Quality Turf Wars
 CE: Medical Devices
 Excel Data Analysis
 Vision Systems
 Vision Buyers Guide
 First Word
 News Digest
 Quality Mgmt.
 One Minute Mgr.
 Quality Standards
 SPC Toolkit
 Book Reviews
 Quality Software
Picture Picture
 Quality Management

Statistical Standards
and Quality

The creation of international statistical
standards has not kept up with demand.

by A. Blanton Godfrey

Since their publication in 1987, the ISO 9000 series of standards have become the most widely used standards ever published by the International Organization for Standardization. The publication of these standards coincided with an intensive, worldwide focus on total quality management as a means of preparing companies for the rapidly expanding globalization of international markets.

During this time, the use of statistical methods in companies also grew dramatically. New national and international statistical standards are needed, but the creation of these standards has not kept up with demand. Even those that have been published are relatively unknown and almost totally unused in industry.

New statistical standards are critically needed in several areas. These areas include supporting methods for developing a balanced scorecard; providing a means for assessments, reviews and audits; and developing a quality control system for management information systems.

One of the most important areas in quality management today involves measuring areas other than just financial performance. Many organizations now realize the critical need for measuring internal and external quality, customer satisfaction and loyalty, environmental performance, supplier quality and human resource management. By creating balanced scorecards, they can more easily measure these "intangible" processes.

A number of interesting statistical problems accompany the construction of these balanced scorecards. The first -- and most fundamental -- is the construction of the primary measurements. People still need guidance on when to use attributes measurements or variables measurements.

The second problem facing many of these organizations concerns combining multiple measurements of different quality characteristics into useful summary statistics. When the characteristics are unequal and should be weighted differently, the problem becomes even more challenging. The third major challenge involves combining measurements of many different products or services into summary measures for business unit heads or corporate officers without masking important information. The measurement plan often is developed by starting at working group levels and deciding which quality, cost and process measurements are important, then combining these at plant, service site,  division, business unit and corporate levels. These measurements often miss critical factors important to external customers and fail to give accurate pictures of costs and process quality across departments and functions.

     The move toward TQM in organizations worldwide has been greatly aided by the development of national and international quality award programs.

     By far the most important use of these award criteria is to perform self-assessments. These self-assessments usually involve numerous highly trained examiners from many different business units and functions performing in-depth reviews. Even within the same company, these examiners face the challenge of understanding the various measurement plans used throughout the company.

When these reviews include external companies such as suppliers, alliance partners or others, the challenge becomes even greater. The examiners may not understand the measurements used, the construction of summary statistics and what each of these means. Even harder for the examiners and auditors is identifying which critical measurements are not being made.

Companies are just beginning to explore information quality. Many organizations report quality, financial and even environmental results with little knowledge of the data's accuracy, the summary statistics or even whether the numbers are meaningful. Although companies have done some preliminary work in this area in the past few years, most still have no information quality management plan and don't even use rudimentary estimates of the accuracy or precision of the numbers upon which they routinely base critical decisions.

The standards organizations must understand what good methods companies now use, what best practices exist and where companies believe they have serious needs for standardized methods. Rather than creating new methods, their first priorities should be to capture what their customers truly need. International standards should reflect the ideas of a wide range of experts and practitioners. These working groups of experts should limit their efforts to clearly documenting the best methods while calling on the leaders in each area from around the world for input articles and books, reviews and discussion.


About the author

A. Blanton Godfrey is chairman and CEO of Juran Institute Inc. at 11 River Road, Wilton, CT  06897.

© 1997 Juran Institute. For permission to reprint, contact Godfrey at fax            (203) 834-9891 or e-mail agodfrey@qualitydigest.com.


e-mail Quality Digest

Copyright 1997 QCI International. All rights reserved. Quality Digest can be reached by phone at (916) 893-4095.

Please contact our Webmaster with questions or comments.


[Current Issue]

[ISO 9000 Database]

[Daily News]

[Phil's Journal]

[Quality Leadership]

[Past Issues]

[Quality Web]


[Media Kit]

[Classified Ads]