Overview

When you calculate with measured values, their uncertainties carry through to the result. This combines the uncertainties of two values for any basic operation.

Enter both values with their ± uncertainties, and choose the operation. It assumes the two errors are independent.

Worked examples

InputsResult
Value A10Uncertainty of A (±)0.2Value B5Uncertainty of B (±)0.1OperationA × BResult: 50 ± 1.4142Relative uncertainty: 2.83%
Show 2 more examples
Value A10Uncertainty of A (±)0.2Value B5Uncertainty of B (±)0.1OperationA + BResult: 15 ± 0.2236Relative uncertainty: 1.49%
Value A10Uncertainty of A (±)0.2Value B5Uncertainty of B (±)0.1OperationA ÷ BResult: 2 ± 0.0566Relative uncertainty: 2.83%

How it works

For addition and subtraction, the absolute uncertainties combine in quadrature: the square root of the sum of their squares. For multiplication and division, the relative uncertainties combine in the same way. Independent errors partly cancel, so the combined uncertainty is smaller than simply adding them.

The formula

δR = √(δA² + δB²) for + and −; δR ÷ |R| = √((δA ÷ A)² + (δB ÷ B)²) for × and ÷
δA, δBUncertainties of A and B
RThe result

Practical tips

  • (10 ± 0.2) × (5 ± 0.1) = 50 ± 1.41, since both have a 2% relative uncertainty.
  • (10 ± 0.2) + (5 ± 0.1) = 15 ± 0.22.
  • If the errors are linked rather than independent, adding them directly gives a safer upper limit.

Frequently asked questions

Why not just add the uncertainties?

That assumes both errors go the same way at once. For independent errors, quadrature is more realistic.

Can I use this for more than two values?

Yes. Combine two values, then use that result with the next one.