Friday, April 28, 2006

What is Uncertainty

“Uncertainty is inherent in any measurement process”

Let us understand this statement.

If I say, “No measurement can be perfect”, will you agree me? Well, I don’t mind if you say NO; just because you haven’t read the following text.

If I give you a piece of metal, say a small rod, and ask you what it’s the length? Then what you will do is called measurement. You will take a scale and tell me its 17 mm long. Well, Let me ask you a simple question at this point, are you sure? If you are wise enough you will say NO and will disappear only to come along with a “venier caliper”. You will measure with it and tell me, actually its 17.1 and not 17.0. But, still I would like to ask you the same question, are you sure? Well, this time with a “micrometer”, you will tell me its actually 17.13 and not 17.10.But, still I would repeat my question. Now, with a “metroscope” you will tell me its actually 17.134 and not 17.130, and also that you are NOT sure about the measurement.

You can NEVER be sure. That’s why I said, “No measurement can be perfect”.


Then, how can we deal with this situation? It’s simple. Instead of specifying result as a value, tell it as a range. What do I mean is that, instead of saying length of rod as 17, just say that it can be somewhere in between 16 to 18. With a scale you can easily tell this. You can judge that the length of rod is more than 16 and less than 18. So, what you tell me is length of rod is 17 ± 1 mm.

This 17 is the result of measurement and ±1 is called uncertainty of measurement.

Similarly, you will say with “vernier caliper”, measurement is 17.1 and uncertainty is ±0.1 and so on.

Thus, no measurement is perfect unless uncertainty is specified along with the results of measurement.


Note –
What we are considering here is only one of the factors contributing to uncertainty in measurement, the least count of instrument. Well, let me tell you one thing. As you use more and more precise instruments, effects of temperature, method of measurement, accuracy of instruments and many more need to be considered. And, that makes uncertainty calculation complex. We learn it in next few articles.

0 Comments:

Post a Comment

<< Home

electronic health record software
electronic health record software