Absolute error is a measure of how far a single measured or estimated value is from the true or accepted value. It shows the magnitude of the difference, regardless of whether the measured value is too high or too low. It measures error using the same units as the original data, making it easy to interpret the size in practical terms.
The absolute error is always expressed as a positive number and provides a simple way to understand how much error is present in a single measurement.
Absolute error gives you a straightforward answer to the question: How far off was the measurement?
The absolute error formula is the following:
This formula takes the absolute value of the difference between the measured value and the true value, so negative signs are ignored.
It does not indicate whether the error is an overestimate or underestimate, and it does not account for the scale of the values involved. For that, you would use relative error or percent error.
For example, a thermometer reads a temperature of 102.4°F, but the actual temperature is 100.0°F. The absolute error is the following:
This result indicates the reading is off by 2.4 degrees, disregarding the direction of error.
« Back to Glossary Index