|
|
 |
 |
|
|
Posted July 30, 2001
Sick SigmaBy Jonathan Walters
It’s been a while since the management consulting world has foisted a really goofy trend on the public sector — probably “management by objectives” was the last real corker (yes, we could argue about TQM, but let’s not). You know the type of trend, though: Invented on some Midwest assembly line that chunks out food mixers, they’re then supposed to be immediately transferrable to some obviously similar public-sector function, such as moving families off of TANF and into self-sufficiency or rehabilitating chronic juvenile offenders.
Actually, I’ve been waiting for “Six Sigma” to rear its trendy little head for about four years — that’s how long ago I first heard of it. It had such a catchy ring to it, I figured it would have to show up sooner or later as the newest in mind-numbing management panaceas. And darn if I didn’t read in a recent issue of U.S. Mayor, the monthly publication of the U.S. Conference of Mayors, that a large Midwestern city that will remain nameless — at least here — had adopted the practice.
For those you who are hopelessly out of it when it comes to management cool, an explanation: “Sigma,” apparently, is some scale that represents deviation from perfection, “One” through “Six” being the perfection continuum. So “One Sigma” would be, say, a 58-car pileup on a fog-bound L.A. freeway. “Two Sigma” would be my office. “Three Sigma” would be the Boston Red Sox pitching staff, etc., until you get to “Six Sigma,” which conjures up images of robots flawlessly shaping McDonald’s hamburgers into perfect circles exactly one quarter inch thick and whirling them, Frisbee-like, into the shipping cart without ever bouncing one off the ceiling, or something like that.
Just to add to the Eastern Mysticism of the practice (and to make sure that consultants can rack up some serious billable hours), Six Sigma comes with attachments — that is, various levels within the Sigmas — so you can be a “black belt” Six Sigma ... ninja, I guess.
Anyway, it’s obvious that these Six Sigma consultants have too much perfection in their lives already. Otherwise, how would they find the time to try to sell these mystical management doodads to the poor schmoos in government who are just trying to hold the enterprise together on a day-to-day basis?
So, for those career managers in that nameless Midwest city who are on the receiving end of this one, please accept my sincere sympathy. But remember, the sillier the name, the faster they disappear. In fact, this one will be gone in about Six Seconds, or by the next election, whichever comes first.
Jonathan Walters is a staff correspondent for Governing and author of Measuring Up: Governing’s Guide to Performance Measurement for Geniuses (and Other Public Managers).
Agree? Disagree? Want to expand on the discussion? E-mail us at mailbox@governing.com, and we'll post your comments here. Please include your name, location, government or business title or job description, and a daytime phone number (for verification purposes).
Readers’ Responses:
Jonathan Walters may be a guru on measurement (I do recommend his book on the subject), but I was surprised at his article. Six sigma is merely a statistical measurement of consistency. It refers to an error rate six standard deviations from the mean on a given distribution curve. It is used in common jargon to represent a very low error rate. Granted, managers can fixate on a single measurement to the neglect of everything else, but I'd rather have a manager focused on a goal than have no error rate measurement, let alone a method to improve it.
The oxymoron of "public management" is that many who are in management positions in government have had little or no management training. In contrast to the business schools, most public management training courses teach precious little about managing and provide virtually no mention of statistical management techniques.
While it is easy to dismiss enthusiastic attempts to improve service and efficiency as merely the "fad of the month," I am reminded of the Peter Drucker quotation that nothing of substance in an organization gets accomplished "...without a monomaniac with a mission." The six sigma concept is nothing more or less than a useful arrow in the management quiver.
Roger D. Cross
I tend to agree with Jonathan Walters, though he's probably being deliberately inflammatory. There's very little in the Six Sigma methodology that didn't already exist in the classic Deming and Juran approaches to quality. Here's what I see as being new(ish):
1) In Six Sigma, the pendulum has swung back away from the quality circle concept, in which everyone was responsible for improving quality in their work centers, and everyone was expected to be well-versed in quality methods. Instead, we have Six Sigma Black Belts, who receive fairly rigorous training in statistical methods and other tools, and whose job it is to be roving troubleshooters. This difference has been watered down somewhat, with the appearance of shorter courses for green belts, etc. The material in the courses is very similar to the quality tools that have been around since the 1950's. I tend to think that there's a balance needed: it's important for everyone to know how to think clearly, monitor data, and make good decisions. It's also important for there to be some experts around who can do more advanced analysis when it is indicated.
2) There's a new emphasis on breakthrough improvements, though this may inherit from the reengineering/breakthrough thinking fad of the mid-90's. Six Sigma teams are generally gunning for big dollar improvements, not just incremental ones. However, if you have 10,000 employees all engaged in making incremental improvements, it wouldn't need each one to be big for it to have a strong impact on the bottom line.
3) There's a new emphasis on making almost no parts or outcomes outside of spec limits. This is a useful idea--if you can reduce the variation in a process so that the variation is much less than that allowed by spec limits, you make very few bad parts, and that saves you money all the way down the line. However, a lot of processes don't lend themselves to the sorts of precise high-volume measurements required for robust six sigma calculations, particularly in the service sector.
The tools of quality work very well, and have done so for many years. The movements, such as TQM, reengineering, and Six Sigma all use the same basic concepts and tools. You can spend a lot of money learning them, or you can read a good book for $20. If you use the tools well, they'll pay off handsomely. If you don't use them, they won't. If the Six Sigma movement helps people get a new interest in the tried and true quality methods, that's a good thing. If it just creates a lot of expectations, and expenses, and people don't or can't use what they learn, it will create new skepticism.
Steve David
While uninspired to join the “six sigma” debate, I was struck by Roger Cross’ comments about “the oxymoron of ‘public management’” and the lack of training in management and statistical analysis by schools of public administration.
As in all fields, certain universities tend to focus on different aspects of public administration. The academically oriented programs may well be light on nuts and bolts, but you won’t find a dearth of such requirements in programs that are practitioner oriented. Granted, there are many people in public sector positions who are trained in technical fields rather than management, but the fault for any lack of additional education lies with those who hire them, not with those trying to do an increasingly difficult job with consistently less support.
Christine D. Wilson
The whole concept of sixth sigma in government is flawed by mathematical reality, to wit:
Sigma measurements (standard deviations) can only be useful in measurement of outputs which are, themselves, subject to the possibility of accurate measurements. Very accurate measurements. Also, whatever it is being measured must be subject to the laws of probability in a continuously varying function subject to the so-called normal distribution. Such measurement is useful in the production of drive shafts for V-4 engines, but a little more difficult to make workable when measuring customer satisfaction in the treasurer's office. And even less workable and useful when nobody really understands what they are talking about. And many customers would be ecstatically happy with the third sigma, if it applies, when it applies.
Government is actually more subject to queue theory...check out the waiting line at DMV. Or the laws of temporarily exponential functions...check your property taxes. Not to mention several other possibilities...such as Poisson.
I maintain that every proposer of a new “management paradigm” should have his or her idea certified by a competent graduate level statistician, a mathematician, an engineer, a physicist, a psychologist, a priest, a high school English teacher, and a Mom.
The statistician is to eliminate false premises related to statistics, the mathematician to eliminate fundamental errors in mathematical logic, the engineer to determine whether or not the so called paradigm can actually be applied, the physicist to eliminate errors in physical reality, the psychologist to eliminate solutions that are unworkable in the field of human endeavor, the priest to rule on morality and say a prayer for the proposer, the English teacher to correct errors of organization, logic, spelling, and grammar in the proposal, and the Mom to tell the proposer to get a real job.
James D. Eagle
Recently in View:
President Pre-emption: why George W. Bush doesn't trust state and local government (posted July 22, 2001)
Kilowatts and the Price of Turkey: why electricity deregulation is failing (posted July 16, 2001)
Down at the DMV: making the least of citizen interaction (posted July 9, 2001)
River Dance: to dredge or not to dredge (posted July 2, 2001)
Crime Among the Kiwis: where gun violence is abhorrent (posted July 2, 2001)
CAVE People: pontificators and public input (posted June 18, 2001)
Complete index of previous columns
|
|
