Relative Error Calculator

Calculate the unit-less ratio between absolute error and the true value — the standard expression of measurement quality in physics, engineering, and analytical chemistry.

Relative Error Calculator

Compute the unit-less |Vobs − Vtrue| ÷ |Vtrue| ratio with decimal and percentage forms — the foundation of percent error and accuracy.

Observed Value

True Value

Try a quick example

What Is Relative Error?

Relative error is the unit-less ratio between the absolute error of a measurement and the true value being measured. By dividing the difference by the magnitude of the accepted value it removes the units, letting you fairly compare the accuracy of a kilogram-scale weight reading against a milligram-scale chemical assay.

This relative error calculator computes |Vobserved − Vtrue| / |Vtrue| for any pair of observed and accepted values, returns both the decimal and percentage forms, and grades the result on a five-level measurement-quality scale from excellent to poor. It is the underlying ratio behind percent error, accuracy, and most published uncertainty figures.

This is one mode of the full Percent Error Calculator — pair it with the percentage calculator and scientific calculator for related scientific work.

How Relative Error Is Calculated

Subtract observed from true

The first step is the signed difference — how far the measurement strays from the accepted value. Positive means the measurement is too high, negative means too low.

Take the absolute value

Relative error reports only the magnitude. Take the absolute value of the difference so direction does not bias the comparison between trials.

Divide by the true value

Scale the absolute difference by the magnitude of the true value. The result is dimensionless — fractions of unity — which lets you compare across scales and units.

Optionally convert to percent

Multiply by 100 to express relative error as a percentage. Both forms are equivalent; lab reports typically use the percentage form for readability.

The Relative Error Formula

Relative error is defined as the absolute error divided by the magnitude of the true value. It is the foundation under percent error and the inverse complement of measurement accuracy.

Relative error

RE = |Vobs − Vtrue| ÷ |Vtrue|

Absolute error scaled by the true value. Dimensionless.

Percent equivalent

RE% = RE × 100%

Same quantity expressed on a 0–100% scale. Identical to absolute percent error.

Accuracy from RE

Accuracy = (1 − RE) × 100%

Complement of relative error — how close the measurement got to the true value.

How to Use the Relative Error Calculator

1
Enter the observed value
Type the value you measured in the experiment, taken from a meter, scale, balance, sensor, or other instrument output.
2
Enter the true value
Type the accepted, reference, or theoretical value — whatever you are comparing your measurement against. The relative error is computed relative to this number.
3
Calculate and read the result
You get both the decimal form (e.g. 0.0454) and the percentage form (4.54%), along with a quality grade and step-by-step working that mirrors the formula above.
4
Compare across units and scales
Because relative error is unit-less, you can compare a 4.5% error on a 1 kg scale to a 4.5% error on a 1 mg balance — both represent the same fractional deviation from the true value.

Key Relative-Error Concepts

Dimensionless ratio

Relative error is unit-less because both numerator and denominator carry the same units. This is what lets you compare measurements across scales and disciplines.

Decimal vs percent

0.045 and 4.5% are the same quantity. Lab reports usually print percentages; physics texts often print decimals or fractions of unity.

Magnitude only

Relative error discards the sign. Use signed percent error or signed difference when direction matters (e.g. systematic bias detection).

Reference dependence

Relative error scales with the true value. The same 1 mL absolute error is 1% on a 100 mL measurement and 10% on a 10 mL measurement.

Where Relative Error Shows Up

🔬

Analytical chemistry

Titration results report relative error to certify whether a method meets a published accuracy spec — typically below 2% for routine assays.

🧪

Physics lab

Standard student lab reports quote a measurement as 'g = 9.71 m/s² (1.0% relative error)'. The 1% number is the comparison metric.

📡

GPS positioning

A 3 m absolute error matters very differently when locating a city versus an indoor robot — relative error makes that scale-dependence explicit.

💉

Drug dosing

Pharmacopoeia specs cap acceptable relative error on dose preparation at 5%, regardless of whether the dose is 10 mg or 10 g.

⚖️

Mass measurement

Calibrating a scale: a 1 g deviation is irrelevant on a truck weighbridge (≈0.000003%) but disqualifying on an analytical balance (≈0.1%).

🌡️

Temperature sensors

Thermocouple specs are quoted as ±0.4% of reading — a relative error spec that yields ±4 °C at 1000 °C and only ±0.4 °C at 100 °C.

Best Practices for Relative-Error Reporting

✓Always state the reference value. Relative error means nothing without the denominator. Report '4.5% relative to the certified 50 ppm standard', not just '4.5%'.
✓Match the precision of the inputs. Reporting 0.045327% from inputs known to three significant figures is meaningless. Quote relative error to one fewer significant figure than your least-precise input.
✓Use signed percent error for bias. Relative error hides direction. When detecting systematic bias, also report the signed difference so you can see whether the measurement is consistently high or low.
✓Guard against zero denominators. Relative error is undefined when the true value is zero — switch to absolute error in that regime.
✓Beware of relative error near zero. Small denominators amplify relative error catastrophically. A 0.01 ± 0.005 measurement is 50% relative error even though the absolute uncertainty is tiny.

Common Relative-Error Mistakes

Dividing by the observed value

Relative error is defined relative to the true value, not the observed value. The percent-difference formula uses an average, which is a different quantity.

Forgetting the absolute bars

Relative error is magnitude only. If you keep the sign you have computed signed relative deviation, not relative error.

Mixing units before dividing

Subtract values in the same units before computing the ratio. A 1 cm versus 1 m mix produces a meaningless answer.

Quoting relative error for zero references

When the true value is zero, fall back to absolute error — the relative form is undefined and explodes near zero.

Why Relative Error Matters

Relative error is the universal language of measurement quality. Every published instrument specification, every regulatory accuracy requirement, and every comparison between lab methods reduces to a relative error figure. A spec like 'accurate to ±0.5%' is a relative-error statement.

It also lets you transfer accuracy claims across scales without re-deriving the math. If a balance is rated 0.1% accurate, it deviates by no more than 0.1 mg on a 100 mg sample and no more than 100 mg on a 100 g sample — the relative number stays constant.

Built for laboratory analysts, physics students, engineers, calibration technicians, and anyone reporting measurement quality on a unit-less scale.

Formulas cross-checked against standard measurement-science references — see our methodology and editorial policy. Educational only — confirm critical scientific or regulatory calculations with your discipline's authoritative source.

Relative Error FAQs

Relative error is the absolute error of a measurement divided by the magnitude of the true value: RE = |Vobs − Vtrue| / |Vtrue|. It is unit-less, scale-independent, and can be multiplied by 100 to give percent error.

Subtract the true value from the observed value, take the absolute value of the difference, then divide by the absolute value of the true value. Multiply by 100 to convert to a percentage.

It depends on the field. Routine analytical chemistry expects below 2%, student physics labs below 5%, and industrial process control often below 0.5%. The calculator grades any input automatically.

Yes, when expressed as a percentage. Percent error is just relative error × 100, with absolute bars so it carries no sign. Some texts reserve 'percent error' for the signed version and 'relative error' for the absolute version.

When the true value is zero, the denominator is zero and relative error is undefined. In that regime report absolute error or, if zero is the expected result, use the noise floor of the instrument as a reference.

Because absolute error depends on the units and scale of the measurement, while relative error does not. A 1 mL discrepancy is huge on a 10 mL dose and trivial on a 1 L volume — relative error makes that explicit.

Accuracy is the complement of relative error: Accuracy = (1 − RE) × 100%. A 4% relative error corresponds to 96% accuracy.

All arithmetic is exact double-precision; the calculator's intrinsic precision exceeds about 15 significant digits, far beyond any real measurement need. Real-world accuracy is limited by your input values, not by the calculator.

Relative Error Calculator