2.2.1FoundationsCore

Errors and Uncertainties

Every measurement you will ever take in this course — and every one you take in the twelve required practicals — comes with some built-in doubt. Reporting a result honestly means reporting that doubt alongside it, and knowing exactly how that doubt behaves when you combine several measurements into a final calculated answer.

40 min Video by Science Shorts Module 2.2 Making measurements and analysing data
Errors & Uncertainties — GCSE Science & A-level Physics Practical SkillsCurated by Kepler, not produced by us. Watch it before the notes below.
Open on YouTube

What you'll be able to do

  • Distinguish random error from systematic error
  • Distinguish precision from accuracy, and state the role of resolution
  • Calculate absolute, fractional and percentage uncertainty
  • Combine uncertainties for sums, differences, products, quotients and powers
  • Find the uncertainty in a mean from a set of repeated readings
1

Random and systematic error

A random error\textbf{random error} causes readings to scatter unpredictably above and below the true value, often due to things outside the experimenter’s full control — reaction time when starting a stopwatch, or a slightly different line of sight each time a scale is read. Repeating the reading and averaging reduces the effect of random error, since the scatter tends to cancel out.

A systematic error\textbf{systematic error} shifts every single reading by roughly the same amount, in the same direction — for example, a metre ruler with a worn-down end that makes every length read 2 mm too long. No amount of repeating the measurement removes a systematic error; it can only be found and corrected by recognising its cause (or recalibrating the instrument).

Tip — If repeating a measurement many times and averaging doesn’t bring your result closer to the accepted value, suspect a systematic error, not "not enough readings".

2

Precision, accuracy and resolution

Precision\textbf{Precision} describes how tightly repeated measurements cluster together (a small random error) — it says nothing about whether they cluster around the correct value. Accuracy\textbf{Accuracy} describes how close a measurement is to the true, accepted value (small overall error, both random and systematic). A set of readings can therefore be precise without being accurate, if a systematic error has shifted them all consistently away from the truth.

The resolution\textbf{resolution} of an instrument is the smallest change in the quantity that it can actually detect — for digital vernier callipers reading to 0.01 mm, the resolution is 0.01 mm. Resolution sets a hard lower limit on how small the uncertainty in any single reading can be claimed to be.

Tip — Picture an archery target: tightly grouped arrows off to one side are precise but inaccurate; widely scattered arrows centred on the bullseye are accurate but imprecise.

3

Expressing and combining uncertainty

The absolute uncertainty\textbf{absolute uncertainty} is the size of the possible error in a measurement, in the measurement’s own units (e.g. ±0.2\pm0.2 mm). The percentage uncertainty\textbf{percentage uncertainty} expresses that same uncertainty as a fraction of the measured value, useful for comparing measurements of very different sizes and for combining uncertainties through a calculation.

When quantities are added or subtracted, their absolute uncertainties add. When quantities are multiplied or divided, their percentage uncertainties add. When a quantity is raised to a power nn, its percentage uncertainty is multiplied by nn.

% uncertainty=Δxx×100%\text{\% uncertainty} = \frac{\Delta x}{x}\times100\%
Percentage uncertainty from absolute uncertainty and measured value.
Example 1The diameter of a wire is measured with digital callipers (resolution 0.01 mm) as (0.82±0.01)(0.82 \pm 0.01) mm. Calculate the percentage uncertainty.
1Use Δxx×100%\dfrac{\Delta x}{x}\times100\%.
2=0.010.82×100%= \dfrac{0.01}{0.82}\times100\%.
Answer
≈1.2%\approx 1.2\%
Example 2A cylinder has radius (1.20±0.02)(1.20\pm0.02) cm and height (5.0±0.1)(5.0\pm0.1) cm. Find the percentage uncertainty in its volume, V=πr2hV=\pi r^2 h.
1Percentage uncertainty in radius: 0.021.20×100%=1.67%\dfrac{0.02}{1.20}\times100\% = 1.67\%.
2Because V∝r2V \propto r^2, this doubles: 2×1.67%=3.33%2\times1.67\% = 3.33\%.
3Percentage uncertainty in height: 0.15.0×100%=2.0%\dfrac{0.1}{5.0}\times100\% = 2.0\%.
4These are multiplied together in the volume formula, so their percentage uncertainties add: 3.33%+2.0%3.33\% + 2.0\%.
Answer
≈5.3%\approx 5.3\% total percentage uncertainty in the volume.

Tip — Powers are unforgiving: a squared quantity doubles its percentage uncertainty, and a cubed quantity triples it — the radius in the example above contributes far more uncertainty to the volume than the height does, despite being measured more precisely in absolute terms.

4

Uncertainty in a mean

When several repeated readings are taken, a simple and defensible estimate of the uncertainty in their mean is half the range\textbf{range}: half the difference between the largest and smallest reading obtained.

Δxˉ=xmax⁡−xmin⁡2\Delta \bar{x} = \frac{x_{\max}-x_{\min}}{2}
Uncertainty in a mean, from the range of repeated readings.
Example 1Five repeated measurements of the time for 10 oscillations of a pendulum give: 16.2 s, 16.5 s, 16.1 s, 16.4 s, 16.3 s. Find the mean and its uncertainty.
1Mean =16.2+16.5+16.1+16.4+16.35=81.55=16.3= \dfrac{16.2+16.5+16.1+16.4+16.3}{5} = \dfrac{81.5}{5} = 16.3 s.
2Range =16.5−16.1=0.4= 16.5-16.1 = 0.4 s.
3Uncertainty =0.42=0.2= \dfrac{0.4}{2} = 0.2 s.
Answer
Mean time for 10 oscillations =(16.3±0.2)= (16.3\pm0.2) s.

Tip — Always quote the uncertainty to the same number of decimal places as the mean — "16.3 ± 0.2 s", never "16.3 ± 0.2214 s".

Equation recap

% uncertainty=Δxx×100%\text{\% uncertainty} = \dfrac{\Delta x}{x}\times100\%
Percentage uncertainty.
Δ(x±y)=Δx+Δy\Delta(x\pm y) = \Delta x + \Delta y
Combining uncertainties: sums/differences.
Δ(xy)xy=Δxx+Δyy\dfrac{\Delta(xy)}{xy} = \dfrac{\Delta x}{x}+\dfrac{\Delta y}{y}
Combining uncertainties: products/quotients.
Δ(xn)xn=nΔxx\dfrac{\Delta(x^n)}{x^n} = n\dfrac{\Delta x}{x}
Combining uncertainties: powers.
Δxˉ=xmax⁡−xmin⁡2\Delta\bar{x} = \dfrac{x_{\max}-x_{\min}}{2}
Uncertainty in a mean from repeated readings.

Common mistakes to avoid

Trying to remove a systematic error by taking more repeated readings.
Repeating a measurement only reduces RANDOM error; a systematic error shifts every reading the same way and must be corrected at its source.
Claiming an uncertainty smaller than an instrument’s resolution allows.
The resolution of the instrument used sets a hard floor on the smallest sensible uncertainty for a single reading.
Adding percentage uncertainties for a sum or difference of two measured quantities.
For sums and differences, add the ABSOLUTE uncertainties; percentage uncertainties are only added for products and quotients.

Key takeaways

  • Random error scatters readings (reduced by repeating and averaging); systematic error shifts every reading the same way (only removed by correcting the cause).
  • Precision is about consistency; accuracy is about closeness to the true value; resolution sets the smallest detectable change.
  • Percentage uncertainty = (absolute uncertainty ÷ value) × 100%.
  • Sums/differences: add absolute uncertainties. Products/quotients: add percentage uncertainties. Powers: multiply percentage uncertainty by the power.
  • Uncertainty in a mean ≈ half the range of the repeated readings.

Test yourself

Answer a few questions, then review anything you missed.

Prefer paper? Open the written self-check

Answer each question before opening its answer. This self-check does not record a quiz score.

  1. A systematic error in an experiment is best removed by:

    • Identifying its cause and correcting or recalibrating for it
    • Taking many more repeated readings
    • Averaging a larger sample
    • Ignoring it, since it cancels out over time
    Show answer and explanation

    Answer: Identifying its cause and correcting or recalibrating for it

    Repeating readings only reduces random error, not systematic error.

  2. A set of repeated readings is tightly grouped but consistently offset from the true value. This set is:

    • Precise but not accurate
    • Accurate but not precise
    • Both accurate and precise
    • Neither accurate nor precise
    Show answer and explanation

    Answer: Precise but not accurate

    Tight grouping shows precision; being offset from the true value shows inaccuracy.

  3. A length is measured as (15.0 ± 0.3) cm. The percentage uncertainty is:

    • 2%
    • 0.3%
    • 15%
    • 30%
    Show answer and explanation

    Answer: 2%

    (0.3/15.0) × 100% = 2%.

  4. Two lengths, (8.0 ± 0.2) cm and (5.0 ± 0.1) cm, are subtracted. The absolute uncertainty in the result is:

    • 0.3 cm
    • 0.1 cm
    • 0.2 cm
    • 0.02 cm
    Show answer and explanation

    Answer: 0.3 cm

    For subtraction, absolute uncertainties still add: 0.2 + 0.1 = 0.3 cm.

  5. A quantity depends on r³, where r has a percentage uncertainty of 4%. The percentage uncertainty in r³ is:

    • 12%
    • 4%
    • 7%
    • 64%
    Show answer and explanation

    Answer: 12%

    For a cube, percentage uncertainty is multiplied by 3: 3 × 4% = 12%.

  6. Repeated readings of a mass are 12.4 g, 12.8 g, 12.2 g, 12.6 g. The uncertainty in the mean, by the half-range method, is:

    • 0.3 g
    • 0.6 g
    • 0.15 g
    • 1.2 g
    Show answer and explanation

    Answer: 0.3 g

    Range = 12.8 − 12.2 = 0.6 g; uncertainty = half the range = 0.3 g.

  7. Two quantities with percentage uncertainties of 3% and 6% are multiplied together. The percentage uncertainty in their product is:

    • 9%
    • 3%
    • 18%
    • 4.5%
    Show answer and explanation

    Answer: 9%

    For multiplication, percentage uncertainties add: 3% + 6% = 9%.

  8. Why does the resolution of a measuring instrument set a minimum possible uncertainty for a reading?

    • Because the instrument cannot detect any change smaller than its resolution
    • Because resolution only affects systematic error
    • Because resolution is unrelated to uncertainty
    • Because higher resolution always means lower accuracy
    Show answer and explanation

    Answer: Because the instrument cannot detect any change smaller than its resolution

    A reading cannot be more precise than the smallest change the instrument itself can register.

Finished this lesson?

Loading your progress…

Next step

Keep practising

Start with the quiz above to check the key ideas. You can also choose another topic from the practice library.

Browse practice topics

Related lessons

All 4 lessons