Relative Error Calculator

What the relative error calculator does

Relative error, usually reported as percent error, measures how far a measured value falls from the true value, as a fraction of that true value. This calculator takes an observed value and a known true value and returns the percent error between them, keeping the sign so you can see the direction of the miss.

It is the standard way to express how accurate a measurement or estimate is. Below is how it works and why scaling the error to the true value is so useful.

How to use it

1. Enter the observed value, the figure you measured or estimated.
2. Enter the true value, the known or accepted figure.
3. Press Calculate for the percent error, or Reset to clear it.

How percent error is worked out

The calculator takes the gap between the observed and true values and expresses it as a percentage of the true value:

Percent error = (observed value minus true value) ÷ true value × 100

The top is the plain error, how far off you were in raw terms. Dividing by the true value turns that into a proportion, so the error is measured relative to the size of what you were trying to hit, and multiplying by 100 makes it a percentage. This is what separates relative error from absolute error: it is always judged against the true value, not left as a bare amount.

What the sign tells you

This calculator keeps the sign of the error, which carries useful information. A negative percent error means the observed value was below the true value, an underestimate. A positive percent error means it was above, an overestimate. A percent error of zero means you hit the true value exactly.

Keeping the sign tells you not just how far off a measurement was, but in which direction, which can matter when a consistent bias is at work. If repeated measurements all come out negative, that points to something systematically pulling the readings low, a clue you would lose if only the size of the error were reported. Some settings care only about the magnitude, in which case the sign can be dropped, but knowing the direction is often the more informative choice.

Why relative beats absolute for comparing

The real strength of percent error is that it makes errors comparable across very different scales. An error of one centimetre is enormous when measuring something ten centimetres long, but trivial when measuring a distance of ten metres. The raw error, one centimetre, is identical in both, yet its importance is completely different.

Percent error captures that difference: the same one-centimetre miss is 10 percent in the first case and one-tenth of a percent in the second. By scaling the error to the size of what is being measured, it tells you how significant the error really is, not just how large. This is why percent error is the natural measure when you want to judge or compare the accuracy of measurements of different magnitudes.

A worked example

Suppose an experiment gives an observed value of 48 when the true value is known to be 50. The error is 48 minus 50 = minus 2, and as a percentage of the true value that is minus 2 divided by 50, times 100, which is minus 4 percent.

The size, 4 percent, says the measurement was off by a modest amount relative to the target. The negative sign says it was an underestimate, coming in below the true value. Had the observed value been 52 instead, the percent error would be plus 4 percent, the same size of miss but in the overestimate direction.

Entering your values

Enter the observed value and the true value. The calculator returns the percent error, keeping its sign to show whether the observation was an under or overestimate. Because the error is divided by the true value, that value sets the scale against which the error is judged.

Questions people ask

What is relative error?

The difference between an observed and a true value, expressed as a percentage of the true value. It measures accuracy relative to the size of what was being measured.

What does a negative percent error mean?

That the observed value was below the true value, an underestimate. A positive percent error is an overestimate, and zero is an exact hit.

How is it different from absolute error?

Absolute error is the raw gap between the values. Relative error divides that by the true value, so it reflects how significant the error is for the scale involved.

Why use percent error to compare measurements?

Because it scales the error to what is being measured. The same raw error can be large on a small quantity and tiny on a large one, and percent error reflects that.

References

A quick note on where the methods here come from. Measurement error and its expression relative to a true value are core topics of the NIST/SEMATECH e-Handbook of Statistical Methods, the US government's public statistics reference and the recognised authority on measurement. OpenStax provides free, widely used science and statistics textbooks covering measurement error and accuracy.

1. NIST/SEMATECH e-Handbook of Statistical Methods (measurement error and accuracy). https://www.itl.nist.gov/div898/handbook/
2. OpenStax (free science and statistics textbooks covering measurement error). https://openstax.org/subjects/math