Percent Error Calculator

Percent error = (|experimental - theoretical| / |theoretical|) x 100. You subtract the theoretical value from your measured value, take the absolute value of that difference, divide by the absolute value of the theoretical value, then multiply by 100. In the default unsigned mode the result is always zero or positive.

Use it whenever you measure a quantity that has a known accepted value, such as the density of water (1.00 g/mL), the boiling point of ethanol (about 173 F), or g in a pendulum experiment (32.2 ft/s^2). Percent error tells you how far your measurement drifted from that reference, giving you a standardized way to judge accuracy across different experiments.

Absolute percent error drops the sign and reports only magnitude. This is the default and what most lab reports expect. Signed percent error keeps the direction: a positive result means your experimental value was above the theoretical, and a negative result means it was below. Use the signed option when you need to track whether your measurements consistently run high or low.

"Over" means your experimental value is greater than the theoretical value, "under" means it is less, and "exact" means they match. This label appears even in absolute (unsigned) mode, so you can still see which way your systematic error leans without switching to signed mode.

Percent error divides by the theoretical value, so a zero denominator makes the formula undefined. If your accepted reference value is zero, report the absolute error in raw units instead, for example "my measurement was 0.05 g off" rather than a percentage.