Through measurement and/or calculation, we obtain determined value `X_o` of something. But this value is rarely equal to the real one `X`. The difference of these two values is error.

`"error"=X-X_o`(1)

Errors exist frequently in reality, and in most cases, inevitable. Sources of errors are varied: measurement devices, method of measurement, environment factors, ...

In general, we do not know the real value `X`, so we do not know exact value of error. We can only estimate it with certain degree of confidence.

Error consists of two components:

• systematic error,
• random error.

Reducing errors, especially systematic ones, as much as possible is a topic in statistical science.

 

In the fields of science, engineering, industry and statistics, accuracy and precision of measurement are two different concepts. The illustration most commonly used to explain their differences is a target and the shots on this target (Fig. 1)

Fig.1 Accuracy and precision

The difference between these concepts can be expressed as:

• Accuracy of a measurement is defined as the degree of proximity of measured values to the true one.
• Precision is used to express the degree of repeatability of measured values.