How do the uncertainties in measured quantities combine when those quantities are added, multiplied or raised to a power?
Combine uncertainties in derived quantities by adding absolute uncertainties for sums and differences and adding fractional uncertainties for products, quotients and powers
A focused answer to the H2 Physics Measurement learning outcome on propagating uncertainty. Adding absolute uncertainties for sums and differences, adding fractional uncertainties for products and quotients, and handling powers.
Reviewed by: AI editorial process; not yet individually human-reviewed
Have a quick question? Jump to the Q&A page
Jump to a section
What this dot point is asking
SEAB wants you to propagate measurement uncertainties through a calculation: add absolute uncertainties when quantities are added or subtracted, and add fractional (or percentage) uncertainties when quantities are multiplied, divided or raised to a power. This is examined in the Paper 3 data-based question and is the analysis backbone of Paper 4.
The answer
Absolute, fractional and percentage uncertainty
For a measured quantity :
- is the absolute uncertainty (same units as ).
- is the fractional uncertainty (no units).
- is the percentage uncertainty.
Rule 1: sums and differences add absolute uncertainties
If or , then:
Note that the absolute uncertainties always add, even for a difference. A difference of two close values can therefore have a very large fractional uncertainty, which is why subtracting nearly equal measurements is poor experimental design.
Rule 2: products and quotients add fractional uncertainties
If , then:
The fractional uncertainties add regardless of whether a quantity is in the numerator or denominator.
Rule 3: powers multiply the fractional uncertainty
If , then:
So a squared quantity contributes twice its fractional uncertainty, a cubed quantity three times, and a square root half. This is just Rule 2 applied to repeated factors.
Getting to the final absolute uncertainty
After combining fractional uncertainties, convert back to an absolute uncertainty by multiplying the fractional uncertainty by the calculated value of :
Then round the uncertainty to one significant figure and the value to match.
Examples in context
Example 1. The danger of subtracting close values. Measuring a small temperature rise as , each with , gives an absolute uncertainty of on a difference of : a uncertainty. This is why experiments are designed to measure large differences.
Example 2. Identifying the dominant error. In the pendulum example above, the period contributed while the length contributed only . Recognising that the squared, larger-fractional-uncertainty quantity dominates tells a student exactly where to improve their technique: time more oscillations to shrink the period uncertainty.
Try this
Q1. Two lengths and are added. State the result with its absolute uncertainty. [2 marks]
- Cue. (absolute uncertainties add).
Q2. A quantity is found from with known to and to . Find the percentage uncertainty in . [2 marks]
- Cue. (square doubles the voltage contribution).
Q3. A cube has side . Find the volume and its percentage uncertainty. [3 marks]
- Cue. ; percentage uncertainty , so .
Exam-style practice questions
Practice questions written in the style of SEAB exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.
Original4 marksThe density of a cylinder is found from . The measurements are , and . Determine the percentage uncertainty in the density.Show worked answer →
For a product or quotient, add the fractional (percentage) uncertainties, and double the one that is squared.
Mass: .
Radius (appears as ): , doubled to .
Length: .
Total percentage uncertainty: .
Markers reward converting each absolute uncertainty to a percentage, doubling the radius contribution because of the square, and summing to a final percentage. The is exact and contributes nothing.
Original3 marksA resistance is found from two voltage readings and across a component, where the voltage drop used is . Determine the voltage drop and its absolute uncertainty.Show worked answer →
For a difference, add the absolute uncertainties.
Voltage drop: .
Absolute uncertainty: .
So the voltage drop is .
Markers reward adding (not subtracting) the absolute uncertainties for a difference, and presenting the result with value and uncertainty to matching decimal places. Note the fractional uncertainty of a difference can be large when the two values are close.
Related dot points
- Distinguish random and systematic errors, relate them to precision and accuracy, and quote results to an appropriate number of significant figures with an estimated uncertainty
A focused answer to the H2 Physics Measurement learning outcome on errors. Random versus systematic errors, the link to precision and accuracy, and how to quote a measurement with an uncertainty and sensible significant figures.
- Recall the SI base quantities and their units, express derived units as products or quotients of base units, and use base units to check the homogeneity of physical equations
A focused answer to the H2 Physics Measurement learning outcome on SI base quantities and units. The seven base quantities, how derived units are built from them, and how to check an equation for homogeneity using base units.
- Use SI prefixes from pico to tera, convert between prefixed units consistently, and make order-of-magnitude estimates to check whether a numerical answer is physically reasonable
A focused answer to the H2 Physics Measurement learning outcome on prefixes and estimation. The common SI prefixes, how to convert safely between them, and how order-of-magnitude estimates catch unreasonable answers.
- Distinguish scalar and vector quantities, add coplanar vectors, and resolve a vector into perpendicular components
A focused answer to the H2 Physics Measurement learning outcome on scalars and vectors. The distinction, vector addition by the parallelogram and triangle rules, and resolving a vector into perpendicular components.