Leadership & Business: Lies, Damn Lies, and Statistics

Interesting Innovations

Street Heat:
Ever burn your foot walking on hot asphalt in the summer? That’s because black absorbs heat—while white reflects it. Well, in case you haven’t noticed, modern cities are covered in the black stuff. Dutch construction firm Ooms is now heading its headquarters by running water pipes under the street. Some of them collect heat in the summer and run deep into the ground where they heat water via a heat exchanger. That heated water is stored for winter—a sort of battery, if you will. In fact to take it a step further, the water is returned to the ground after heating the building, by passing under the street again. The residual heat in the water, now only a few degrees above freezing, melts any snow or ice on the road surface. The water is then stored—used cold to cool the building—before being run under the asphalt again to prepare for winter. Brilliant!

Lies, Damn Lies, and Statistics

Rarely do I find the need to directly blast another writer, but this is one of those times. In a recent article published by Chief Executive Magazine the value and legitimacy of six sigma as a means of improving business performance is challenged by Charles Holland, Ph.D.

One doesn’t have to read very deep to see that this article is really nothing more than an advertisement for Dr. Holland’s book. As a PhD and statistician, Dr. Holland knows all too well the problems with his analysis. But for everyone else, I think they’re important to discuss.

First, there is absolutely no reference to where you can find the details of his research and there is no mention of the methods used. For example, he cites research involving 58 companies. Yet thousands have implemented six sigma. What science was used to ensure these 58 are truly and statistically representative of the rest. The lack of any discussion or reference to where this discussion can be found leads me to ask whether they were chosen to reinforce his point.

Then, while Dr. Holland uses the entire article to hype Multivariable Testing (MVT)—which by its very name implies that many variables are necessary to truly understand a business process, he would like to credit six sigma will all or none of a company’s success, rather than acknowledge that the implementation of six sigma is but one of many variables that determine a company’s success. He also fails to acknowledge that what he brands as multivariable testing is really nothing new, and is included in the six sigma tool set used by any well-trained six sigma expert.

It is true that six sigma alone won’t lead to success. And it is true that a myopic view of six sigma can be detrimental. But it is equally true that when used in combination with strong leadership, a sound business strategy, and in some cases alignment of the starts, six sigma has been proven to deliver exceptional results.

The point is that, as Mark Twain is often quoted, “There are lies, damn lies, and then there are statistics.” Dr. Holland should be ashamed of the way he has tried to use statistics to distort the truth. And Chief Executive should be equally ashamed that they published his work.