Accuracy Calculator
This accuracy calculator tells you how close a measurement sits to the value it should be, expressed as a percentage. Enter your measured or observed value and the true, accepted or target value, and the calculator finds the gap between them, scales it against the true value, and reports the accuracy as 100 percent minus the percent error. It is the everyday complement of a percent error calculation: where percent error tells you how far off you were, accuracy tells you how much you got right. Science and chemistry students use it to grade experimental results against a textbook value, technicians use it to check an instrument against a calibration standard, and anyone comparing a forecast, an estimate or a sensor reading against a known outcome can use it to put a single clean number on the quality of the match. The calculator also shows the underlying absolute error and percent error so you can see exactly where the accuracy figure comes from. Keep in mind that accuracy is about closeness to the true value, which is different from precision, the closeness of repeated readings to each other. If your measurement is off by more than the true value itself the accuracy can go negative, which is simply a loud signal that something has gone badly wrong rather than a score to interpret literally.
Accuracy is defined as 100 percent minus the percent error, where percent error is the absolute difference divided by the true value. Accuracy measures closeness to the true value, not the consistency of repeated readings. Enter a non-zero true value.
How it works
The absolute error is the size of the gap between your measured value and the true value, ignoring which way it points. Dividing that gap by the true value and multiplying by 100 gives the percent error, the standard way of expressing how far off a measurement is relative to what it should be. Accuracy is then 100 percent minus the percent error, so a small error means high accuracy and a large error means low accuracy. Using the true value as the denominator, rather than the measured value, is the convention in science because the true value is the fixed reference you are comparing against. If the two values match exactly the error is zero and the accuracy is a full 100 percent.
Worked example
You measure the boiling point of a sample and read 98, while the accepted value is 100. The absolute error is the difference, which is 2. The percent error is 2 divided by 100, multiplied by 100, which is 2 percent. The accuracy is 100 minus 2, which is 98 percent, a good result. If a second reading came in at 107 against the same true value of 100, the absolute error would be 7, the percent error 7 percent, and the accuracy 93 percent, telling you at a glance that the first reading was the closer of the two.
Related calculators
- Percent Error Calculator: the error side of the same comparison.
- Absolute Error Calculator: the raw size of the gap.
- Relative Error Calculator: error as a fraction of the true value.
- Confusion Matrix Calculator: classification accuracy from TP, FP, FN, TN.
- Percentage Calculator: any part-of-a-whole percentage.