The Biggest Statistical Mistake of 2007 (and Every Other Year!)

The most common statistical mistake in Business, Research and Development, and Manufacturing is failing to repeat measurements. Day in and day out, people make decisions based on a single measurement. Some do not realize the importance of making repeated measurements. Others want to save time and money by making as few measurements as possible. Still others don't realize that measurements of exactly the same thing in the same way by the same person vary. Measurements vary. This is a fact of Nature, whether or not we like it. You can easily prove this to yourself. Pick something to measure -- anything you like. Try to measure it exactly the same way three different times. Did you get the same number each time? If so, find a more sensitive measuring device and try this again. If the device is sensitive enough, you will see variation. Interestingly, people who realize that measurements vary often assume that this variation is small. If it is small, repeated measurements may be skipped without much risk. If the variation is actually large, you can very easily be misled by a single measurement and make costly mistakes. In my experience, measurement variation is usually much larger than you expect. Suppose you visit the doctor and he measures your cholesterol level. He measures 245 mg/dL -- should you be concerned? An in-office cholesterol determination can have a standard deviation that is 5% of the value measured. This means that if you have a true cholesterol level of 215 mg/dL you could receive a single measurement of your cholesterol anywhere from 185 to 245 mg/dL. The single high measurement could cause your doctor to begin treatment with pharmaceuticals. If he had received a single low measurement of 185, he might conclude that you are fine. If he had received the 215 measurement he might prescribe diet and exercise improvements. In other words, he might take completely different action simply because of measurement variation. If your doctor were to repeat the measurement and use the average value, he would more likely take the best action. The same thing can happen to you! Fortunately, it is easy to avoid this mistake by repeating your measurements. If you make a habit of looking not only for a measured value, but also the variation in that value, you will be able to make more informed, less risky decisions. Why not eliminate this mistake from your work in 2008? Objective Design of Experiments workshops will teach you to use DOE in your work. Design of Experiments is a fundamental technique for industrial experimentation. You will learn to apply DOE easily without excessive math and theory. We will help you be even more successful!