Systematic bias: Encyclopedia II - Systematic bias - Systematic vs. random

Systematic bias - Systematic vs. random

An example of systematic bias would be a thermometer that always read three degrees colder than the actual temperature because of incorrect initial calibration or labelling, whereas one that gave random values within five degrees either side of the actual temperature would have random error.

Once detected, systematic effects are easier to take into account than random effects: in the example just given, if you know that your thermometer always reads three degrees below the correct value, you can simply make a systematic correction by adding three degrees to all readings; there is no equally simple correction for random error.

The existence and causes of systematic bias may be difficult to detect without an independent source of information; the phenomenon of scattered readings resulting from random error calls more attention to itself from repeated estimates of the same quantity than the mutually consistent incorrect results of a biased system.

However, systematic can additionally sometimes be used imply planned human agency. Systematic bias therefore can also mean that the system produces bias as a consequence of consistent, deliberate and planned human interference.

Adapted from the Wikipedia article "Systematic vs. random", under the G.N U Free Docmentation License. Please also see http://en.wikipedia.org/wiki