To reduce random error
Measurements & Uncertainties - OCR A-Level Physics
- To reduce random errorAn error that causes readings to scatter unpredictably above and below the true value. Can be reduced by averaging repeated measurements.: take more repeat readings and calculate the mean, use instruments with higher resolutionThe smallest change in a quantity that an instrument can detect. For example, a ruler has a resolution of 1 mm., use data-logging equipment instead of human timing.
- To reduce systematic errorAn error that shifts all readings by the same amount in the same direction. Cannot be reduced by repeating measurements.: recalibrate instruments, check for zero errors before use, use a fiducial marker to reduce parallax.
- To reduce percentage uncertaintyUncertainty expressed as a percentage of the measured value: (absolute uncertainty / measured value) × 100%. in a measurement: measure a larger quantity.
- Example: timing 1 oscillation gives high % uncertainty. Time 10 oscillations and divide by 10 -- the absolute uncertaintyThe uncertainty expressed in the same units as the measurement. Written as +/- a value after the reading. stays the same but the measured value is 10x larger, so the percentage uncertainty is 10x smaller.
Common Mistake
MEDIUM
Wrong: Saying 'use a more accurate stopwatch' to reduce timing uncertainty.
Right: The resolutionThe smallest change in a quantity that an instrument can detect. For example, a ruler has a resolution of 1 mm. of a digital stopwatch (0.01 s) is not the main source of uncertainty -- human reaction time (~0.2 s) dominates. Time multiple oscillations to reduce the percentage uncertainty, or use a light gate.
Right: The resolutionThe smallest change in a quantity that an instrument can detect. For example, a ruler has a resolution of 1 mm. of a digital stopwatch (0.01 s) is not the main source of uncertainty -- human reaction time (~0.2 s) dominates. Time multiple oscillations to reduce the percentage uncertainty, or use a light gate.
Examiner Tips and Tricks
- When asked to suggest improvements, be specific.
- Do not write 'reduce errors' -- state WHICH error (systematic or random), HOW you reduce it (e.g. 'time 20 oscillations instead of 1'), and WHY this works (e.g. 'the percentage uncertainty in the periodThe time taken for one complete oscillation or wave cycle. Measured in seconds (s). decreases because the total time measured is larger').