Skip to content
Omnikalc

How uncertain is your calculated result?

Enter two measurements with their uncertainties and pick an operation: get the result, its absolute and relative uncertainty and the range it falls in.

Updated on
Your inputs

Value out of range

Value out of range

Value out of range

Value out of range

Results

Result

60.0000

Absolute uncertainty (±)

1.5761

Relative uncertainty
2.63%
Good precision: below 5%
Lower bound
58.4239
Upper bound
61.5761

Every measurement has an uncertainty, and when you combine two measurements in a calculation, their uncertainties combine too. This calculator applies the standard rules of error propagation for independent measurements: enter two values with their uncertainties and choose a sum, a difference, a product or a quotient. You get the result, its uncertainty and a range in which the true value is likely to lie.

How the calculation works

  • Sum or difference: the uncertainty is √(σx² + σy²)
  • Product x × y: the uncertainty is √((y × σx)² + (x × σy)²)
  • Quotient x ÷ y: the uncertainty is √((σx ÷ y)² + (x × σy ÷ y²)²)
  • Relative uncertainty = absolute uncertainty ÷ |result|

Example

A rectangular plate measures 12.5 ± 0.2 cm by 4.8 ± 0.1 cm. The area is 12.5 × 4.8 = 60 cm². The uncertainty is √((4.8 × 0.2)² + (12.5 × 0.1)²) = 1.576 cm², a relative uncertainty of 2.63%. The area lies between 58.42 and 61.58 cm². Adding the uncertainties directly would give 1.7 cm², which overstates the error.

Where it is used

Laboratory reports, engineering tolerances, surveying and cooking at scale all use the same idea. Physics students propagate errors to see which measurement limits the result: in a product, the one with the largest relative uncertainty usually dominates, so improving the other instrument will not help much. Quality teams use the same rules to stack tolerances in an assembly.

Good to know

These formulas are first-order approximations that work well when the uncertainties are small compared with the values. They assume independent, random errors. A systematic error, such as a badly calibrated scale, affects all measurements in the same direction and is not captured here. When the denominator in a quotient comes close to zero, the relative uncertainty explodes, which is a warning that the result is unreliable. Report a result with no more digits than its uncertainty justifies.

For the uncertainty of a sample mean, use the standard error calculator. For the gap between a measured and an accepted value, try the percent error calculator, and to round to the right number of digits, the scientific notation calculator.

Frequently asked questions

Why not just add the uncertainties?

Adding them gives the worst case. When errors are independent, they are as likely to cancel as to add up, so they combine in quadrature, as the square root of the sum of squares.

What is the rule for a sum or a difference?

The absolute uncertainties combine in quadrature: the uncertainty of x + y or x − y is the square root of the sum of the squared uncertainties.

What is the rule for a product or a quotient?

The relative uncertainties combine in quadrature: the relative uncertainty of x × y or x ÷ y is the square root of the sum of the squared relative uncertainties.

Are the measurements assumed independent?

Yes. If the errors are correlated, for example from the same instrument, the real uncertainty can be larger or smaller than this calculator shows.

Was this calculator helpful?

Thanks for your feedback!

Results are estimates provided for information purposes. They do not replace professional advice.