Measurement

Since science is really only concerned with what is measurable, measurements of various kinds are very important and pop up all over the place in almost any study of anything, but the numbers are only as good as the equipment and the people using the equipment. If you need centimeters and all you have are rulers with inches on them, you're in trouble. Or you could have the fanciest gas chromatograph on the market but only somebody like me to use it. It wouldn't do you any good.

If you've ever needed a really straight line, you have experience with this. The saying goes "Measure twice. Cut once." If you want a straight line you probably need more than two measurements. You start at the edge of your material and go... say ten inches (Why are you using inches instead of centimeters, anyway?). You make four or so marks along the length at ten inches and then try to connect them, but they aren't really all lined up. Some are a little farther out than others and you end up making a "best fit" line, which, if you are really lucky, looks straight.

What you are lacking is precision. Your lines are all pretty close to ten inches so your accuracy isn't bad, but sometimes you marked toward at the far end of the little black measuring tick and sometimes at the close end, also the ruler might have been at a slight angle sometimes, making slight variations in the measurements. Precision is the ability to come up with the same answer over and over again when measuring the same thing over and over, and it has a lot to do with both the equipment and the training of the person using it.

Accuracy is important too. Just because you came up with ten inches on the dot, every time, doesn't mean you have ten inches. Your ruler could be wrong or you could be looking at it and seeing the number ten where it should be twelve because the six-year-old took a pen to your ruler.

Precision sounds wonderful. It's a word that makes me think of professionalism, but precision without accuracy can be disastrous. Just think of the surgeon that could make a cut within a millimeter of where he intended, but took out the wrong organ. A little more accuracy and a little less precision might make the patient a little more alive.

Links:

If you ever need to collect measurements for the EPA... here are the methods.

The National Institute of Standards and Technology's Center for Nanoscale Science and Technology.

Here's how Australia measures things.

Word of The Day:

Bias: There are several ways this word is used in statistics (and we are talking about some very simple statistics here) but in measuring things, your group of measurements would be biased if they were wrong in some way that caused them not to fall into a "normal" bell curve... for instance, if they were precise but inaccurate because of that pen mark on the ruler's number two that made the 12 look like 10.