Content By Davis Balestracci

Davis Balestracci’s picture

By: Davis Balestracci

W. Edwards Deming has been dead for almost 16 years. In my opinion, he and Joseph Juran were the true quality giants of the 20th century. No one seems to talk about Deming much any more except to relate stories from the past and pine for the “good ol’ Deming 4-day seminar days.” (I don’t. Nor do I think he would want us to.) He also seems to have a fundamentalist cult following who revel in smug running commentary, quoting chapter and verse a la Deming on any current quality effort.

I’m going to say a couple of things that might not make me popular. However, I’m also going to preface these remarks with my utmost awe and thanks for the extraordinary influence Deming has had on my career, including a warm hand-written personal note of encouragement at the low point of my career 25 years ago. I feel that all the current improvement manifestations come directly from Deming theory—with a healthy dose of Juran’s practical wisdom—but they are missing the point, which is what he tried to put in context at the end of his life with his system of profound knowledge. (See my column “TQM, Six Sigma, Lean and…Data?” www.qualitydigest.com/july06/departments/spc_guide.shtml.)

Davis Balestracci’s picture

By: Davis Balestracci

Do top managers still view financial performance as the sole indicator of success, despite mouthing platitudes about dazzled customers and fulfilled employees? Is there a point when reductions are done excessively in the name of squeezing out a few more percentage points of profit, moving companies from their “ideal weight” to a state of near-anorexia?

Shouldn’t success factors include happy customers, more motivated and committed workers, investment in communities, and concern for the environment? Is it possible to create a company whose objectives are worth sacrifice by those who work in it and by the society it serves?

Focusing solely on the numbers must be replaced by a philosophy of focusing on what drives the numbers. Happier customers require happier employees. Has today’s “bigger… better… faster… more… now” society come to the point where it is not only ignoring human needs, but demeaning them?

Work has become increasingly cerebral, and companies can’t treat new employees the way they treated those who worked with a pick and shovel—people won’t let themselves be treated like parts of a machine.

Davis Balestracci’s picture

By: Davis Balestracci

“I suffer simultaneously from amnesia and déjà vu. I have the feeling that I keep forgetting the same thing over and over again.”
Steven Wright (surreal comedian)

It all seems so logical, doesn’t it? Focus on processes, improve your organizational decision making through utilizing quality improvement tools, give people good technical and administrative information, and the organization “should” get better. It’s so tempting, interesting, and dramatic to lead in the vein of “Star Trek: The Next Generation” with Capt. Jean Luc Picard’s: “Make it so.”

However, in more than 25 years of facilitating quality improvement, I have learned two things:
1. Change would be so easy if it weren’t for all the people.
2. Logic + Humans = Change? Think again!

Davis Balestracci’s picture

By: Davis Balestracci

Click here to read part 1 of this series.

Analytic statistical methods are in very strong contrast with what is normally taught in most statistics textbooks, which describe the problem as one of “accepting” or “rejecting” hypotheses. In the real world of quality improvement, we must look for repeatability over many different populations. Walter Shewhart added the new concept of statistical control, which defines repeatability over time sampling from a process, rather than a population.

For example, the effectiveness of a drug may depend on the age of the patient, or previous treatment, or the stage of the disease. Ideally we want one treatment that works well in all foreseeable circumstances, but we may not be able to get it. Once we recognize that the aim of the study is to predict, we can see what range of possibilities are most important. We not only design studies to cover a wide range of circumstances, but to make the “inference gap” as small as possible.

Davis Balestracci’s picture

By: Davis Balestracci

This is an expanded version of an article that Balestracci wrote for Quality Digest in December 2007. 
--Editor

I discovered a wonderful unpublished paper by David and Sarah Kerridge several years ago (Click here to get a pdf). Its influence on my thinking has been nothing short of profound. As statistical methods get more and more embedded in everyday organizational quality improvements, I feel that now is the time to get us "back to basics"—but a set of basics that is woefully misunderstood, if taught at all. Professor Kerridge is an academic at the University of Aberdeen in Scotland, and I consider him one of the leading Deming thinkers in the world today.

Deming distinguished between two types of statistical study, which he called "enumerative" and "analytic." The key connection for quality improvement is about the way that statistics relates to reality and lays the foundation for a theory of using statistics.

Davis Balestracci’s picture

By Davis Balestracci

As this is my last column for Quality Digest, I’m delighted to announce that my distinguished predecessor, Donald J. Wheeler, will be writing this column again. You’ll be in good hands.

When I got my master’s degree in statistics in 1980, jobs were plentiful for internal statistical consultants in corporate research and manufacturing. Their jobs involved educating scientists and engineers in the power of applied statistical methods, especially experimental design. Excellent research in applied methods was performed and collegially disseminated by distinguished groups, most notably DuPont, Kodak, and General Electric. I was part of a strong internal group at 3M. Those groups, for all intents and purposes, have now vanished.

Statistical process control (SPC) became more dominant during the late 1980s. Everything became “bigger, better, faster, more, now!” Suddenly there wasn’t time to do response surface experimental designs, except maybe 22 or 23 factorials and a replication--if you were lucky. See “Using Design of Experiments as a Process Road Map” ( www.qualitydigest.com/feb06/articles/02_article.shtml ).

Davis Balestracci’s picture

By Davis Balestracci

Take a look at the control chart in figure 1. There are no observations outside the common-cause limits, but there are five special-cause flags:

Observation 5: Two out of three consecutive points greater than two standard deviations away from the mean

Observations 21 and 30-32: Four out of five consecutive points greater than one standard deviation away from the mean

 

So, what do you do? Treat them as five individual special causes? Say, “Well, if you look at it realistically, there really seems to be three ‘clumps’ of special cause?” Because the last seven observations all fall below the mean, some readers might want to call them special causes as well. Maybe nothing should be done because no individual points are outside of the limits.

What have been your experiences?

As I’ve tried to emphasize time and again in this column, always do a run chart of your data first (as seen in figure 2).

Davis Balestracci’s picture

By Davis Balestracci

Suppose you had 16 months of data on an important process (as plotted on the run chart seen in figure 1). For improvement purposes, an intervention was made after the sixth observation to lower this key process indicator. This intervention is equivalent to creating a special cause for a desired effect.

There is no trend as defined statistically (despite the trend downward of length five from observations 5 to 9; with 16 data points, one would need length six, or five successive decreases). Neither is there any run of length eight either all above or below the median. So, would you want to conclude that the intervention had no effect? I doubt it!

Think of the median as a reference, and consider each data point as a 50-50 coin flip (heads = above the median; tails = below the median). The question is: Would you expect to flip a coin 16 times and obtain the specific pattern of seven heads (run of length seven above the median), immediately followed by seven tails (run of length seven below the median), then a head (run of length one above the median) and, finally, a tail (run of length one below the median)? Intuition would seem to say, “No.” Is there a statistical way to prove it?

Davis Balestracci’s picture

By Davis Balestracci

This column is in honor of the first anniversary of my late father’s death. In his last days, Dad enjoyed watching golf, and I’d often join him. Watching the recent British Open, I thought I would apply some basic statistical principles to the final scores.

For example, 83 people made the cut, and the ANOVA of their individual round scores is shown in figure 1.

The two ANOM plots are shown in figures 2 and 3.

Another interesting statistic is the standard deviation of an individual round: square root 8.975 ~ 3. Using the standard Bartlett and Levene tests for equality of variances, I tested the 83 golfers as to whether this was consistent for all of them:

p-value Bartlett: 0.891

p-value Levene: 0.983

Depending on luck and other random factors, an individual’s score could swing by ± 6-9 strokes in a round!

Davis Balestracci’s picture

By Davis Balestracci

Eighty-four doctors treated 2,973 patients, and an undesirable incident occurred in 13 of the treatments (11 doctors with one incident and one doctor with two incidents), a rate of 0.437 percent. A p-chart analysis of means (ANOM) for these data is shown in figure 1.

This analysis is dubious. A good rule of thumb: Multiplying the overall average rate by the number of cases for an individual should yield the possibility of at least five cases. Each doctor would need 1,000 cases to even begin to come close to this!

The table in figure 2 uses the technique discussed in last month’s column, “A Handy Technique to Have in Your Back Pocket,” calculating both “uncorrected” and “corrected” chi-square. Similar to the philosophy of ANOM, I take each doctor’s performance out of the aggregate and compare it to those remaining to see whether they are statistically different. For example, in figure 2, during the first doctor’s performance, one patient in the 199 patient treatments had the incident occur. So, I compared his rate of 1/199 to the remaining 12/2,774.