Method
Percent error = |observed − accepted| ÷ |accepted| × 100. Signed relative error retains the direction of the difference.
Math worksheet
Percent Error Calculator
Compare an observed value with an accepted reference and separate absolute error, relative error and signed bias.
The calculation runs locally. Exact-integer tools avoid floating point; decimal and transcendental tools report rounding limits.
Method and verification guide
How the Percent Error Calculator works
Percent error = |observed − accepted| ÷ |accepted| × 100. Signed relative error retains the direction of the difference. The calculator keeps the intermediate values visible so you can reproduce the answer independently.
Interpret the result
Absolute percent error reports magnitude; signed error shows whether the observation is above or below the reference.
Check the assumptions
Percent error is undefined when the accepted reference is zero.
Local calculation
Everything runs in this browser. Inputs are not sent to an external calculator.
Rounding
Displayed values may be rounded for readability while calculations use full floating-point precision unless the tool explicitly uses exact integers.
Scope
Percent error is undefined when the accepted reference is zero.
Simple by design
Three steps to a clearer number.
Enter your details
Use realistic values in each field. You can change them anytime.
Calculate instantly
The formula runs locally, so there is no account or waiting time.
Use the result
Treat the answer as a practical estimate for your next decision.