Accuracy, precision and errors in measurement

We've left a very important topic until the end of this unit, because by now you have had the opportunity to make several different sorts of measurement, and you should have developed a 'feel' for measurement.

Accuracy, precision and error are important in all types of measurement. In other words, these things apply to measurement in general. Remember the importance of generalisation in scientific method!

Accuracy

'Accuracy' in measurement means getting close to the real or correct measurement. Imagine you used your ruler to measure the length of a pencil. Let's say you got a measurement of 7. 4 cm. Then a scientist, who is an expert at measuring things, came and measured the pencil with a very special, sensitive, high-quality instrument. The scientist finds the pencil to be 7.38 cm long. Your measurement was very close to the value obtained by the scientist. Your value was very accurate!

The opposite of accurate is inaccurate. If you had found the pencil to be 7.0 cm long, you would have been ‘out' by about 0.4 cm. This would have been an inaccurate result.

Precision

The word 'precision' is close in meaning to accuracy, but there are some differences. Imagine someone used a special instrument to measure the mass of a body. Because she uses a special, sensitive balance, she is able to measure the mass with great precision. Suppose she finds the body has a mass of 6.2753 kilograms. We would say that this is a very precise result.

It has a great amount of 'detail' in it. The mass has been measured to four decimal places of a kilogram. If we used a simple instrument to weigh the same body we might have found a mass of 6. 3 kg. There is less precision in this result.

Is the scientist's precise result very accurate? Perhaps it is. But it might not be. Suppose the very sensitive instrument had not been kept properly. Perhaps it had been slightly damaged, and was not working perfectly. In this case the result may not be very accurate. When scientists use very sensitive, complicated instruments, they check them carefully before use, to ensure that they are working properly.

Errors

In scientific work, an ‘error' is not the same as a ‘mistake'. If your name is Patrick but I call you Peter, that is a mistake. It is simply wrong. It is not an error.

If you measure something, such as a mass, a length, a temperature, even if you are careful, and use a good instrument, there will always be some amount of error in your result. If you work carefully, the size of the error will be small, but there will always be some errors. Scientists have ways of working out how big an error in a measurement is likely to be. If the error is small enough, it won't matter.

A carpenter making a garden bench needs to measure and cut wood to ‘more or less' the right lengths. Even quite big errors don't matter. An engineer making parts for an engine, or building a bridge, will need to measure and cut the pieces with much greater care, because the errors will need to be much smaller.