SR: Individual concern?
In the May edition of QP (InBox), an interesting letter by Professor J. Douglas Barrett was published addressing the issue of social responsibility (SR). His letter was written in reaction to the March issue of QP, which contained several contributions on this very hot topic related to business ethics.
In my view, there is something fundamentally wrong with all of these discussions. There is no such thing as a socially responsible or irresponsible company, nor is there such thing as an ethical or unethical one. A company is a legal entity created to allow people to do business within a given legal framework. Companies do not have a conscience, a feeling of responsibility or ethic. These are human characteristics. These are qualifications that can be given to actions taken by humans.
Enron was not an unethical company. Certain people within Enron acted in a manner that was later deemed to be unethical and illegal. I am fairly certain the vast majority of Enron employees were people who behaved in a highly ethical manner with a high degree of responsibility toward society.
Ethics will never come from companies because ethics cannot come from companies. They will always come from people. It is our responsibility as individuals to think about the kind of society we want and to be involved in creating that society. We can do this at a company level, but it should be the people who defi ne SR for the entity, not the other way around.
In Zen and the Art of Motorcycle Maintenance by Robert M. Pirsig, there is a quote from Plato’s Phaedrus: "And what is good, Phaedrus, and what is not good—need we ask anyone to tell us these things?"
Maybe we should ask ourselves this question rather than creating a standard for SR that allows companies to make meaningless statements in the same way that so many companies make meaningless statements about quality based on a standard. We certainly do not need to make the same mistake all over again.
One question for Professor Barrett: Is a person who has seven children lying when he states that overpopulation is a major problem? The comment about Al Gore at the end of the letter, which suggests that the value of any statement is directly related to the personal behavior of the person making that statement, is a gross intellectual dishonesty. Sinners have said many inconvenient truths throughout history, just as would-be saints have often lied.
Six Sigma illusions
As a quality executive, I am dismayed at misconceptions about Six Sigma that undermine the years of work quality professionals have devoted to define their field of expertise. I still see companies treating quality management as an easy assignment that can be filled by any capable engineer or Six Sigma Black Belt (BB).
Recently, two encounters have amused me. In conversation with a BB serving in a corporate quality role, I bemoaned the demise of Joseph Juran. She asked, "Who’s that?" I explained that Juran was a visionary for his work in quality management and quality engineering. I further explained that when I studied for my certified quality engineering exam, Juran’s Quality Control Handbook was the source text I used.
A few weeks later, a supplier told me it had just fired its new quality manager and couldn’t explain the manager’s inability to perform his duties. After all, the company paid top dollar for a recent college graduate who was a BB, thinking this designation was the highest honor bestowed on any quality practitioner.
Don’t get me wrong. Six Sigma has had a tremendous impact on corporate America. Its implementation by executives has resulted in major product and process improvements, and significant cost savings. As I worked on my first Green Belt project, however, I experienced déjá vu. Isn’t this just a repackaging of tools that have been around for decades, with a few new approaches added? There is so much more to our trade than just these tools. Let’s not let Six Sigma silliness overshadow the value of credentialed quality managers, engineers and auditors.
Gary D. Lizotte
In the April 2008 edition of QP, the author of the article "Test Drives and Data Splits" (3.4 per Million) uses the term confidence interval when he is referring to a prediction interval. I’ve found this to be a common mistake in statistics that should be clarified.
A confidence interval is the range in which the population’s characteristic is expected to fall based on the sample data. For example, a 95% confidence interval on a mean signifies there is a 95% chance the population’s average will fall within the values predicted by the sample data.
A prediction interval is the range in which individual values of the population are expected to fall based on the sample data. Just as the name implies, we are predicting where individual data values will fall.
The author of the article uses the term confidence interval when he is actually referring to prediction intervals. If he had used the confidence interval, the result of this exercise would have been much different. Using the data in the article, I compiled a Minitab graph of the 95% confidence interval and the 95% prediction interval to help illustrate the difference (see Figure 1).
Notice that the confi dence interval range is much smaller. This is because we expect the regression of the population to fall within these lines. The prediction interval is much wider because we expect individual values of the population to fall within that range.
Tinley Park, IL
Carroll’s point reinforces the importance of being as clear as possible regarding one’s choice of terms. One should always be aware of the difference between a confidence interval for the mean of future observations and a confidence interval for a single future observation.
To simplify the discussion, the former is frequently referred to as a confidence interval and the latter as a prediction interval. In my article, I meant to refer to a confidence interval for a single future observation, and I believe the discussion in the article supports that intention.
If one is accustomed to conversing in terms of confidence and prediction intervals, the lack of that term in the article is certainly a possible source of confusion for which I apologize. If the reader traces the steps in Appendix 2 as described in the online link provided with the article, I believe you will find they correspond to the formula for a prediction interval.
My thanks go out to Carroll for his helpful point.
Joseph D. Conklin