Percent Error Calculator
See how far a measured value lands from the accepted true value — as a clear, never-negative percentage.
One number, clear accuracy
Enter your measured value and the true value to get the percent error instantly — lower means more accurate.
The true value can't be zero
Percent error divides by the true value, so it is undefined when the true value is 0. The measured value may be negative or zero.
What is percent error?
Measured vs. accepted, as a percentage
Percent error tells you how far an experimental (measured) value falls from a theoretical (true or accepted) value, expressed as a percentage of that true value. It answers one practical question: how accurate was my measurement? Because the formula uses the absolute value of the difference, the answer is never negative — it captures the size of the error, not whether you measured too high or too low. That makes it the standard accuracy check in lab experiments, physics and chemistry classes, instrument calibration, and quality control.
Enter the measured value and the accepted true value to get the percent error instantly.
One short formula, built from the difference between your two values and the size of the true value.
percent error = |measured − true| ÷ |true| × 100%Take the difference between the measured and true values, drop the sign with the absolute value, divide by the absolute true value, and multiply by 100 to turn the ratio into a percentage. The absolute values are what keep the result positive and meaningful no matter the direction or sign of your numbers.
Suppose you measure the acceleration due to gravity as 9.8 m/s², while the accepted value is 9.80665 m/s².
Difference
9.8 − 9.80665 = −0.00665 — take the absolute value to get 0.00665.
Divide by the true value
0.00665 ÷ 9.80665 = 0.000678 — the error as a fraction of the true value.
Convert to a percentage
0.000678 × 100 = 0.067811% — a very accurate measurement.
A percent error is a plain accuracy score: the closer it sits to 0%, the nearer your measurement was to the true value. In the worked example, 0.068% means the reading of 9.8 was almost spot-on — well within what you would expect from a careful experiment. A result of 20%, by contrast, signals that something is meaningfully off: a mis-read instrument, a flawed method, or genuine variation in what you measured. Because the formula uses absolute values, the percent error never tells you the direction of the miss — measuring 8 against a true 10 and measuring 12 against a true 10 both yield 20%. To know whether you over- or under-shot, look back at the raw difference before the absolute value was applied. What counts as "good" depends entirely on context: a school physics lab might happily accept under 5%, while precision metrology or industrial quality control may demand a small fraction of a percent. Always judge the number against the tolerance for your specific task rather than treating any single threshold as universal.
The formula is exact, but a couple of practical points are worth keeping in mind.
Needs a true value, and it can't be zero
Percent error only makes sense when you have a genuine accepted or true value to compare against. If you are comparing two equally-valid measurements with no reference, use percent difference instead, which divides by their average. Percent error is also undefined when the true value is 0, because the formula divides by it — the calculator returns no result in that case.