Skip to content

Measurements and their Errors

Uncertainties and Error Bars

AQA A-level Physics


Absolute, fractional and percentage uncertainty

fractional uncertainty = absolute uncertainty ÷ valuefractional uncertainty = Δx / x
percentage uncertainty = uncertainty ÷ value × 100 %percentage uncertainty = Δx / x × 100 %
  • The absolute uncertainty has the same unit as the quantity. The value in the denominator is the mean of the repeated readings.
  • Write a percentage uncertainty with %. Never write a percentage uncertainty as a decimal.
  • Same absolute uncertainty, larger value: absolute uncertainty is the same but value is larger, so the percentage uncertainty is smaller.
  • For a small measured value write percentage uncertainty is large, naming the quantity. Never write small distances are hard to measure.

Finding the uncertainty

Repeated readings

uncertainty = half the rangeΔx = (largest reading − smallest reading) / 2
  • Remove an anomalous reading before taking the range.
  • Percentage uncertainty = ½ range / mean × 100.

One reading and two readings

  • The uncertainty in one reading is half the resolution of the scale when a scale is read. Where a spot or line has to be located, it can be a whole division.
  • A length found from two readings (one at each end): the uncertainties in each reading are added, so absolute uncertainty = 2 × uncertainty in each reading. Never write because the smallest division is 1 mm as the reason.
  • Locating the exact position of a spot or line adds to the uncertainty: uncertainty in locating exact position.
  • Timing many oscillations: the uncertainty of the one timing is spread over all of them, so percentage uncertainty = uncertainty in the timing / total time × 100 %.

Reducing the percentage uncertainty

  • Make the measured quantity larger: larger values, so the percentage uncertainty is reduced.
  • Time many oscillations, such as 20T: this reduces the percentage uncertainty.
A larger value is what reduces a percentage uncertainty

To reduce the percentage uncertainty in a distance, make the distance larger: increase distance, decreasing the percentage uncertainty. Never write repeat and average to reduce the percentage uncertainty in a distance or a ruler with smaller divisions to reduce the percentage uncertainty in a distance. Neither makes the distance larger, and a percentage uncertainty falls when the same absolute uncertainty is a smaller fraction of a larger value.

Combining uncertainties

sum or difference, y = a + b or y = a − bΔy = Δa + Δb
product or quotient, y = a b or y = a / bpercentage uncertainty in y = percentage uncertainty in a + percentage uncertainty in b
power, y = anpercentage uncertainty in y = n × percentage uncertainty in a
absolute uncertainty from percentage uncertaintyΔy = percentage uncertainty in y / 100 × y
Method: uncertainty in a calculated result
  1. Find the percentage uncertainty in each measured quantity.
  2. Multiply each by the power of that quantity in the equation: for d2, 2 × % uncertainty in d.
  3. Add them: the total is the percentage uncertainty in the result.
  4. Convert to an absolute uncertainty: value × percentage uncertainty / 100.
Percentage uncertainties add

For a product or quotient the percentage uncertainties of the quantities are added. Never write the mean of two separate percentage uncertainties: averaging halves the combined uncertainty, but every quantity contributes its full uncertainty to the result.

Error bars

  • An error bar extends the absolute uncertainty above and below the point.
  • Draw a straight line of best fit passing through all error bars.
  • A constant absolute uncertainty: the error bars are the same length. A constant percentage uncertainty: the bars grow longer as the value grows.
  • Compare the lengths of the error bars with a number: how many times longer one bar is than another.

Uncertainty in a gradient and an intercept

xySteepest lineLine of best fitShallowest lineError bar
The steepest and shallowest lines each pass through every error bar.
uncertainty in gradient = (maximum gradient − minimum gradient) / 2ΔG = (Gmax − Gmin) / 2
  • Steepest line: ruled through bottom of first error bar and through top of last error bar it can pass through. Shallowest line: top of the first, bottom of the last.
  • Both are lines that pass through all the error bars, each a thin, single, continuous ruled line.
  • Best gradient: the mean of Gmax and Gmin.
  • Percentage uncertainty in the gradient = ΔG / best gradient × 100 %.
  • Uncertainty in an intercept: the same method, using the intercepts of the steepest and shallowest lines.

Significant figures

  • Give the result with sf consistent with uncertainty: the value to the same decimal place as its absolute uncertainty, written as value ± uncertainty with its unit.
  • Give a percentage uncertainty to two significant figures.

Quantities and units

QuantitySymbolUnit
Absolute uncertainty in xΔxunit of x
Fractional uncertaintyΔx / xnone
Percentage uncertaintynone%
Gradient uncertaintyΔGunit of the gradient

That was one page of 90

The rest of AQA A-level Physics is written the same way

The other 89 pages of AQA A-level Physics are written exactly like this one, with the mark scheme wording highlighted throughout. A year of the whole course is $9.99.

Subscriptions open as soon as payment processing is approved.