Techniques that increase the accuracy of your measurements
Why techniques must be written down, not just used
Merit requires you to use technique(s) that increase the accuracy of the measured values — and to say what you did and why it helped. Using a good technique silently earns nothing; the write-up has to name it.
- Accuracy means how close a measurement is to the true value.
- Precision means how finely you can read the instrument, and how closely repeated readings agree with each other.
- A technique improves accuracy when it reduces the size of an error relative to the quantity being measured.
Techniques worth using (and why each works)
- Measure a multiple, then divide. Time 10 oscillations and divide by 10, rather than timing one.
- Your reaction time error (about s) is the same whether you time one swing or ten, so dividing by 10 makes that error ten times smaller as a fraction of the period.
- Repeat and average. Take at least three readings at each value and use the mean.
- Random errors scatter above and below the true value, so averaging cancels much of them out. Repeats also let you see the spread, which becomes your uncertainty.
- Use a fiducial marker. Put a fixed mark (a pin, a line on paper) at the centre of the swing and start/stop timing as the pendulum passes it.
- The bob is moving fastest at the centre, so it spends the least time near the mark — your eye can judge the moment of passing much more sharply than at the turning points, where it hangs almost still.
- Avoid parallax error. Read a scale with your eye directly in line with the mark, and use a mirror scale where one is available.
- Reading at an angle makes the pointer appear shifted along the scale, which produces a systematic error in one direction.
- Zero the instrument before use. Check a balance, a newton meter or a micrometer reads zero with nothing on it, and subtract any offset if it does not.
- An un-zeroed instrument adds the same error to every reading, which shifts the whole line on your graph rather than scattering the points.
- Measure the largest quantity you can. Where you have a choice, measure over a large distance, mass or time.
- The instrument's uncertainty is fixed, so the percentage uncertainty falls as the measured value gets bigger.
Writing the technique up
A Merit-level sentence has three parts: what you did, which error it reduces, and why that helps.
- Weak: "I timed 10 swings."
- Strong: "I timed 10 complete oscillations and divided by 10, because my reaction-time error of about s is the same for one swing as for ten, so dividing by 10 reduces its effect on the period to about s."
Worked ExampleChoosing and justifying a technique
A student is measuring how the time for a trolley to roll down a ramp depends on the ramp's height. Their first attempt uses a stopwatch to time a single run over a cm track, and the readings vary by up to s. Describe two techniques that would improve the accuracy of the timing, and justify each.
Step 1 — Identify the error that dominates
Over cm the trolley takes well under a second, and the readings vary by s. That scatter is about the size of human reaction time, so reaction time is the dominant error — not the stopwatch itself, which reads to s.
Step 2 — First technique: lengthen the timed run
Extend the track to the full length available, say m. The reaction-time error stays at roughly s regardless, but the measured time is now several times larger, so the percentage uncertainty in the time is several times smaller.
Step 3 — Second technique: remove the human from the timing
Use light gates (or a video recording analysed frame by frame) instead of a hand-held stopwatch. This removes reaction time from the measurement entirely, rather than just reducing its relative size, because the timing is triggered by the trolley itself.
Step 4 — Say what to do if neither is possible
If light gates are unavailable, repeat each run at least three times and take the mean, since reaction-time error is partly random — sometimes early, sometimes late — so averaging reduces it, although it cannot remove any consistent tendency to press late.