
other industrial, commercial and retail
settings.
Call: 800-544-3354; visit: www.
itwswitches.com.
Tinted desiccator
A manufacturer of laboratory products has
released a desiccator that reduces visible
light penetration more than 50%.
The Secador Amber Tinted Auto-Des-
iccator also blocks 99% of UV light and is
resistant to staining, crazing and chemical
attack.
Produced by Bel-Products, the unit
reduces relative humidity to 25% and circu-
lates dry air automatically to ensure safe,
dry storage. The desiccator is available in
four sizes and can be stacked together to
save space.
Call: 1-800-423-5278; visit http://
ww.belart.com.
Crush sensor
Endevco, a manufacturer of vibration,
shock and pressure measurement instru-
ments, has introduced a crush test ac-
celerometer.
The Endevco model 7287, primarily used
in vehicle crash tests, is placed on bum-
pers, fenders, doors and trunk lids to gather
critical data on vehicle crush zones.
The new accelerometer features an in-
put voltage of 10Vdc and output sensitivity
of 0.10 mV/g. Its full scale range is 2000 g,
and its frequency response is 4,000 Hz with
a resonant frequency of 20,000Hz.
Three form factors are available to allow
maximum exibility: two with cable and
one without cable for users who prefer to
perform their own wiring.
Call: 949-295-5515; visit www.endevco.
com.
data with real-time control of laser inten-
sity. This allows for the accurate measure-
ment of data from most surfaces that vary
in texture, color and reectivity. The sensor
has a measurement range of 1,250 mm and
can provide resolutions to 0.15 microns.
Call: 770-888-6586; visit www.
lmitechnologies.com.
Corrosion data collector

Sensors for tires and rubber
A series of in-process noncontact sensors
for the tire and rubber industry has been
introduced by LMI Technologies.
Called the Selecom SLS5000 series,
the sensors feature a fast, automatic light
control circuit for measuring on varying
surfaces from mat topography to shiny
black materials, such as rubber.
The sensor provides high-speed, 16 kHz
Theres a new handheld portable corrosion
data collection instrument to help collect
data in hazardous settings.
The Checkmate DL, developed by Rohr-
back Cosasco Systems, lets users read cor-
rosion data from a data logger or remote
data collectors. Users can later transfer the
data to a computer for graphing and corro-
sion analysis.
The portable unit runs for about 10
hours on six AA batteries and can oper-
ate in temperatures of -18 to +50 C. The
instrument can capture and store as many
as 50 readings of different corrosion data
loggers.
Visit: www.cosasco.com.
Output beam expanders
be specied at single and dual wave-
lengths, from 248 nm to 1550 nm, and have
damage thresholds of 500 MW/cm2.
The beam expanders allow for proper
collimation and a more controlled focus or
defocus of the beam. They can be used in
several laser processing and micromachin-
ing tasks, such as cutting solar cells, mark-
ing diverse materials, drilling metal sheets,
structuring metal foils and scribing ceramic
substrates.
Call: 973-366-7289; visit: www.
specialoptics.com.

A new family of large output laser beam

expanders has been released by Special
Optics, a lens systems manufacturer.
The new beam expanders feature:
Diffraction-limited optical designs with
large output clear apertures of 10 mm
to more than 100 mm.
Expansion ratios from 5X to 40X.
Input apertures of 1.3 mm to 20 mm.
The large output beam expanders can

GOT A QUALITY PRODUCT?

Send your product description and photo to vfunk@asq.org.

June 2008 QP 79
 QPREVIEWS
Six Sigma for Technical Processes
Clyde M. Creveling, Prentice Hall, 2006,
358 pp., $34.99 (book).
It is hard not to feel as if every new Six
Sigma book will become a dust collector on
your bookshelf. After all, with so much in
print about the subject, surely there arent
any new ideas to be found. Six Sigma for
Technical Processes, however, does a great
job of disproving that notion by offering a
new perspective on what is becoming an
old conversation.
Creveling does not bore his readers
by rehashing old ideas. Everyone knows
Six Sigma is a toolbox of problem-solving
techniques meant to be applied to any type
of organization. Most quality practitioners
do not need to read
more about how to
use the ve whys or
construct a Pa-
reto analysis when a
problem exists. What
they do need to know
is what Creveling
provides: instruction
for how to develop a technical process
designed to prevent, if not totally eliminate,
many problems before they occur.
This is a grand aspiration. After all, the
problems faced by practitioners in the
real world are far more complex than
the simplied problems written about by
many authors. Creveling tries to cope with
that complexity by proposing a triangle
of process excellence composed of a
product development process, a project
management plan and a collection of tools
and best practices. These are a mouth-
ful to say together in one sentence, but
the author does a ne, albeit sometimes
wordy, job of reducing the steps to bite-
sized chunks.
80 QP www.qualityprogress.com
Six Sigma for Technical Processes takes
time to read and digest, but the book is
loaded with new ideas worthy of consider-
ation. I think you will be just as surprised as
I was by how fresh Crevelings ideas are.
Reviewed by Tim Knight
Evergreen Park, IL
Commercializing Great Products
With Design for Six Sigma
Randy C. Perry and David W. Bacon, Prentice
Hall, 2007, 656 pp., $89.99 (book).
The purpose of this book is to assist
people in identifying, developing and com-
mercializing new products and services us-
ing the design for Six Sigma (DFSS) method.
The book is divided into ve sections that
take the reader through the DFSS process:
getting started, preparing the business plan,
obtaining the voice of the customer, devel-
oping the product or process, and analyzing
the product launch and project postmortem.
Each of the subsequent sections guides the
reader through the details of using the DFSS
toolset, including KJ analysis, quality func-
tion deployment (QFD), failure mode effects
analysis and statistical optimization tools.
The main strength of the book is how
it lays out a roadmap of the DFSS process,
which is broken into six stages and shows
the reader how and when to use the vari-
ous DFSS tools, as well as how each tool
logically links to the next. This is important
because a lot of Black Belts run into trouble
with how the tools relate to one another.
The book goes into much detail on
KJ analysis and using QFD to translate
customer needs into key process param-
eters. The book also emphasizes the use
of nancial analysis to scope out a project
and stage-gate tools to monitor it, and uses
Minitab and Excel to perform the needed
statistical calculations.
The book could use improvement in
two areas. The rst is how it relates the
roadmap to one of the established DFSS
sequences. It is confusing to read about all
of the known DFSS sequences, see where
this roadmap could possibly t one of
them, and then proceed through the rest of
the book using a totally different roadmap.
The second area that could be improved
has to do with how the authors fail to
relate DFSS to arenas other than manufac-
turing. The development of a new candy
wrapper is the main
example used to ex-
plain how the entire
DFSS toolset can
be employed. This
can lead people to
believe DFSS is only
good for manufac-
turing applications.
Having examples from nonmanufacturing
areas would have shown that DFSS can be
used to develop new products or process-
es in other industries such as healthcare,
insurance or nancial services.
Overall, the book does a ne job showing
how DFSS can be used in new product or
process development. It is detailed enough
that someone could use it to learn DFSS or
as a reference while on a DFSS project.
Reviewed by Brian Cocolicchio
Teterboro, NJ
A Lean Guide to
Transforming Healthcare
Thomas G. Zidel, ASQ Quality Press, 2006,
169 pp., $68.25 list, $40.95 member (book).
The healthcare industry poses unique
challenges to managers and practitioners
interested in process and quality improve-
ment. Those looking for guidance could
turn to this short book, which provides an
overview and examples of how to apply The rst half of the book presents an In this new edition, the structure of the
lean methods associated with the Toyota overall business process improvement rst half of the book has been reworked,
Production System. model. It deals some new tools have been added, and
The book consists of 12 chapters that with understanding a new, extensive case study has been
comprise a roadmap for implementing lean and modeling your included. Those changes make for an enjoy-
methods in healthcare, including the use of current business able and clear presentation. As a result, the
value-stream mapping, the Five Ss, mistake processes us- book is a goldmine of structured support,
proong, Six Sigma, conducting a lean event ing performance suitable for employees and managers at any
and a case study from Yale New Haven measurement in organizational level in any type of industry.
Hospital. Covering these topics in a short improvement work, Reviewed by Bengt Klefsj
book prevents Zidel from providing anything creating a business Lule University of Technology, Sweden
but an overview of each subject. Anyone improvement roadmap and organizing
interested in analysis that digs more deeply improvement work. RECENT RELEASES
will have to seek out other references. The second half presents a toolbox Kaizen and the Art of
Zidel is a Six Sigma Black Belt certied related to the improvement model. For Creative Thinking
by the Juran Institute who held the position each phase, a selection of suitable tools is Shigeo Shingo, Enna Products Corp., 2007,
of director of continu- presented, along with their backgrounds, 256 pp., $59.40 (book).
ous improvement at steps for how they can be applied and Story of a Lean Journey
a major hospital for examples of their use. The nal chapters James K. Lewis, Society of Manufacturing
several years. That contain two extensive case studies il- Engineers, 2008, 182 pp., $23 (book).
real-world experience lustrating the full method and a number of Thinking Tools for Kids
allows him to take on templates that support most of the tools Barbara A. Cleary and Sally J. Duncan, ASQ
a complex topic in a presented. Quality Press, 2008, 182 pp., $15 (book).
challenging industry
while still making
the book interesting to read, relevant and
informative.
Advertisers Index
Reviewed by Ron S. Kenett
ADVERTISER PAGE PHONE WEB
Raanana, Israel AssurX, Inc. 7 408-778-1376 www.assurx.com
EtQ, Inc. 1 516-293-0949 www.etq.com
Business Process Improvement Euroquest 57 800-355-3876 www.euroquest.net
Toolbox, second edition InnityQS International, Inc. 75 703-961-0200 www.innityqs.com
Bjrn Andersen, ASQ Quality Press, 2007, Membership, ASQ 65, 66 414-272-8575 www.asq.org
290 pp., $63 list, $37.80 member (book). Minitab, Inc. IFC 814-238-3280 www.minitab.com
Systematic use of toolsstatistical Mitutoyo America Corporation 27 630-820-9666 www.mitutoyo.com
and othersfor structuring and analyz- PQ Systems, Inc. 74 800-777-3020 www.pqsystems.com
Quality Council of Indiana 10, 11 812-533-4215 www.qualitycouncil.com
ing data to achieve quality improvements
Southern Polytechnic State University 71 678-915-7778 www.msqa.edu
has been emphasized by quality greats W.
Sparta Systems, Inc. 5 732-203-0400 www.sparta-systems.com
Edwards Deming and Joseph Juran. The aim
Statistical Process Controls 2 865-584-5005 www.spcpress.com
of Business Process Improvement Toolbox
StatSoft, Inc. 90 918-749-1119 www.statsoft.com
is to give readers practical insight into Terry Holley Consulting, Inc. 9 317-878-4800 www.terryconsulting.com
how different tools can create a coherent Upper Iowa University 64 515-369-7777 www.uiu.edu
improvement system. World Conference Call for Papers, ASQ 89 414-272-8575 www.asq.org

June 2008 QP 81
 QPCALENDAR
To receive information or to register for
ASQ Education Courses and Conferences,
contact Learning Offerings, ASQ, 600 N.
Plankinton Ave., Milwaukee, WI 53203,
call 800-248-1946 or 414-272-8575, fax
414-272-1734 or visit www.asq.org.
JULY
14 ASQ Education Course. Six Sigma
Executive Training. Milwaukee.
14-18 ASQ Education Course.
Facilitating Continuous Improvement in
Education. Milwaukee.
15-17 Shock and Vibration Mea-
surement. Juan Capistrano, CA. Call Yeni
Hoo at Endevco at 949-493-8181 or e-mail
yeni.hoo@endevco.com.
15-18 ASQ Education Course. Six
Sigma Champion Training. Milwaukee.
16-18 Statistical Analysis for Pro-
cess Improvement Using Minitab. San
Francisco. Call Exsilon at 800-930-8528 or
e-mail training@exsilondata.com.
16 8D Problem Solving Seminar. Au-
burn Hills, MI. Call Sybeq at 248-233-6104
or e-mail info@sybeq.com.
21-25 Coordinate Metrology Sys-
tems Conference. Concord, NC. Call Jan
Rode at the Coordinate Metrology Society at
425-802-5720 or e-mail registrar@cmsc.org.
SAVE THE DATE
If youd like your event included in QP Calendar, submit information at least three
months in advance to vfunk@asq.org. Non-ASQ organizations may list one event
per issue.
82 QP www.qualityprogress.com
22-24 Principles and Practices
of Process Improvement. Denver. Call
the International Society for Performance
Improvement at 301-587-8570 or e-mail
register@ispi.org.
23-24 Supply-Chain World Asia Pa-
cic conference. Singapore. E-mail Shelly
Koh at skoh@supply-chain.org or visit www.
supply-chain.org.
24-25 Product Safety and Liability
Conference. Milwaukee. Call Randall Good-
den at 262-594-5198 or e-mail rgoodden@
ipslp.com.
28-29 Planning and Leading Lean
Kaizen Events. Atlanta. Call Stat-A-Matrix
at 800-472-6477 or visit www.statamatrix.
com.
31 Lean Purchasing Seminar. Man-
chester, NH. Call the Assn. for Manufactur-
ing Excellence at 224-232-5980 or e-mail
dcrawford@ame.org.
AUGUST
4-8 Transactional Design for Six Sig-
ma. Denver. Call the International Society
of Six Sigma Professionals at 480-368-7083
or e-mail info@isssp.com.
18-19 ASQ Education Course. Lean
for Service. Addison, TX.
18-20 ASQ Education Course. Root
Cause Analysis. Addison, TX.
18-21 ASQ Education Course. Guide
to Process Improvement and Change.
Addison, TX.
18-22 ASQ Education Course. Auditor/
Lead Auditor Training for ISO 13485. Ad-
dison, TX.
18-22 ASQ Education Course. ISO
9001:2000 Lead Auditor Training.
Addison, TX.
18-22 ASQ Education Course. Lead
Auditor Training for AS9100. Addison, TX.
18-22 ASQ Education Course. Design
for Six Sigma. Addison, TX.
18-22 ASQ Education Course.
Introduction to Quality Management
Revised. Addison, TX.
18-22 ASQ Education Course. Soft-
ware Quality Engineering. Addison, TX.
20-22 ASQ Education Course. Imple-
menting Statistical Process Control.
Addison, TX.
18-22 ASQ Education Course. Imple-
menting and Auditing an ISO 9000:2000
Quality System. Addison, TX.
21-22 ASQ Education Course.
Customer-Supplier PartnershipsAn
Introduction. Addison, TX.
SEPTEMBER
16-19 Lean Six Sigma Improvement
Week. Chicago. Call the International Quality
and Productivity Center at 800-822-8684 or
e-mail enquire@icpc.co.uk.
 For information on placing an ad,
contact Ramona Garcia at 800-248-1946,

PROFESSIONALSERVICES or 414-272-8575, or e-mail rgarcia@asq.org

What does earning a degree mean to you?

Career advancement? Greater opportunities? Personal development?
With distance learning, all these things are within your reach.
California National University for Advanced Studies (CNU) is an accredited distance-learning university designed for technical and business
professionals. The CNU College of Quality and Engineering Management includes a Bachelor of Quality Assurance Science and a Master of
Engineering Management. Combining distance learning with one-on-one instruction, CNU degree programs provide maximum flexibility for maximum
growth. The Bachelor of Quality Assurance Science prepares the individual to be able to manage, plan, procure, design, and maintain an effective
Quality Assurance Program within a company. The Master of Engineering Management provides the individual with the breadth of knowledge required
for succesful project management, and with the ability to integrate knowledge from diverse disciplines.
CNU also offers seven other degree programs and courses by extension for those who prefer to begin their continued education gradually.

Let CNU help you take it to the next level.

Contact one of our admissions representatives at 800-782-2422 or e-mail cnuadms@mail.cnuas.edu.

Multi-Vari Chart and

Analysis Book
- A Pre-Experimentation
Classroom Training Aids Technique
Quincunx Boards
Sampling Bowls www.mpcps.com
Catapults
Deming Funnels
And lots more James Highlands,
Visit us at: www.qualitytng.com RLA, CQA
Email sales@qualitytng.com for brochure MSA Audits-Training Consulting
Management Reviews
Ph: 248-641-7030 Fax 248-641-7031
PO Box 611 Troy, MI 48099-0611 Quality Program Development
ISO 9000, ISO 14000, NQA-1, ASME III,
Quality Technology Company jhighl@aol.com
Achieving Dramatic Performance Improvement www.ManagementSystemsAnalysis.com
TRAINING & CONSULTING
Corporate Performance Improvement
CMMAdviceFree.com Management & Quality Consultants
Six Sigma
P.O. Box 136 Royersford, PA 19468
Team-Based Problem Solving Ph. 610-409-0168 Fax 610-409-0167
Management Systems (ISO/QS/TS)

Best Seller in ASQs Six Sigma Beef Up Your Ineffective

Forum Magazine
Quality Management System!
To purchase visit: www.amazon.com Advanced Internal Auditor Training
Reduce QMS Documentation
Call: 847-884-1900 (or) 800-680-0700 Effectively
Visit us at: www.qtcom.com Integration of Lean Practices
into QMS
Effective Root Cause Analysis
McRINK FAR EAST SERVICES Tough Audits to Challenge Past

QualityQuest, Inc.
Downsizing inhouse Q.A.? Michael J. Micklewright, CSSBB, CQMgr, CQA, CQE
Arlington Heights, Illinois
Changing Suppliers? PH: 847-870-0822, Fax: 847-870-0872
mike@mikemick.com; www.mikemick.com
Can you afford to forgo Inspection
outsourcing? DBS QUALITY MANAGEMENT
Final Random Inspection, INTERNATIONAL
Vendor Assessments, etc. 894 Carriage Way, Lansdale, PA 19446
215-368-6266, Fax 215-393-4873
Established for 40 years RAB, IRCA, AIAG, & AAQG Certified Staff
in the Far East. ISO9000-QS9000-TS16949-AS9100
We offer fast, efficient, and Professional Auditing Training Services:
professional low cost services Services & Assessment:
Internal Auditing
Internal Auditing
Quality Systems -
Pre-certification ISO/Aerospace/Automotive
in the Far East Gap Assessments Publications:
Contact us in Hong Kong at Supplier Audits Documentation Packages
Process Audits CD Training Packages
e-mail: mcrinkhk@netvigator.com Conformance Audits Consulting Services
Web site: www.mcrink.com Registration Procedure Writing Services
Tel: 852-2389-3770 Leading Companies to Certification Since 1991
Website @ www.dbsqualitymgmt.com
Fax: 852-2357-4770 E-mail: mikedocrt@aol.com

June 2008 QP 83
 For information on placing an ad,
contact Ramona Garcia at 800-248-1946,

PROFESSIONALSERVICES or 414-272-8575, or e-mail rgarcia@asq.org

Greenville, SC 29609
864-271-1988 Fax 864-271-1988
ISO9001-TS16949-QS9000-ISO14001
Professional Services Training Services
Internal Auditing Process Audits
Gap Analysis Quality Systems
Pre-registration Consulting Services
Root Cause Analysis Process Improvements
Supplier Audits Systems Development
Proven track record
billhoellrich@aol.com | www.Hoellrich.org

Quality Manufacturing Dr. Deming Inspiration Speeches

QMA Associates
Dr. Deming Impersonation
What Would Deming Say
*Leadership Effectiveness Training of Todays Business World?
*Six Sigma/Lean/TQM coaching and training Deming Evangelist
*In-house and public Green/Yellow Belt Preaching the Word of Deming
from the Out of the Crisis Bible Future
Training
R U a Member of SAD (Society of Anti-Deming)?
*Paperless Quality System Software Funny, sarcastic look at the anti-14 Points
*Spanish/English Bilingual training Keynotes, Corporate Events, ASQ Meetings, Conferences
*Assistance with Latin American Manufacturing
Call: 1-954-856-8810 or QualityQuest, Inc.
E-mail: emb109@aol.com Michael J. Micklewright, CSSBB, CQMgr, CQA, CQE
Enrique Bekerman, MS, MBA, Arlington Heights, Illinois
ASQ CQMgr, ASQ CSSBB PH: 847-870-0822, Fax: 847-870-0872
www.webuildqualitytogether.com mike@mikemick.com; www.mikemick.com

Uniquely qualified to assist you

ASTON TECHNICAL CONSULTING SERVICES For all your needs regarding India Gauge R&R Studies
TEAM 2000 Book
- For Destructive &
www.hownwhy.com
Phone: 1-877-HOWNWHY
Non-Destructive Testing
Rated in the Top 10% by Customers www.mpcps.com
SERVING OIL AND GAS, AEROSPACE
AND MANUFACTURING INDUSTRIES ACCREDITED HUMIDITY CALIBRATION
Quality Management Systems IS O / IE C 17 02 5
(ISO 9001:2000, API Q1 and ASME)
Development & Implementation NIST TRACEABLE STANDARDS
QMS Manual & Procedures
Onsite Staff Training or Orientation Relative Humidity.................. 0.5% to 99%
External & Internal QMS Auditing Services Dew / Frost Point.............. -90 C to 70 C
Inspection Services Volume Ratio, V (PPM)........ 0.1 to 250,000
E&P Project Inspection Support
Supplier Evaluations/Assessments
Source Inspections Thunder
Operations Inspections Scientific
ASQ/RABQSA Certified Auditors and Engineers Corporation Lab Code 200582-0
AWS/ASNT Certified Inspectors
Bill Aston, Principal & Managing Director 623 Wyoming S.E., Albuquerque NM 87123
1-800-872-7728
Office: (936) 653-5257 or (877) 653-5257 www.thunderscientific.com
E-mail: quality@astontechconsult.com
Web site: www.astontechconsult.com sales@thunderscientific.com

Trouble in China?
Give us a call.
Six ASQ certifications at your service.
Shanghai-American Quality Services
Michael Mead, Ph.D., President
Michigan China
269-944-1914 86-769-22425547

84 QP www.qualityprogress.com

PROFESSIONALSERVICES
Six Sigma Training + Cert.
Business Process Mapping
Process Characterization
Design of Experiments
Machine & Process Capability
TM
Advanced Systems FMEA, 8D, Corrective Action
Consultants Minitab Training
480-423-0081
info@mpcps.com MARIO PEREZ-WILSON
www.mpcps.com - www.ascsixsigma.com
Build Quality By Choice
Not By Chance
TEAM 2000
www.hownwhy.com
Phone: 1-877-HOWNWHY
Rated in the Top 10% by Customers
TQM ASSOCIATES INC.
WERE HERE TO SUPPORT YOU
Established in 1994
Women Owned
Quality Assurance Professionals
Across the U.S. and Worldwide
Temporary or Permanent
Source Inspection
Surveys
Audits
Expediting
(800) 424-4729 3990 Old Town Ave. #C109
Fax (619) 297-3251 San Diego, CA 92110
e-mail: stephk@tqminc.net
SAE AS9100/AS9110/AS9120, ISO 9001,
and ISO 14001 Consulting and Training
Documenting your quality or environmental
management system
Mapping and improving your
management system
Improving your communication processes
Gap Analysis
Internal Audits
Supplier Audits
Training needs, including root cause analysis
Selecting an Accredited Registrar
QUAL-TECH, INC.
Certified QMS and Aerospace Auditors
Ph: 321-453-0637, Fax: 321-453-5842
contact@qual-techinc.com | www.qual-techinc.com
EASTERN MICHIGAN UNIVERSITY
MS Degree in Quality Management
Join other professionals in an online MS
degree program in Quality Management.
Our students are typically in mid-career with
quality-related responsibilities in manufacturing
and various service industries.
The 30 semester hour curriculum consists
of a 10-hour core, specialization courses, and
a research project/thesis.
Program applicants must have a BS/
BA degree with a GPA of 2.50. Get more
information at: http://cot.emich.edu/quality, and
http://www.emuonline.edu/, or contact Dr. Tom
Soyster at: tsoyster@emich.edu, 734-487-2040.
For information on placing an ad,
contact Ramona Garcia at 800-248-1946,
or 414-272-8575, or e-mail rgarcia@asq.org
Q.A. SERVICES
IN THE FAR EAST
Can you afford to buy products
out of Hong Kong, China (PRC),
and the Far East
without some assurance they
meet your specifications and
quality requirements?
Our services are fast, efficient,
definitely affordable and reduce
your in-house Q.A. costs.
McRink Surveyors Company Limited
Source Inspection: PPAP: Manufacturing 1-2, 2/F., Hung Tai Industrial Bldg.,
Audits: Supplier Qualification: Process 37-39 Hung To Road, Kwun Tong,
Audits: Product Technical Evaluation: Kowloon, Hong Kong.
Supplier Control Management Tel No: 852-2389-3770
Fax No: 852-2357-4770
E-mail: mcrinkhk@netvigator.com
Web site: www.mcrink.com
CHINA Suppliers Inspection
Shenzhen: Guangzhou: Dongguan: Zhuhai:
Xiamen: Shanghai: Suzhou: Nanjing: Ningbo:
Tianjian: Qingtao: Xi-an: Chongqing: Dalian: Serving the Carolinas, Virginia
and East Tennessee
www.amrepchina.com
USA: 1 800 692 8539 ISO & TS Implementations
E-mail: gochina@amrepchina.com Auditor Training/Coaching
Continuous Improvement
Root Cause Analysis
Kimberly Maggie
Documentation Development
President and CEO
LEAN Culture Development Internal Audits
Lean Culture Training Statistical Process Control
Certified by RABQSA, ASQ & Plexus
Lean Assessments and Action Plans
Inspiring Lean Office & Mfg. Office: 828-337-4857
Simulations john@jcarpenterandassociates.com
www.jcarpenterandassociates.com
Integration into ISO QMS
Training/Facilitating in all Tools
Present
Five Day Kaizen/VSM Events
Dimensions
QualityQuest, Inc. Consulting Services
Michael J. Micklewright, CSSBB, CQMgr, CQA, CQE
Arlington Heights, Illinois Quality Systems Management
PH: 847-870-0822, Fax: 847-870-0872 Roy G. RoperConsultant
mike@mikemick.com; www.mikemick.com
Six Sigma Design of Experiments
Data Analysis Statistical Process
Process Tek - Sterility by Design Control
For sterile products, packages & processes
P.O. Box 342, Ellenton, FL 34222
Kailash S. Purohit, Ph. D. T 941.776.8369, F 941.776.8431
www.processtek.net e-mail: RRoper3@tampabay.rr.com
kaipurohit@processtek.net www.DimensionsConsultingServices.com
June 2008 QP 85
 For information on placing an ad,
contact Ramona Garcia at 800-248-1946,

PROFESSIONALSERVICES or 414-272-8575, or e-mail rgarcia@asq.org

UHRIG CONSULTING 200+ Custom / Std. Workshops

Including Lean and Six Sigma
Redondo Beach, CA
310.798.8442 TEAM 2000
lisa@uhrigconsulting.com www.hownwhy.com
Phone: 1-877-HOWNWHY
www.uhrigconsulting.com
Rated in the Top 10% by Customers
Process Mapping/Improvement,
Training, Auditing, Documentation Amazon Consulting, INc.
Development & Facilitation: www.amazonconsultinginc.com

ISO 9001 TL9000 ISO 9001

AS9100/9110/9120 6Sigma Tools Online Auditing
for Global
ISO13485/21CFR820 CAPA & Error Companies
TS16949 Proofing
208-853-6988 Corvallis, OR

Machine/Process
Capability Study Book
- A Methodology For
Optimizing Processes
www.mpcps.com

Business Process
Improvement Services
Business Process Excellence
Consumer Product Safety
Development of Quality System
Factory Evaluation & Inspection
Written Guaranteed Supplier Quality Management
Statistical Process Control
Save up to 80%
Training/Certification/Degrees
Certification Testing & Training

Consulting & Training
100% Success Rate as little as 60-90 Days
Online/Onsite/Offsite/Customize
All STANDARDS
ISO9001, AS9100, ISO14001, TL9000, FDA, CGMP, TickIT,
GMP, TS16949, ISO13485, HACCP, ISO17025, Z540-1, ISO20000,
ISO27001, OSHAS18001, Six Sigma, ISO22000, Sarbanes-Oxley,
ROHS, CMMI, Sony Green, Lean Manufacturing, CMDCAS, etc.
709 Lyndsie Dr., Coppell, TX 75019 Via Online Quality University
P: 972-315-2695 All Standrds/Regulations
mehtaconsulting@gmail.com
Degrees, Certificate, Internal Auditor, Overview,
www.mehtaconsulting.net
Implementation & Executive Review
We also offer certificates in Quality Environment,
Health & Safety, OSHA, Six sigma and others.

AccelCert! Quality University

FOR PROFESSIONAL STUDIES
1-888-476-9103
www.qualityuniversity.com
www.accelcert.com
support@qualityuniversity.com
www.qualityuniversity.com

86 QP www.qualityprogress.com
 For information on placing an ad,
contact Ramona Garcia at 800-248-1946,

PROFESSIONALSERVICES or 414-272-8575, or e-mail rgarcia@asq.org

The most innovative Resources for Quality

& Productivity Improvement
and valuable consulting Books, Videos, CDs, DVDs,
tool Ive ever used. Software, Calibration Lables,
ASQ Certification Preparatory Materials,
Jean Kulig, Intelicorp Tags, Computer Based Training,
Implementation Kits, Learning Packages

Service Education Healthcare

Government Manufacturing
TM Customer Relationship Mgt.,
Product Development, 5 S/Visual Factory,
Lean Enterprise, Team building, Six Sigma,
For information ISO 9000, Statistics, & Much More
on placing an ad,
www.QualityCoach.Net
contact Ramona Garcia
1-800-648-9510
at 800-248-1946
www.ONTHESYSTEM.com or 414-272-8575,
or e-mail rgarcia@asq.org.
Free Consultant Resources:
onthesystem.com/advisors

TraKDocument
from VSV
CLASSIFIEDADS
Business for sale: Business for sale:
TraKDocument is a document control management
system software. It emphasizes workflow and audit Established since 1984 quality consulting Small Metrology Lab
and ecommerce business for sale. in Mooresville, NC.
management. It compares two revisions of text
documents with extensions txt, rtf, doc, pdf. Log on to qcss.com for details. Owner disabled,
Contact qcss@qcss.com Was ISO accredited
Visit www.vsv.com.vn today and find out more. or call: 949-388-7686. Contact CharlieSaunders@winstream.net

CALL FOR ARTICLES

Write an article for the top publication in the quality eld. Go to www.qualityprogress.com.
Click on Author Guidelines at the bottom of the page and youll nd QPs 2008 editorial
planner. There are also submission guidelines and contact information. Here are some of
the topics QP plans to feature in the coming months:

September October November

Innovation & Technology Standards & Auditing Quality Culture/Change
Management

If the topic you would like to write about does not appear in the editorial planner, please
do not let that dissuade you from submitting your article. We will use all accepted articles,
whether they align with a cover theme or not. www.qualityprogress.com

June 2008 QP 87
 ONE GOOD IDEA
High Probability of Success
Use Excel to determine if data are normally distributed
SHAPE, LOCATION and spread are
three attributes of a distribution that
need to be determined when working
with any data. The rst attributeshape
or type of distributionis needed to
know which statistics are appropriate to
use. Some of the distributions used for
quality purposes include normal, bino-
mial, Poisson, Weibull, exponential and
hypergeometric.
Normal is the most frequently occurring
type of distribution, but to use the statistics
relating to normal distribution, it is neces-
sary to prove the data are, in fact, normally
distributed. There are several ways of
conrming that fact, including:
Comparing the data to the empirical
distribution of the data using stan-
dard deviations and data count. Plus
or minus one standard deviation is
approximately 68% of the data. Plus
or minus two standard deviations is
approximately 95.5% of the data. Plus
Scatter plot to determine
normal distribution / FIGURE 1
Daily high temperatures
100
90
Temp-F
80
70
60
-3.00 -2.50 -2.00 -1.50 -1.00 -0.50 0.00 0.50 1.00 1.50 2.00 2.50 3.00
Standard deviation (z)
THE PLOT THICKENS
Is Excel the easiest way to create a normal probability plot? Do you have other
tricks of the trade? Visit the QP website at www.qualityprogress.com and post
them in the columns comment section.
88 QP www.qualityprogress.com
BY BART HAMILTON
or minus three standard deviations is
approximately 99.7% of the data.
Showing the location of the median and
mean to be within one-half standard
deviation.
Analyzing the distribution of the data and
comparing it to chi-square distribution.
Creating a normal probability plot of
the data and comparing the result to
a straight line. If a straight line is pro-
duced, the data are normally distributed.
In the case of the nal method, the prob-
ability plot is a scatter plot of a calculated
plotting position for each data item based
on its rank position in the data set and the
associated data value. Plotting is done on
specialized graph paper, a sample of which
is available at www.weibull.com.
One of the axes, usually the horizontal
one, is scaled to the normal distribution.
Plotting positions are assigned a place on
this axis. The vertical axis is linear and is
used to plot data values. The scatter plot
results from the intersec-
tion of the plotting posi-
tions and data values.
The difculty in plotting
arises when a computer
is used to create the scat-
ter plot. Many graphing
programs do not have
an option for creating a
normal probability plot.
This can be overcome by
using a transformed value
for the plotting position.
The transformation is
performed by converting the calculated
plotting position to a z-value, which is
obtained using the Excel function NORM-
SINV. Then, a linear scatter plot can be
constructed by plotting the data values
versus the z-value.
The procedure for creating a normal
probability plot is:
1. Sort the data in ascending order.
2. Assign ranks to the data. The minimum
data value is assigned a rank of 1.
3. Calculate a plotting position using the
formula PP = (I - 0.5)/n in which I is the
rank and n is the number of data values.
4. Convert the plotting position to z-values
using the Excel function NORMSINV.
5. Construct a scatter plot using the z-values
as the horizontal axis and the sorted data
value as the vertical axis.
6. Check the scatter plot to see if a straight
line can represent it. If it can, then the
data are normal.
An example of those efforts can be
seen in Figure 1, and the data used to cre-
ate the plot can be found in an online-only
table at www.qualityprogress.com.
The plot is made using the sorted data
column and the standard deviation in a
scatter plot. The conversion of the plot po-
sition to standard deviations permits the
use of a linear axis for each variable. Thus
we see how, with a modicum of effort, Ex-
cel can be used to produce a probability
plot for determining the normality of data.
BIBLIOGRAPHY
Dale H. Bestereld, Quality Control, seventh edition, Pearson
Prentice Hall, 2003, pp. 161-163.
Nancy R. Tague, The Quality Toolbox, second edition, Quality
Press, 2005, pp. 365-370.
BART HAMILTON is an instructor in the University of Akrons
engineering and science technology department. He retired
from Uniroyal Goodrich Tire Co. after more than 35 years in tire
research and development. Hamilton earned a masters degree
in industrial management from the University of Akron and is
ASQ certied as a quality engineer and quality manager.
2009 World Conference on Quality and Improvement

Call for Papers

World Conference on Quality and Improvement

May 18 20, 2009 Minneapolis, MN

The Technical Program Committee of the 63rd World Conference on Quality and Improvement invites you to
submit proposals for presentation in Minneapolis. Focus areas for the 2009 conference include quality basics;
customer service; driving quality throughout the organization; improved performance; and 21st century needs,
competencies, and issues.

The deadline for submission is August 1, 2008.

All forms must be submitted online at http://wcqi.asq.org.

Please visit this site to learn more about the call for papers submission process.

Thank you for your support of ASQ and interest in the World Conference on Quality and Improvement.

Note: The World Conference on Quality and Improvement is not able to provide any support to selected primary speakers
beyond a complimentary registration. There is no provision for expenses, honorarium, or other monetary support.
Business Benefits www.statsoft.com
! Takes six sigma/process improvement to
the next level by delivering comprehensive
analysses and re
eports to targete
ed user groups
! Empowers data-driven decision making by
delive
ering the data and analysiis shortcuts to
th
he knoowledgee workers who need them

User Benefits
STATISTICA Enterprise delivers:
! Streamlined access to databases and
other data sources
! Single-click analysis shortcuts, targeted
to specific groups of users
! Auto-updating analyses
! Comprehensive graphics for data
exploration and understanding
Version 8
Now Available
Downloadable Demo
Technical Benefits
! Fully web-enabled architecture solution
! Integrated windows authentication Compliance Benefits
! Central configuration/management of: ! 21 CFR Part 11 compliant system
! Users/groups/permissions configurations

! World-wide data access ! Integrated document management

system for storage of related
! Analyses templates
docuumennts and records (Sarbanes
s-
! Reports templates Oxleyy Compliance)
! Integrated, cutting-edge process intelligence
StatSoft offers a large selection of desktop
(predictive modeling, process optimization),
and web-enabled, enterprise-wide solutions
and MSPC C options for manufaacturing and other applications
supported with installation and training
services by a network of 24 full-service offices
on all continents.

2300 E. 14th St. Tulsa, OK 74104 USA (918) 749-1119 Fax: (918) 749-2217 info@statsoft.com www.statsoft.com
Australia: StatSoft Pacific Pty Ltd. France: StatSoft France Italy: StatSoft Italia srl Poland: StatSoft Polska Sp. z o. o. S. Africa: StatSoft S. Africa
a (Pty) Ltd.
Brazil: StatSoft Brazil Ltda. Germmany: StatSoft GmbH Japan: StatSoft Japan Inc. Portugal: StattSoft Iberica Ltda Sweden: StatSoft Scandinavia AB
Bulgaria: StatSoft Bulgaria Ltd
d. Hungary: StatSoft Hungary Ltd. Korea: StatSoft Korea Russia: StatSoft Russia Taiwan: StatSoft Taiwan
Czech Republic: StatSoft Czech Rep. s.r.o. India: StatSoft India Pvt. Ltd. Netherlands s: StatSoft Benelux BV Spain: StatSoft Iberica Ltda UK: StatSoft Ltd.
China: StatSoft China Israel: StatSoft Israel Ltd. Norway: StatSoft Norway AS
StatSoft, STATISTICA, and MultiStream are trademarks of StatSoft, Inc. Copyright StatSoft, Inc. 1984 - 2008