No. 1742: ACCURACY AND PRECISION Today, our meters fool us. The University of Houston's College of Engineering presents this series about the machines that make our civilization run, and the people whose ingenuity created them. In an older episode about analog and digital watches, I made the offhand remark that digital watches report the time with greater precision. When a listener wrote to point out that a digital watch can be just as wrong as an analog watch, I realized that I'd used the word precise in its technical engineering sense. We use the two terms accurate and precise to mean quite different things. Imagine for example, two police officers on the target range. When they look at their respective targets, the first discovers a very tight grouping of hits, four inches to the left of the center. The second officer finds a random scatter of hits, eight inches in diameter, centered about the aiming point. And so the first one's shooting is extremely precise, but it's inaccurate. The second one is quite accurate and very imprecise. The word precise actually means no more than well-defined. We save the word accurate for close-to-being-correct. This issue took on special importance back in the mid 1970s when we all gave up our slide rules in favor of pocket calculators. Slide rules were precise within only a percent or so. Now, in a blink, we could divide two numbers and get a ten-decimal-place result. The result was both precise and accurate. Or was it accurate? Our work almost always depends on measured data -- like conductivities or yield-stresses -- which are usually a few percent off. Students were now reporting calculations to a precision of a ten thousandth of a percent, when they could not possibly be accurate within one percent. This sort of thing doesn't stop in the engineering office. I always grow nervous when someone praises a person's precision of language, for the same problem applies. Back in 1950, Senator Joe McCarthy made his famous speech in Wheeling, West Virginia. He announced that 205 State Department employees were communists who reported directly back to the Kremlin. His statement was wonderfully precise, and wholly inaccurate. On the other hand, when poet John William Burgon called the ancient city of Petra, A rose-red city half as old as time, he spoke without any precision, but with a curious kind of accuracy. This issue dogs us all as we read political statements, advertising copy, or any attempt to win us over to various causes. We are constantly dazzled by precision, while accuracy is absent. And I'm back to digital readouts. Film critic Roger Ebert likes to talk about the hackneyed scene where clock with a red LED ticks down toward zero. That's when, say, scientists have predicted that a chemical reaction will complete itself and blow up the universe. The despairing hero looks at the reactor's wiring and says, Shall I cut the green wire or the red? Wouldn't it be something to see, just once, a hero who shrugs and says, Wait a minute; the prediction's only approximate. I can take another minute and get it right. I'm John Lienhard, at the University of Houston, where we're interested in the way inventive minds work. (Theme music) These issues are treated very early in any engineering measurements textbook. See, e.g., T. G. Beckwith, R. D. Marangoni, and J. H. Lienhard, V, Mechanical Measurements, 5th ed.. Reading, MA: Addison-Wesley Publishing Co., 1993. The Engines of Our Ingenuity is Copyright © 1988-2002 by John H. Lienhard.