What is an instrumental failure

to directory mode

Types of errors

In addition to the so-called gross errors (imprecise folding rule, measuring tape for micro measurements), two types of errors play a role: statistical errors and systematic errors.

Statistical (random) errors (engl. error)

You determine the precision (engl. precision) and thus the reproducibility of a process. They are unavoidable and cannot be characterized.

Reproducibility is a measure of the spread between random variables.

Systematic errors (engl. bias)

They influence the correctness (engl. accuracy, trueness) of an analysis process.

They are the deviations of the results from the true value and are caused by disruptive influences (unexpected components) or faulty measuring technology (wrong method, faulty device, warped folding rule). You can only by comparative measurements

  • with different technology
  • in other laboratories
  • through round robin tests

can be discovered, but not by repeated measurements.

A frequently used representation of the expected value of a test result is the mean value. In this case the bias represents the deviation between the mean value of a number of results and the recognized reference value. In analytical chemistry this is usually referred to as systematic error.

Often accuracy is also used as an overall error. Unfortunately, the DIN and ISO standards are inconsistent.

Since the true value is usually not known, an attempt is made to approximate it as closely as possible by reducing all systematic sources of error and to reduce the uncertainty of the determined value through multiple measurements.

Applet: Systematic and random error